CHAIR OF

COMPUTATIONAL

MECHANICS

Vertical stacking of integrated circuits (IC), known as the 3-D integration, has emerged as a breakthrough solution to overcome the limitations of traditional continuous scaling of individual components. 3-D integration enables superior performances due to shorter interconnections, reduced system sizes and improved system heterogeneity. Circuit layers can be fabricated separately, enabling the combination of incompatible manufacturing processes into a single 3D IC (e.g., memory and logic). Among several 3-D integration schemes, wafer-to-wafer hybrid bonding is distinguished as the key technology for achieving high-density interconnections, required for fine-partitioning 3-D system-on-chip applications. In this bonding technology, both wafers are finished with a dielectric layer with embedded Cu pads. The wafer pair is then accurately aligned with a small vertical distance between them, in the order of a few tens of nanometres. The top wafer is pushed through a localized area at its centre to establish the initial contact between wafers. The bonding is then propagated in a wave-like pattern due to the interaction forces between opposing wafers.

A major challenge associated with this process is acquiring a sufficiently low alignment error between the bonded wafers, for which the current and future industry demands are extremely stringent. It is not a trivial task to meet these demands since the final alignment is affected by various parameters, some of which can be listed as: wafer properties (shape, residual stress, mechanical properties), the dielectric material choice, extrusion/recession of Cu pads, the chuck design holding the wafers, the bonding recipe, initial distance between wafers, the magnitude of the point contact force, adhesion forces between dielectrics, air viscosity and gravitational effects. Considering the presence of numerous parameters and their potential interactions, optimization attempts that only focus on experimental approaches will have limited capabilities due to their slow-paced and expensive nature. Therefore, it is necessary to develop supplementary methodologies based on simulation techniques.

The purpose of this PhD topic is to deepen the understanding on both the physics of bonding propagation and the impact of mechanical boundary conditions using modelling and simulation techniques. To achieve this, a physics-based mechanical modelling environment (based on the finite-element method) will be developed to study the bonding phenomena. Modelling activities will first be initiated using simplified 2-D models, which will later be extended to full 3-D wafer bonding simulations. Where possible, bond wave metrology data along with experimental information will be provided to calibrate the simulations. The learnings obtained through simulations will be utilized to guide the industry and academia in terms of wafer preparation and bonding configurations.

View project in the research portal

The Modular Peristaltic Surface Conveyor (MPSC) is an entirely new device that conceptually enables the separation and sorting of flexible small packages (polybags) for the first time, providing an alternative to costly manual processing. For the first time, alongside the development of the actual MPSC, an AI-based Digital Twin (DT) is to be developed, which, based on AI-optimized simulation models, will allow predictions of system behavior and automated parameterization of the actuators and sensor data processing.

View project in the research portal

The DFG-funded research project aims to increase the numerical prediction capability for technical systems by implementing a holistic simulation methodology that enables efficient coupling between a multi-body simulation and a non-linear FE model. An extension of the physically motivated dynamic flocculation model is used to fully and precisely map the non-linear material behavior of elastomeric bearing elements. The focus here is primarily on the changes in the properties of the bearings under multi-axial loading, which are often neglected in current modeling approaches at . Since the integration of a detailed FE model leads to an increase in the necessary computing resources, different levels of detail of the solver coupling are implemented and analyzed in this project with the aim of allowing a reduction in computing time with an acceptable loss of accuracy. The resulting different levels of complexity of the developed methodology are comprehensively compared with conventional modeling strategies. The individual coupling strategies are evaluated with regard to the implementation and parameterization effort as well as the physical interpretability and the required computing resources. The developed and validated FE models based on the DFM are also examined with regard to their suitability, to what extent and with what reliability certain material parameters can be transferred once to other geometries and load scenarios. Finally, the accuracy of all investigated strategies for coupling the FEM and MBS is assessed with the aid of test results from real applications. The FEM is integrated into the MBS both directly via various solver couplings and indirectly by generating a characteristic map or a surrogate model with the aid of the FE model for use within the MBS. The first application example is a laboratory centrifuge, whose vibration amplitudes and operating resonances are measured and compared with the numerically obtained results of the respective coupling strategies. Furthermore, the developed methodology is applied and validated in the context of a vibration analysis of chassis components of an electric vehicle.
This text was translated with DeepL

View project in the research portal

The aim of the project is to research ways of optimizing the fluid dynamic treatment of intracranial aneurysms in order to shorten the occlusion time, reduce the need for follow-up treatment and eliminate the risk of ruptures. To this end, novel neurovascular implants with improved flow-modelling properties are to be developed (target values: locally reduced porosity, optimized adaptability to the anatomy). Possible individualized solutions include the further development of braided structures or the use of novel polymer nonwovens on the support structure. On the other hand, "intelligent" software tools are being developed that enable improved planning and implantation based on virtual catheter guidance through complex 3D vessel models of patients. Deformation states of both the catheter and the crimped implant are simulated on their way to the brain aneurysm. In addition, a blood flow simulation is carried out to assess the effectiveness (intra-aneurysmal thrombosis) of the implant. The results will be used to provide interventionalists with information on handling the implant before and during treatment. Such software enables targeted optimization of the implant properties, for example to achieve localization-dependent reductions in velocity and vertebral strength of up to 50 % compared to the untreated state.

View project in the research portal

Adaptive mesh refinement (AMR) methods are absolutely essential in many industrial and scientific applications in order to reduce the numerical effort and thus make complex problems manageable in the first place. However, a look at the current literature on AMR reveals a number of shortcomings that still need to be resolved. In order to achieve local mesh refinement, either hybrid meshes consisting of simplex and tensor product elements or constraints must be used. However, both approaches inevitably lead to local accuracy losses. Furthermore, in industrial applications, linear approach functions are often used, which is why only algebraic convergence can be achieved. In the scientific environment, there are of course also approaches for complete hp-adaptivity. However, due to the complexity of their implementation, these methods are designed for networks with one hanging node per element edge/surface and have weaknesses when applied to highly dynamic processes (explicit time integration), as diagonal mass matrices are not available. It should be noted, however, that exponential convergence rates can be achieved compared to simple h-refinements. The mentioned problems can easily be eliminated by higher-order transition elements derived on the basis of the so-called mixed (transfinite) interpolation. The element formulation is based on quadrilateral or hexahedral elements in the reference domain and can couple arbitrary discretizations. In principle, a wide variety of element families can be coupled, which differ not only in size or order of approach. Since the functional space does not have to be restricted by constraints, there is no need to compromise on accuracy. For high-frequency, transient calculations, suitable methods for diagonalizing the mass matrix are also being developed in this project. The resulting element family forms the basis for dynamic mesh refinements. The special feature of this approach is the targeted combination of refinement and coarsening steps, which are carried out in each time step of the simulation. This allows optimal convergence rates to be achieved with the lowest possible numerical effort . In order to further increase the efficiency of the developed technique, the algorithms are prepared for high-performance computers. The outstanding properties of the proposed methodology are illustrated using selected examples of wave propagation. For this purpose, continuous structural monitoring by means of guided waves in microstructured materials and the analysis of seismic activities are used.
This text was translated with DeepL

View project in the research portal

Ankle-Foot Orthoses (AFOs) are those devices used for rehabilitation of a pathological gait, which is caused for instance by a stroke. This research aims to design, model, simulate, manufacture, and test a novel AFO, which is designed to ensure ease of use, freedom of movement, and high performance for high-level activities at relatively low costs. Research problems are inherent in the increasing demand for AFOs based on polymers, which have relatively low biomechanical properties and may cause skin sweating and irritation in the long term. Furthermore, there are problems related to the high costs of recent AFOs made of advanced composites or carbon fiber, the market needs (orthopedic workers) and users alike, and the necessity of a novel AFO that meets the demands and helps to produce orthoses for fitting each patient. Therefore, orthotists could save time and obtain a more convenient AFO prototype, which helps them in patients' treatment.
This study includes, from an applied point of view, the design, modeling, and simulation of a novel ankle-foot orthosis based on silicone, shape memory alloy (SMA), and elastic bands. This, in turn, ensures freedom of movement and high performance for high-level activities. It also includes, in practical terms, the manufacturing of the ankle-foot orthosis, based on the aforementioned design and materials, and conducting appropriate mechanical and biomechanical tests. This study also includes a literature review and description of the materials, methods, and equipment used in the design, modeling, simulation, manufacturing, and testing of a novel dynamic ankle-foot orthosis.

View project in the research portal

Neural networks are already used extensively in the field of data analysis. Common material models consist of physically based equations to describe the real behavior as good as possible. Measurements are used to adjust the material parameters, but the accuracy of the model depends on the complexity of the constitutive equations. Neural networks offer the possibility to describe a material with the same test data without the necessity to derive complex and physically based material laws.
Considering a uniaxial stress-strain curve of a hyperelastic material, a classical neural network can be easily set up to describe this behavior. During training, the network finds a good fitting function that depends mainly on the number of weights and biases and the amount of training data. These overall parameters are not physically motivated, as they only connect the stress values to the strain values via multiplication and the sigmoid transfer functions in the range of the trainings set. This is the reason why classical neural networks have a very poor extrapolation performance.
In contrast, autoregressive neural networks can train a time series, such as the stress curve with a constant strain rate, using previous stress values to calculate the next one. Instead of training a stress-strain function, these networks attempt to find a recursive formulation between stress values. With external inputs, other variables can also be used in the recursive formulation, such as the strain rate. If the training data contains different strain rates, the network can take them into account. In addition, other variables are possible, for example, different temperatures.
Due to the recursive or regressive functionality, the network can calculate stress-strain curves, even beyond the range of the training data. With a sufficiently large training data set, it is thus possible to describe more complex material behavior better than with classical material models.
In this project the properties of viscoelastic materials shall be estimated with an autoregressive neural network. Calculating a stress-strain curve with different strain rates and training the networks can be done in a few minutes. Prediction with different strain rates and stress values outside the range of the training data works very well with only a small error and much less computation time. In addition to optimizing the network architecture, the possibility of other external inputs such as temperature or training with a real measurement data set will also be investigated.

View project in the research portal

Predicting the acoustic behavior of systems containing materials with complex microstructure is a major challenge for several reasons. On the one hand, it is very time-consuming to build high-resolution numerical models using geometry-conforming discretizations and, on the other hand, all physically relevant interactions of the structure with both the surrounding and the enclosed fluid must be taken into account. The geometry-conform discretization of heterogeneous materials with complex microstructure usually leads to a very high number of finite elements and thus to unacceptable computing times. In recent years, fictitious domain methods, such as the Finite Cell Method (FCM), have emerged as an effective alternative. To capture the acoustic or vibroacoustic properties, the FCM must be extended in some aspects for the new field of application. First, the acoustic wave equation for calculations in the time domain and the Helmholtz equation for analyses in the frequency domain must be discretized using fictitious domain methods. Furthermore, suitable coupling strategies between the structural and fluid domains must be developed. The subfields can be coupled both weakly (without feedback) and strongly (with feedback). The advantage of fictitious domain methods is, in addition to the highly accurate resolution of the geometry (despite non-conformal discretization), the possibility of superimposing structural and fluid elements. This makes it possible to develop an effective strategy for the vibroacoustic coupling of heterogeneous materials. The numerical effort of these complex simulations is still very high, even when using fictitious domain methods. Therefore, another goal is to derive simplified models based on numerical homogenization methods in addition to the microstructurally resolved models. Despite the strong abstraction of reality, it is expected that useful results can be achieved for various applications. The final focus of the project is the experimental validation of the numerical methods developed. Various test rigs will be used for this purpose. The vibration behavior of the structure is crucial for the implementation of the vibroacoustic coupling. This can be investigated using a 3D laser scanning vibrometer. In addition, the frequency-dependent acoustic parameters are measured using various simple measurement setups, such as a Kundt's tube, and compared with the simulated results. Furthermore, the sound radiation is measured in a free-field room using microphone arrays and far-field microphones. On the basis of this data, the performance of the implemented models can be verified. Finally, guidelines for their use are derived.
This text was translated with DeepL

View project in the research portal

The project is dedicated to the development of an efficient calculation method for solving three-dimensional vibroacoustic problems using porous insulation materials. The aim is to resolve the microstructure of the insulation material in order to overcome the current limitations of Biot's theory, which is often used and seems particularly unsuitable for modeling closed-cell foams. In order to enable the extremely complex geometry-resolved modeling we are aiming for, fictitious domain methods with higher-order approach functions are to be used. On the one hand, these can be applied very advantageously to voxel data and, on the other hand, a high efficiency for wave propagation problems can be expected.

View project in the research portal

Simulation-based and sensor-based functionalized housing design

The ZIM network INSTANT is primarily concerned with medical issues. Within the network, the COCOON R&D project focuses on reducing noise pollution during diagnostic and interventional image-guided procedures.
Various medical studies show that persistently high noise levels can lead to poor concentration, stress, impaired memory, a general reduction in performance and other symptoms, including burnout syndrome. Such stress and anxiety situations are detrimental to the recovery of patients and lead to longer treatment times and therefore increased costs. On the part of clinical/medical staff, noise pollution can lead to loss of concentration and treatment errors, for example during interventions lasting several hours or several consecutive interventions.
In many machines, the generation of loud noises cannot be prevented or can only be prevented by interfering with the existing structure. However, technical measures can be taken to hinder the propagation and transmission of noise and thus minimize the disturbing noise emissions. The COCOON project is researching methods for designing and manufacturing acoustically optimized housings for large medical devices, which also results in very high standards with regard to approval and the materials used.
Furthermore, the ambitious approach of researching a "diagnostic system" for recording the status of product functionality is being pursued. The early alerting of malfunctions is intended to minimize device failures and can thus contribute to product monitoring after the product has been placed on the market.
This text was translated with DeepL

View project in the research portal

Filler reinforced elastomer blends play a key role in the design and optimization of high performance rubber goods like tires or conveyor belts. In most cases, a phase separated, anisotropic blend morphology develops during the last processing steps (extrusion, calendering, injection molding), which lowers its free energy by coagulation and relaxation processes, before the morphology is frozen by cross-linking. The development of the detailed phase morphology and its influence on the high-frequency viscoelastic response, affecting e.g. friction, fracture and wear properties, is not well understood at present but of high technological and scientific interest.
Accordingly, one main objective is the physically motivated modeling and numerical simulation of the thermo-chemically driven phase separation of filled elastomer blends with realistic, microscopic input parameters obtained from independent physical measurements. Beside the chemical compatibility of the polymers and the fillers, also the effect of mechanical stress on the phase dynamics shall be investigated. In combination with elaborated experimental methods, the phase field modeling for Cahn-Hilliard and Cahn-Larché type diffusion shall be applied. The local phase field equations, considering at the end three phases, must be implemented into the isogeometric analysis, allowing for the study of complex interaction of multi-phase materials with different material characteristics. The experimental focus lies on the evaluation of thermodynamic polymer-polymer- and polymer-filler interaction parameters that govern the phase morphology and filler distribution. For the simulation of phase boundary dynamics, the collective chain mobility shall be estimated as an input parameter of the Cahn-Hilliard type dynamic equation.
A second objective is the modeling and numerical simulation of the high-frequency linear viscoelastic response of unfilled and filled elastomer blends, which shall be based on the distinct phase morphology including domain and interphase size, filler distribution and cross-linking heterogeneities. The non-linear response will be analyzed in a future project.
The results of phase field simulations shall be compared to experimental investigations of phase mixture processes and numerically evaluated viscoelastic moduli shall be correlated with experimentally constructed viscoelastic master curves.
The sum of the both objectives leads to a complete numerical procedure with which it is possible to simulate the complete cycle of producing and using a new polymer blend for later engineering applications by optimizing the involved process and distinctive material parameters.

View project in the research portal

The aim of the dissertation is the development and further development of finite element technologies in the field of mixed formulation. The focus here is on the displacement-compression-strain formulation (u/p/e), as it enables both the mastering of incompressible material behavior and increased accuracy in the calculation of stresses and strains.

View project in the research portal

The aim of the joint project DampedWEA is to increase the acceptance of wind turbines. The aim is to open up new regions for wind turbines, particularly in the vicinity of inhabited areas. This requires a reduction in the radiated noise level. In this joint project, the focus is on tonal emissions, which are increasingly coming to the fore due to the successful optimization of aeroacoustic emissions and now pose a problem. In order to reduce these sufficiently, innovative concepts for vibration and noise reduction are used. The main source of tonal noise is the generator, as the vibrations from the generator propagate via the bearings and the drive train or via the generator support structure into the entire wind turbine and are ultimately emitted as sound. Tonal noises are particularly critical for public acceptance, as they are perceived as much more annoying than broadband noise.

The aim of this project is to investigate transmission paths where research into the potential for noise reduction is promising. In addition, many different concepts will be tested, some of which go far beyond the current state of the art. The project is being carried out in a consortium consisting of WRD/Enercon with the research partners DLR, Fraunhofer IFAM, Otto von Guericke University Magdeburg and Leibniz University Hanover.
This text was translated with DeepL

View project in the research portal

This project proposal focuses on innovative acoustic metamaterials. These are, for example, acoustically effective foam materials in which local resonance effects are to be generated by additionally introducing solid bodies with high rigidity. The aim is to significantly improve the insulating and damping effect of these materials, particularly in the low-frequency range. However, general guidelines on how an acoustic metamaterial should be designed in order to achieve the best possible and in particular a broadband effect are still lacking. The aim of the proposed project is to develop a reliable and efficient numerical tool in order to carry out a comprehensive analysis of the mechanisms, influencing factors and design parameters as well as targeted topology optimizations of acoustic metamaterials in further research work. A coupling of the finite cell method (FCM) and the boundary element method (BEM) is to be developed for the vibroacoustic analyses . The FCM is to be used for the structural-dynamic calculations in order to adequately and efficiently map the heterogeneous structure of the metamaterials. The resulting sound pressure in the surrounding air volume and the radiated sound power are used to evaluate different acoustic metamaterials. The sound radiation is calculated using the BEM, as this is an efficient way of calculating the acoustic field, particularly for evaluation in the far field, compared to volume-discretizing methods. The advantages of higher-order approach functions are also to be utilized within the scope of the project. After successful implementation, commercial FE-based calculation programs, analytical comparison solutions and experimental investigations will be used to verify and validate the developed methods in detail.
This text was translated with DeepL

View project in the research portal

Robotics has developed by leaps and bounds over the last few decades and many of the challenges of medium to large scale robotics have found suitable solutions. However, at the mesoscale, on the order of a millimeter to centimeters, few of these challenges have been addressed, chief among them, fabrication and actuation. Due to favourable scaling characteristics, piezoelectric actuation becomes more appropriate than electromagnetic actuation at small scales. Piezoelectric materials provide an actuation as they are materials that generate strain when a voltage is applied to them. They also generate a voltage when strained, which gives them the capability to operate as sensors or actuators, or both simultaneously. Due to their small total displacement, large bandwidth, and lack of friction, they have the ability to generate fast and precise movements.

The overall goal is to optimize a new class of piezoelectric motors based on a series of unimorph (a piezoelectric material bonded to a substrate) arms. The Canadian partner, Assistant Prof. Dr. Ryan Orszulik, has recently designed and fabricated a series of prototypes of a piezoelectric motor which has a planar rotor diameter of 9 mm, stator diameter of 8 mm, a total integrated motor thickness of 0.8 mm, weighs approximately 200 milligrams, and is capable of producing bidirectional motion with relatively low rotational speeds but high torque. However, a number of challenges remain, the most important of which is optimizing the torque density of the motor. For this purpose a numerical optimization will be used, which considers the mass and volume limitations, in order to achieve much higher torques without compromising structural integrity. This multi-objective optimization is a very challenging task, especially on such small scales. For mesoscale robotic applications, it is the torque that is of the greatest interest as it mitigates the need for a gearbox, which is very difficult to manufacture and integrate at these small scales. The unimorph based piezoelectric motor that is the focus of this project is simpler to construct, as it relies on non-standard planar fabrication techniques, and requires only a single drive source at a lower frequency to produce a high torque. In this research program, the goal is to leverage new fabrication techniques to create and miniaturize these piezoelectric motors, test them, and optimize them via analytical and finite element techniques. By employing the developed design, modeling, and fabrication techniques, a number of applications will be pursued including miniature autonomous vehicles and surgical instruments. The most promising possible application, which would create further opportunities for collaboration with the satellite design laboratory at York University, is to use these motors as the actuator for single gimbal control moment gyroscopes in pico to femto class satellites.

View project in the research portal

This project is a cooperation between the Chair of Multibody Dynamics and the Chair of Computational Mechanics with one research assistant from each partner. The core objective of the project is the development of a practical simulation methodology for calculating the noise emissions of engines and their psychoacoustic evaluation. This makes it possible to directly trace the effects of structural modifications (stiffness, mass distribution) and tribological system parameters (bearing clearances, viscosity, deaxialization and filling level) back to the excitation mechanisms and the internal structure-borne sound paths and to preventively combat them in terms of acoustic optimization through design and tribological measures. This purely virtual engineering approach is intended to do entirely without real prototypes and thus enable an acoustic evaluation early on in the engine development process. In this way, design measures to improve acoustic quality can be implemented in coordination with the development groups of adjacent subject areas without negatively influencing other important design criteria such as performance, pollutant emissions or total mass.
In contrast, passive measures to combat noise emissions through insulation, for example, are generally cost-intensive, as they require additional material as well as additional assembly steps and therefore have an impact on the production process. At the same time, this runs counter to the idea of lightweight construction, reduced consumption and environmental friendliness and leads to additional installation space being required, which is usually a very scarce resource in the development of modern engines and automobiles. The fundamental problem with these insulation measures, which are being used more and more frequently these days, is their symptomatic approach, which combats the effect but ignores the causes of the acoustic disturbance.
The holistic methodology that is the focus of this project, on the other hand, makes it possible to directly analyze and combat the cause of the disruptive noise emissions. In addition, the psychoacoustic evaluation of the sound emission allows it to be categorized into disturbing and less disturbing sound emissions. In this way, the design can be specifically modified so that the resulting noise is classified as more pleasant by people; after all, a quiet noise can still be perceived as more disturbing than a loud one.
This text was translated with DeepL

View project in the research portal

The aim of the project is a comprehensive investigation of elastomer pads that are used in a vehicle joint from Siemens Mobility GmbH. For this purpose, finite element analyses are carried out to qualitatively evaluate the deformation properties of the joint and in particular the installed elastomer pads. In addition, experimental investigations are to be carried out on the elastomer pads from Siemens Mobility GmbH in order to characterize the corresponding material properties more precisely. This will allow more precise correlations between material selection and structural properties to be determined in the FE analyses.

View project in the research portal

The aim of the project is to develop a method for generating simple geometries of vessel models that contain only essential information that can be used for the subsequent reconstruction of simplified simulation models for the finite element and CFD methods.

The focus here is on the geometry compression and reconstruction of the inner vessel wall with the help of parameterized NURBS. The centerline of the vessel is represented by the NURBS. Other important parameters (such as the vessel diameter, the curvature of the vessel and also the vessel thickness) are stored parameterized at the individual support points of the NURBS. In this way, the geometry is reduced to the essentials, but contains the most important information for recovering the required 3D geometry of the vessel model in a reconstruction process. This geometry can then be used for a wide variety of software systems to carry out corresponding simulations. Furthermore, it is possible to vary the parameters as required in order to generate new realistic vessel models for comparative simulations.

View project in the research portal

The eMobility Competence Center project addresses the structural challenges and develops solutions in key areas as part of a newly established competence center, which will significantly strengthen cooperation between SMEs and university research and teaching. The knowledge can be transferred directly to the affected supplier industry, where it can help to successfully manage structural change and exploit new economic opportunities. In addition to the primary objective of building up and transferring core know-how, the main focus is on the long-term anchoring of the knowledge gained in economic structures that create jobs.
Based on a multi-patented, globally unique lightweight engine concept developed by OVGU, the work in the DRIVE TRAIN research area focuses on the further development and prototypical presentation of the new engine technology, its integration into the drive train and its operation in accordance with given safety and comfort requirements (driving dynamics). At the same time, there are further innovative steps in the area of basic research to increase the performance of the engine technology, which are to be developed and implemented in prototypes during this funding period.

Content of the AR4 sub-project:
The emitted noise is a central problem of all electrical machines. This is mainly due to the fact that the typical sound emission of an electric motor is very tonal and very high-frequency and is therefore, on the one hand, in the range of the auditory surface in which humans hear best and, on the other hand, is perceived as particularly annoying. For this reason, methods and solutions are to be developed as part of this sub-project in order to significantly improve the acoustic behavior of electric machines. The aim is not only to reduce the sound pressure level but also to achieve a noise that is as unobtrusive or pleasant as possible, which is why human perception is included in the considerations. State-of-the-art commercial simulation methods and proprietary software extensions are used for the developments, as well as extensive experimental studies and listening tests. The experimental investigations include vibration analyses using laser vibrometry in a stationary and rotating system (derotator measurements), sound pressure measurements with far-field microphones and measurements with microphone arrays (acoustic camera) in an anechoic chamber. The aim of the experimental investigations is to validate the simulation models on the one hand and to demonstrate the added value of the solutions developed on the other. In addition to acoustics, the focus is on lightweight construction. The concepts to be developed should be both acoustically inconspicuous and have a minimal mass.
Among other things, alternative materials (aluminum foam structures, metamaterials, GFRP, CFRP), innovative damping strategies, novel construction designs (e.g. additive manufacturing), as well as the inclusion of add-on parts (e.g. gearboxes) in terms of additional excitation sources are investigated. Stress analyses and strength calculations are carried out to ensure that structural integrity is guaranteed despite the lightweight construction measures taken. These include both static and dynamic load cases. The dynamic stress analyses are absolutely essential in order to take account of the inertial forces acting as a result of the highly variable processes over time and the impulsive excitations during typical operating scenarios.

View project in the research portal

Welding is considered as one of the most indispensable processes in many industrial sections for joining. In many structures, welds are known as a critical sections led to mechanical failures. There are a variety of physical defects such as undercut, insufficient fusion, excessive deformation, porosity, and cracks that can affect weld quality. Of those defects, cracks are considered to be the worst since even a small crack can grow and lead to failure. All welding standards show zero tolerance for cracks whereas the other defects are tolerated within certain limits. There are three requirements for cracks to form and grow: a stress-raising defect, tensile stress, and material with low fracture toughness. Microscopic defect locations are available in practically all welds including geometric features and weld chemistry that can raise the local stress enough to induce a crack. That leaves the engineer to work with the stress environment and toughness: if either of the two can be effectively controlled then cracks can be prevented from initiating and growing. Toughness is a measure of resistance to crack growth; resistance can be provided by blunting of the crack tip in ductile materials. However, if applied strain rate is very high (as would be the case when a spot weld cools at the end of the pulse) and the stress field is multi-axial, even ductile materials exhibit poor toughness and produce rapid crack growth. Hard materials, such as martensite formed during cooling of steels, are brittle and have poor toughness. Having a deep understanding of the residual stresses in welding, micro structure and mechanical behavior of HAZ, multi axial fatigue strength, crack progress behavior and the effect of improvement techniques on welded structures will result in manufacturing more reliable and minimizing weight and increasing structural strength.
The following objectives of this project are:
- Modeling welding process by considering the phase transformation changes occurred in base and weld metal during the heating and cooling process.
- Effect of weld material strength and number of weld passes on the fatigue strength.
- Influence of heat treatment process like stress releasing, annealing hardening on fatigue behavior.
- Development of damage mechanics rules based on numerical analysis for predicting the ductile failure, fatigue life crack initiation.
- Numerical modeling of fatigue crack initiation and propagation based on phase field theory.
- Achieving experimental data by carrying out on universal servo hydraulic machine to investigate the influence of multi axial stresses on fatigue strength and fatigue life.
- The effect of residual stresses caused by welding on the fatigue life.
- Investigating HFMI process on residual stresses and fatigue strength by means of numerical and experimental work.

View project in the research portal

The aim of the BELUCCI project is to establish and validate a novel approach for the treatment of intracranial aneurysms with flow diverters, which includes individualized and simulation-based planning, implant selection/manufacturing and consultation based on patient-specific anatomical selection parameters. The project aims to develop a standardized individualization process in order to provide each patient with the optimal implant for the individual aneurysm and thus substantially improve the efficacy and safety of the procedure. The approach will be clinically evaluated as part of the project using patient-specific aneurysm models. In the sub-project at IFME, a computer-aided design tool for the numerical investigation and design of individualized flow diverters is being developed.

View project in the research portal

The Meisterbock is primarily used for new vehicle start-ups as a means of measuring and analyzing exterior components. These include sheet metal add-on parts such as fenders, doors, front and tailgates and side panels. In order to evaluate and qualify these components and their interaction in the installed state, each part is mounted on the master jig and aligned with repeat accuracy using the standardized reference point system (RPS). The aim of this project is to optimize this qualification process through the use of numerical simulation using the finite element method (FEM) in order to reduce the effort required for physical assembly and thus increase efficiency.

View project in the research portal

The aim of the project is a comprehensive parameter study as part of deformation analyses of a new type of high-performance adhesive tape. The main point here is the selection of a suitable material model for the core and the adhesive layers. In subsequent FE analyses of the face tensile test, the material parameters and layer thicknesses are varied in order to assess its influence on the overall behavior both qualitatively and quantitatively.

View project in the research portal

Fracture under thermomechanical load is a complex failure pattern that has serious consequences in materials and components. The prediction of fracture behavior through crack initiation and propagation in metals using numerical methods has become increasingly important in technical applications. The classical theories of fracture mechanics only include the criteria for crack propagation, but cannot be used to predict crack initiation. Furthermore, no statements can be made about curved cracks or crack branching. Over the past ten years, the phase field method has been transferred and further developed to describe crack formation and propagation. This method offers a powerful and flexible framework for investigating the fracture behavior of materials under arbitrarily complex thermomechanical loads. By defining an additional degree of freedom, the so-called order parameter, the crack description is carried out in the model. The heat conduction equation can also be included, for example if thermal stresses dominate the crack propagation. Both slow and sudden heating can be considered here. Analogous to the crack analysis, the temperature field is treated as an additional degree of freedom. The resulting equations can be solved using the finite element method. The aim of this doctoral thesis is to develop a model that can describe the mathematical relationship between thermomechanical loads and crack initiation and propagation at high temperatures. The starting point of the multiphysical model is formed by the constitutive equations of thermoelastoplasticity, which are solved using the phase field method. The degrees of freedom of the model include the displacement, the temperature and the phase field for the crack description.

View project in the research portal

Elastomer diaphragms are used as flat diaphragms in oscillating pumps or for pressure-actuated short-stroke actuators and control elements. Compared to metal diaphragms, elastomer diaphragms are very soft and flexible. Fabrics are often inserted into the elastomer to make elastomer diaphragms stronger and more resistant. The diaphragms are often exposed to a large number of complex and highly stressed switching cycles and must have optimum service life properties due to their important function.

Due to the complexity of elastomer membranes, it is hardly possible to reliably estimate the mechanical and service life properties based on empirical values alone. The aim of this project is to use the finite element method (FEM) to develop a simulation tool that can be used for the realistic deformation and service life analysis of fabric-reinforced elastomer membranes.

View project in the research portal

The modeling of phase fields and size effects in solids, such as the width of shear bands or the grain size dependence of the plastic flow in poly-crystals, need to be based on non-standard continuum approaches which incorporate length-scales.
With the ongoing trend of miniaturization and nanotechnology, the predictive modeling of these effects play an increasingly important role.
The mixed multi-field representation of gradient-type problems is a recently introduced thermomechanically consistent framework for modeling such kind of phenomena. The key idea is to extend the field of constitutive state variables by micromechanical independents and further to derive the macro and micro balance equations in a closed form.

View project in the research portal

For several years now, the cause of death statistics in Germany have been dominated by cardiovascular diseases. According to the Federal Statistical Office, these were responsible for around 39% of all deaths in 2015. These include strokes, which can be caused by a subarachnoid hemorrhage. This is when blood enters the subarachnoid space surrounding the brain. These bleedings are mainly caused by the rupture of cerebral aneurysms. These are balloon-like dilatations of arterial blood vessels that develop in approx. 2-6% of the Western population in the course of their lives. Approximately 10 out of 100,000 people per year experience a rupture.
Various measures are intended to prevent such a rupture. Surgical (clipping) or endovascular (coiling, balloon angioplasty, stenting, placement of flow diverters or WEB devices) interventions are used to reduce the blood flow into the aneurysm. This is aimed at the formation of thrombi, which cause a natural occlusion of the vessel. These measures are neither risk-free nor necessarily successful. This motivates the development of new procedures and the continuous improvement of established ones.
The aim of the project is to develop a stent with a novel mode of action for therapeutic deformation of the carrier vessels of intracranial aneurysms. As a result of the targeted guidance of the blood flow, more favorable hemodynamics are achieved and the blood inflow into the aneurysm interior is reduced. This in turn increases the dwell time of the blood in the aneurysm and promotes natural thrombosis, which closes the aneurysm.
This is a completely new concept in a) the treatment method and b) the necessary stent design. For this reason, the simulative methods are to be developed within this framework in order to determine the expected individual effectiveness of this concept.

View project in the research portal

FE simulations of a newly designed impact protection coupling under mechanical and thermal loads are being carried out as part of the project. This coupling consists of a modified shaft-hub connection that is to be used to transmit a constant maximum torque over several cycles.
For this purpose, validation simulations are to be implemented using the finite element method (FEM) for a simplified test setup. This axial test setup consists of two identical test specimens that are axially loaded with a contact force on a circular ring surface and then twisted against each other.
Extensive material and system characteristics as well as the practical boundary conditions are taken into account in order to enable an appropriate comparison between existing experiments and FE simulations. In addition, parameter studies are subsequently carried out in order to understand their influence on the system response. These parameters include, for example, the layer thickness and the friction coefficient. In addition to varying the contact pressure, simulations under changing temperatures are also taken into account.

View project in the research portal

Cardiovascular diseases are the main cause of death in Western countries today. There are various treatment methods for such pathologies, but the trend of the future is percutaneous minimally invasive therapy. Here, high-tech endoprostheses are inserted into the pathological area via an endoluminal path. One of the best-known families of such implants are vascular stents. They are characterized by their complex geometry and non-trivial material properties. The safe use of these stents requires continuous technological improvement in terms of material, design and operating conditions in order to achieve safe implantation, efficient drug release and optimal long-term behavior. In addition, the concept of predictive medicine, i.e. the prediction of alternative treatment methods for individual patients, is becoming increasingly important, which is not possible without robust and cost-efficient simulation methods.
This project aims to contribute to the efficient simulation of the deformation behavior of carotid stents in the carotid artery. The long-term goal is the real-time simulation of stent behavior during synchronous surgery on humans, so that various processes can be tested virtually shortly before real placement and optimally performed with regard to the individual patient.

View project in the research portal

Shape memory alloys (SMA) can undergo phase transformation between a high-ordered austenite phase and a low-ordered martensite phase, as a result of changes in the temperature and the state of stress. Consequently, SMA exhibits several macroscopic phenomena not present in the traditional materials. Two significant phenomena are the shape memory effect (SME) and the pseudoelastic effect (PE). These unique features of SMA have found important fields of applications especially in medical technology. The increasing use in commercially valuable applications have motivated a vivid interest in the development of accurate constitutive models to describe the thermomechanical behavior of SMA. In this project a thermomechanical 3D model for SMA, which includes the effect of pseudoelasticity as well as the shape memory effect will be extended with regard to fatigue behavior and crack resistance.

View project in the research portal

In this project, an in-depth investigation of the service life of technical rubber materials under multi-axis loading conditions and, in particular, under shear with rotating axes is being carried out.

In addition to experimental investigations, a theoretical concept for predicting service life is already being developed in the early phase of the project which, based on shearing with rotating axes, can cover a much wider range than previous conventional concepts.

The concept is to be validated by means of further targeted tests for double-sided shear, shear with rotating axes and under single-sided shear and tension. The load amplitude will also be varied.

View project in the research portal

The aim of the project funded by the Luxembourg Research Foundation (FNR) is the numerical investigation of the efficiency of special elastomer materials with regard to their rolling resistance properties. The so-called dynamic flocculation model (DFM) is being used and further developed for this purpose. This physically motivated material model can realistically represent the inelastic material behavior of filled elastomers under cyclic load history (e.g. Mullins effect and stress-strain hysteresis) in a large strain range. The extension of the material model to time- and temperature-dependent phenomena enables a more accurate representation of the dissipative properties of the material under dynamic loads, as they occur in rolling tires. Finally, the material model is used to establish a correlation between the dissipation that occurs and the rolling resistance, which can be used for the targeted selection of materials for tire treads.

View project in the research portal

Double-sided adhesive tapes are characterized by their viscoelastic and particularly good adhesive properties on a variety of substrates. They are either a multi-layer system consisting of a thin adhesive layer on the top and underside applied to an inner backing layer, or a single material is used that serves as both the adhesive layer and the backing material.
In this research project, a simulation tool is being developed that enables a better estimation of the application limits, taking into account the complex material characteristics, such as strong non-linearity and viscoelasticity of the material. With the help of this tool, the model parameters regarding material variation, time-dependent changes in the external boundary conditions and long-term behavior can be easily adapted and realistic predictions can be made about the complex structural behavior of single and multi-layer high-performance adhesive tapes.

View project in the research portal

The quality and informative value of FEM simulations of technical components is limited by the suitability of both the material laws used and the assigned material parameters.
 
A material model suitable for industrial application is not necessarily the most accurate and complete simulation of real material behavior. Rather, the practical suitability of a material law requires a balanced compromise between problem-specific requirements regarding the scope, accuracy and combination of properties of the material description on the one hand and economic restrictions regarding the required computer capacities and calculation times on the other.

In most cases, the corresponding material parameters are adjusted using homogeneous tests on laboratory test specimens. However, technical components and associated laboratory test specimens usually have very different geometries and are also often manufactured in different ways. In many cases, this causes serious deviations in material behavior. Component simulations with material laws that have been adapted to measurements on such test specimens are therefore prone to errors from the outset.

The core objective of the research project is the realization of a computer program suitable for industrial use for the identification of material law parameters, which enables the efficient use of measurement data from tests on test specimens close to components with inhomogeneously distributed stresses and distortions. In this way, the above-mentioned disadvantages of the restriction to homogeneous reference measurements are avoided, and the possibility is opened up to take into account specific characteristics of product groups and loading processes when adapting the material laws. The inevitable increase in computing times associated with this approach is of secondary importance given the performance of today's standard computers, provided that the potential of efficient algorithms and clever programming is fully exploited.

View project in the research portal

As part of this project, a well-founded investigation of the deformation behavior of a rolling lobe bellows under realistic load conditions is being carried out. During installation and operation, the rolling lobe undergoes large deformations, which can lead to complex contact conditions, among other things. Under operating conditions, this can lead to undesired early failure of the rolling lobe. Since the rolling lobe consists of filled elastomer, an extended material model must be used that can map the inelastic properties (such as material softening, permanent elongation and loading and unloading hysteresis).

View project in the research portal

This project, funded by the German Rubber Society (DKG), aims to increase the development potential of rubber components whose function is largely dependent on friction properties. To this end, the understanding of friction processes involving a rubber surface is to be improved on the basis of computer simulations. In particular, the significance of adhesive force components is to be reassessed and researched in detail in connection with compliant contact surface roughness. For the simulations, a model of a representative section of a contact pairing with realistic surface roughness will be created. Under contact pressure, the increase in the effective contact area due to deformation of the roughness is to be observed. A load tangential to the contact surface is then simulated. In both phases, the force components from elastic deformation, adhesion and dissipative effects are balanced.

View project in the research portal

The institute has developed a material model to describe the mechanical behavior of foamed elastomers. This model can realistically depict the extremely complex material behavior of foamed elastomers under any mechanical stress. A functional relationship between the mechanical properties and the pore content is taken into account using a homogenization approach. The material model has so far been developed for quasi-static loading conditions, i.e. time- and frequency-dependent properties of both the elastomer matrix and the pore structure cannot yet be modeled. The aim of this project is to extend the model with regard to the time-dependent properties that can occur in particular under high-frequency loads due to the spontaneous pressure build-up within the pore structure. The model will also be implemented in a suitable finite element program so that it can be used for the FE simulation of more complex, multi-dimensional load conditions.

View project in the research portal

In dem von der Deutschen Kautschukgesellschaft geförderten Projekt wird ein neues Materialmodell für thermoplastische Elastomere entwickelt, das die mechanischen Eigenschaften von TPEs, wie z.B. Inelastizität, Viskoelastizität und Temperaturabhängigkeit der Materialparameter, realitätsnah abbilden kann. Das Modell beruht auf einer Homogenisierungsmethode in der explizit der Volumengehalt der elastomeren und thermoplastischen Phase einfließt. Das Modell wird in die Finite-Elemente-Methode implementiert und kann somit in Zukunft für die realitätsnahe Simulation des Strukturverhaltens von TPE-Bauteilen benutzt werden.

View project in the research portal

The prediction of damage behaviour and residual strength as part of a damage tolerance assessment is crucial for the design, evaluation and approval of safety-relevant lightweight structures. For fibre-reinforced composites (FRP), sufficiently precise and robust methods for evaluating progressive damage have been lacking to date. The main challenge for analysing FRP structures in comparison to metallic materials is the heterogeneity of FRPs, which leads to complex failure mechanisms. A simulation methodology for strength assessment must therefore be able to depict both the initiation and progression of damage, including all the mechanisms at work and their interaction.

The aim of the DFG project is to develop an improved damage analysis methodology for FRP. For this purpose, a new adaptive solution approach is proposed, which consists of a coupling of the peridynamics (PD) for potentially damaged model areas with the FEM for the undamaged areas. The aim of the project is to increase the prediction accuracy of the load-bearing behaviour and thus to develop more robust, safe and resource-efficient structures.
PD is a promising non-local method for describing damage and dynamic crack growth, especially in brittle materials. However, the computational effort is extremely high in order to achieve a sufficiently accurate description of the crack behaviour. In order to reduce the computational effort, peridynamics is only used in those parts of a structure in which cracks can potentially occur. The remaining structural areas are modelled using the classical finite element method (FEM). In the project, suitable methods for coupling the PD with the FEM are developed, tested and evaluated and used for crack propagation. Initial good results were achieved with the Arlequin Method, the Alternating Schwarz Method and the Splice Method. The Splice Method has provided the best test results and is characterised by comparatively simple coupling equations. The software developed in the project will be made freely available (open source software) in accordance with the DFG goal of "sustainability of research software" as part of the "e-Research Technologies" funding programme in order to enable further use by other researchers.

View project in the research portal

The emitted noise is a central problem of all electrical machines. This is mainly due to the fact that the typical sound emission of an electric motor is very tonal and very high-frequency and is therefore on the one hand in the range of the auditory surface in which humans hear best, and on the other hand is perceived as particularly annoying. For this reason, this sub-project aims to develop methods and solutions to significantly improve the acoustic behavior of electric machines. The aim is not only to reduce the sound pressure level but also to achieve a noise that is as unobtrusive or pleasant as possible, which is why human perception is included in the considerations. State-of-the-art commercial simulation methods and proprietary software extensions are used for the developments, as well as extensive experimental studies and listening tests. The experimental investigations include vibration analyses using laser vibrometry in a stationary and rotating system (derotator measurements), sound pressure measurements with far-field microphones and measurements with microphone arrays (acoustic camera) in a low-noise chamber. The aim of the experimental investigations is to validate the simulation models on the one hand and to demonstrate the added value of the solutions developed on the other. In addition to acoustics, the focus is on lightweight construction. The concepts to be developed should be both acoustically inconspicuous and have a minimal mass.
Among other things, alternative materials (aluminum foam structures, metamaterials, GFRP, CFRP), innovative damping strategies, novel construction designs (e.g. additive manufacturing), as well as the inclusion of add-on parts (e.g. gearboxes) in terms of additional excitation sources are investigated. Stress analyses and strength calculations are carried out to ensure that structural integrity is guaranteed despite the lightweight construction measures taken. These include both static and dynamic load cases. The dynamic stress analyses are absolutely essential in order to take account of the inertial forces acting as a result of the highly variable processes over time and the impulsive excitations during typical operating scenarios.
This text was translated with DeepL

View project in the research portal

Chain drives are often used in petrol engines to drive camshafts, balancer shafts or pumps. During engine operation, they are subjected to complex mechanical stresses, which lead to wear between the pin and link plate in individual chain links and thus to overall elongation of the chain. Above a certain level, this elongation can no longer be compensated for by the tensioning system used. As a result, unpleasant noises such as howling or rattling occur and, in extreme cases, engine damage can also occur, e.g. if the chain jumps over individual teeth or if the chain breaks. Current wear problems make it clear that wear mechanisms occur in real customer operation that have not yet been mapped in the design.
The aim of the project is therefore to create a physically motivated wear model for chain drives that will enable improved predictions to be made about the service life of a chain drive in real customer operation. The model should basically be based on the relevant physical parameters of the tribological system (e.g. geometry, surface characteristics, hardness, residual stresses, etc.), but should also reflect the influence of key boundary conditions in engine operation such as driving profile and oil quality (e.g. viscosity, acid or soot content, etc.). With such a model, it should be possible to determine wear rates for the chain drive depending on the operating condition of the engine.
In the project, the chain wear is considered with the aid of a dynamic contact problem between pin and link plate under changing lubrication boundary conditions, with the additional complication that the contact partners pin and link plate permanently twist against each other in a certain angular range due to the rotation of the chain around the sprockets under a periodically fluctuating normal force (run force). The physical modeling leads to a complex time-dependent non-linear differential equation system that can only be solved numerically using the FEM (finite element method). The required input parameters for the calculations must be determined experimentally [1]. The time-varying loads on the pin-sleeve connection during a chain rotation are obtained with the aid of multi-body simulations. Special considerations and developments were necessary in order to (a) arrive at a physically based wear model, (b) describe the resulting non-linear contact problem between sleeve and pin appropriately [2]-[6], © develop a sufficiently accurate 3D FEM model with a reduced number of elements and (d) limit the extreme calculation times for the time sequence calculation to an acceptable level by means of suitable extrapolation without significantly worsening the accuracy of the results. Only through these developments was it possible, for example, to calculate the chain wear after an engine running time of 50,000 km. The numerically calculated chain wear is evaluated with results obtained from measurements on real vehicle chains from customer vehicles. The initial simulation results show that it is possible to obtain realistic predictions of chain wear using the new methodology developed for the wear calculation [7].

[1] Tandler, R., Bohn. N., Gabbert, U., Woschke, E.: Analytical wear model and its application for the wear simulation in automotive bush chain drive systems, Wear, Volumes 446-447, April 15, 2020, 203193, https://doi.org/10.1016/j.wear.2020.203193.
[2] Tandler, R., Bohn, N., Gabbert, U., Woschke, E.: Experimental investigation of the internal friction in automitive bush chain drives systems, Tribology International, Vol. 140 (2019), Article 105871, https://doi.org/10.1016/j.triboint.2019.105871.
[3] Gabbert, U.: Berücksichtigung von Zwangsbedingungen in der FEM mittels Penalty-Funktions-Methode, Technische Mechanik 4, 1983, Heft 2, S.40-46.
[4] Buczkowski,R., Kleiber,M., Gabbert,U.: On linear and higher order standard finite elements for 3D-nonlinear contact problems. Computers and Structures Vol. 53 (1994), No. 4, pp.817-823.
[5] Gabbert, U., Graeff-Weinberg, K.: Eine pNh-Elementformulierung für die Kontaktanalyse. ZAMM 74 (1994), 4, pp. 195-197.
[6] Buczkowski, R., Gabbert, U.: Finite element formulation of the 3D contact problem under consideration of a solidifying friction law. ZAMM Vol. 76, 1996, Supplement 5, pp. 81-83.
[7] Weinberg, K., Gabbert, U.: An adaptive pNh-technique for global-local finite element analysis. Engineering Computations: Int. J. Comp.-Aided Engng and Software, Vol. 19 No. 5, 2002, pp. 485-500.
This text was translated with DeepL

View project in the research portal

In recent years, as a result of rapid development in the fields of aerospace, optics, telecommunication, etc., demands for ultraprecision devices have been ever-increasing. A practical solution to achieve ultraprecision devices is to integrate piezo actuators and sensors  into the structure and to develope a methodology for control the accuracy of the device. Piezo materials belong to a class of so called smart materials that are capable of changing their physical properties, such as the shape, in response to an externally applied stimulus. In comparison with traditional electric motors, the smart material-based actuators have the advantages of lightness, low noise levels, low power requirements and high reliability. Therefore, they are widely used in applications of micro/nano-robots, micro-manipulation and micro/nano positioning stage. However, their characteristics (nonlinearities, badly damped vibrations, etc.) require the use of advanced control techniques. In addition to these characteristics, the particularity of working at the micro/nano-scale makes their  control even more challenging.

The current research work is control of multiple piezoelectric actuators (PEAs) in Fabry-Perot spectrometer (FPS).  The FPS can provide multispectral mappings for research on atmospheric science and planetary mineralogy, such as measuring the Earth’s O2-A band, aerosol, surface albedo, and pressure. The developed FPS uses three PEAs to provide spectral tuning of the desired optical signal transmittance by selecting the gap spacing of a tuneable optical filter. The PEAs are required to be controlled at nanometer steps with high accuracy.

One challenge we are currently working on at OVGU is the implementation of the inverse compensator for the Preisach model in F-P system. If the Preisach model is utilized to describe the hysteresis effect, it requires at least more than 10K elementary relay operators. The inverse compensator is built based on the model, which also needs more than 10K relay operators. Therefore, it causes the implementation problem when implementing the inverse compensator. To overcome this problem, we applied the model order reduction approach to reduce the number of  the relay operators in the Preisach model. In our current work, we have successfully used 200 relay operators to describe the hysteresis effect and at the same time to preserve the model accuracy. This work has been published in IEEE Transactions on Industrial Electronics (Impact factor: 7.168).  The next step will be applying the proposed approach to construct the inverse compensator for the three PEAs in the F-P system.

View project in the research portal

The Research and Development (R&D) Project COmpetence in MObility (COMO), a joint research and transfer project of the key area Automotive of the Otto-von-Guericke-Universität of Magdeburg, Germany, deals with the electrification of automobiles, including the energy supply, the energy conversion and the power train engineering as well as general problems in connection with the electro-mobility.
The project "Near-Series Drive System for Automobiles with Wheel Hub Motor, Subproject Acoustics" focuses on the improvement of the acoustics of wheel hub motors which are developed in the project. The main aim of the project is the creation of an electric wheel hub motor with an acoustic optimized behavior.As basis of the design process an overall modeling approach is developed, which is able to simulate the noise radiation of the wheel at different operation conditions. Besides the mechanical and the acoustical field such a model has also be able take into account the thermal and the electrical influences. As basis of the project experiences of previous project can be used, where an overall model for simulating the acoustic behavior of engines has been developed and successfully testet.

View project in the research portal

The overall aim of the project is to develop and test a method platform for aluminum casting, with the help of which both the technological process and the components can be optimally designed for the first time, so that a minimum component weight is achieved and at the same time the requirements with regard to mechanical properties (strength, service life, dynamics, temperature, etc.), costs and casting boundary conditions are met. The method platform is being tested using real die-cast car components.
The sub-project "Pore morphology and component strength" is concerned with determining the influence of real pores on strength and service life. To this end, the finite pore method (FPM) is coupled with computed tomography (CT) so that the pores can be recorded in the form of STL data. In addition to the use of CT data, artificially generated virtual pores and pores from casting simulations can also be taken into account in component optimization. In the future, this will make it possible to take casting criteria into account when developing optimum lightweight components manufactured using aluminum die casting.
This text was translated with DeepL

View project in the research portal

From a comfort point of view, the acoustic perception of cars plays a decisive role; at the same time, the permitted sound levels are increasingly strictly regulated. An optimal distribution of the acoustic treatments becomes therefore necessary, so that the demanded requirements are met. The prediction of the acoustic effect in the early, virtual development stages would enable to reach a better compromise with the weight, room and costs constraints. The support of this design process using simulative approaches to describe the sound transmission is hence the goal of the doctoral project. Focusing on the front-end area of a car, the main sound transmission mechanisms will be identified and it will be investigated, how each of them should be characterized. For this purpose serves the creation of simple models that allow the understanding of the appearing phenomena, as well as the comparison between measurements and simulative results. Based on this, the phenomena will be later coupled and the resulting interactions will be considered. Another objective is the integration of simulative partial models in the chain of effects of a measured complete car model.

View project in the research portal

The project is aimed to the development of virtual engineering (VE) methods and comprises the holistic computational development of a product. This process is ranged from 3D-CAD modeling, via the computation and simulation of the product characteristics up to realistic representations. Not only does this lead to a reduction of development time and costs - the quality of a product can also be increased during this process of product engineering. Component geometry, simulation models and results of computations are a basis for decisions concerning the design of a product. If the visualizations are adapted in a suitable way, it is much easier to perceive complex relations for faster identification of problems. Thus, it is necessary to create a clear structured customized presentation of the focused data sets. Investigations concerning the mechanical properties of a product provide a multitude of multi-physical resulting data. Exemplarily, the models can specify the characteristics of the product in the domain of structural mechanics, acoustics or fluid mechanics. Due to the variety of the received data structure and characteristics, the concepts of visualization have to be adapted in a suitable way. Methods representing such non-geometrical model data are developed in the paper. The multi-physical resulting data is investigated for essential information to emphasize the demands for the visualization.

View project in the research portal

The two years research project of Dr. Ryan Orszulik, financially supported by an Alexander von Humboldt Fellowship, is carried out in collaboration with Prof. Gabbert from the Otto-von-Guericke University of Magdeburg, and Prof. Monner of the DLR Braunschweig. During the last decade piezoelectric actuators have drawn considerable attention from the research community.But the development of accurate nonlinear models, specially a rate-dependent one, that is valid over a range of operating conditions is still an open problem. The limiting effect of the actuators performance is due to the asymmetrical and rate-dependent nature of the hysteresis effect. In the focus of the research project is the development of a nonlinear controller for the actuator, that can guarantee stability of the system in the presence of the hysteresis nonlinearity, is still necessary for highly precise positioning of the actuator over a large bandwidth.

View project in the research portal

Composites are of enormous importance to the industry. The usage of such materials for industrial

products has rapidly increased over the last years. Therefore, there is high interest
in gaining a better understanding of these materials and their physical behaviour. Aside
from performing experimental studies, this can also be achieved by using homogenisation
methods.
The objective of the project is to develop advanced numerical homogenisation methods
which are based on the finite element method (FEM). These methods are applicable to calculate the effective properties of unidirectional fibre reinforced composites with a periodic fibre distribution. In the developed numerical models repeated unit cells (RUCs) are used, whose cross sections can even be parallelogram shaped. The signicant advantage of these models, especially those with the parallelogram shaped cross section, is the capability to simulate a wide range of unidirectional bre reinforced composites with different fibre arrangements. This also includes the special cases of hexagonal and square fibre arrangements, which are commonly used in the literature. The numerical models are extended by employing an imperfect contact formulation between
the matrix and fibre phase to represent the presence of a very thin interphase, which is
for instance caused by chemical reactions in manufacturing processes.
The developed methods give a better insight into the material behaviour of composites as well as the modelling techniques.

View project in the research portal

The project TP3 aims to develop computational methods for the simulation and analysis of guided Ultrasonic/Lamb wave propagation in typical aircraft lightweight structures, such as sandwich structures with a heterogeneous core layer (honeycomb, open foam, closed foam. For the experimental investigation of wave propagation a 3D-Laser-Scanning-Vibrometer is applied. The project TP3 aims to explain how ultrasonic guided waves propagate in such structures, how they interact with inner boundaries and failures, such as delamination between the core layer and the top layer due to impacts, damages of the core layer, and how to design an optimal structural health monitoring (SHM) system.  For the simulation of ultrasonic guided waves a very fine spatial and temporal resolution is required, which exceeds the computational possibilities of today s commercial FEA software packages, operating with classical linear or quadratic finite elements, different new methods for an effective analysis of wave propagation in complex structures have to be developed. Consequently, the objective of the aims to develop and to test (a) Semi-analytical finite element methods, (b) Coupling of analytical and computational approaches,
© Simplification of models,(d) Higher order finite elements (spectral FEM, p-FEM, NURBS-FEM), (e) Extension and application of the finite cell method (FCM) to wave propagation analysis.

View project in the research portal

An evaluation of vehicle noise solely by measuring physical parameters is not sufficient to decide, for example, whether a noise is perceived as pleasant or annoying. However, the value of a sound has a direct influence on the purchasing decision of customers and should therefore be able to be determined objectively and influenced in a targeted manner. The aim of the doctoral project is therefore to use psychoacoustic methods, supplemented by physical measurements, to objectify acoustic target values. The relationship between physical sound events and the human perception based on them results in psychoacoustic parameters, which are examined in detail as part of the doctoral project. The focus of the project is on the objective evaluation of singular, impulsive noises, such as those produced when unlocking vehicle doors.
This text was translated with DeepL

View project in the research portal

WP1: Intake module: In this work package, a new intake module for a range extender is being developed, manufactured and experimentally tested. The intake module should absorb significantly less heat from the cylinder head and increase the air mass in the cylinder. The sub-task to be worked on here is to design the intake module in such a way that the noise emission is as low as possible.
WP2: Exhaust gas aftertreatment system: The aim of this work package is to develop, test and optimize a new concept for exhaust gas aftertreatment so that the performance, acoustic and emission criteria are optimally met. 
WP3: Oil pan: The aim of the work package is to develop, manufacture and test a new type of lightweight oil pan for a range extender. The focus of the sub-project is on the development of a lightweight structure optimized according to thermal and acoustic criteria. The finite element method is used for the calculation and optimal design. Sandwich structures made of AL foam are initially used as the lightweight construction material.
This text was translated with DeepL

View project in the research portal

The research focus COMO - COmpetence in MObility, a joint project in the
The Automotive research and transfer focus of Otto von Guericke University Magdeburg is concerned in the broadest sense with the electrification of motor vehicles, including energy supply, energy conversion and drive technology as well as fundamentally new questions in connection with electromobility. The Range Extender sub-project is dedicated to efficient energy conversion for a sufficient range. The focus of the research and development project is the question of the "optimal" range extender.b]R3: Range Extender - Acoustics sub-project aims to ensure that the very low noise and vibration level achieved by the electric drive is maintained by developing active and passive measures for noise and vibration damping. One approach that is currently being pursued is to achieve the thermal and acoustic targets of range extender operation through thermo-acoustically optimized intelligent encapsulation, which should also be characterized by minimal weight and a small installation space. The targeted weight reduction is to be achieved on the one hand through structural optimizations and on the other hand through the use of new structural concepts (e.g. aluminum foam sandwiches & plastics).
This text was translated with DeepL

View project in the research portal

Eine CO2-Reduktion von Motoren erfordert unter anderem eine Effizienzsteigerung im Antriebsstrang und im Verbrennungsmotor.  Unter anderem muß der Anteil der Motorreibung und hier insbesondere des  Kolben-Zylinder-Systems am Kraftstoffverbraucherreicht reduziert werden. Im Fokus der Promotionsprojektes steht daher die Reduktion der Motorreibung und damit die Effizienzsteigerung von Verbrennungsmotoren. Es soll untersucht werden, wie sich die vorhandenen Zylinderverformungen auf die Motorfunktion auswirken (Reibung, Blow-by, Ölverbrauch) und was die relevanten Formanteile und Effekte sind. Darüber hinaus soll mit Hilfe geeigneter numerischer Modelle unter Nutzung von Optimierungsmethoden untersucht werden, welches Potential eine Reduktion der Verzüge (Montageverzüge, warmstatische Verzüge) auf den Kraftstoffverbrauch hätte.

View project in the research portal

Das Promotionsprojektes soll einen Beitrag zur genaueren Analyse von Materialeigenschaften von Kompositstrukturen zu leisten. Komposite haben gegenüber den traditionellen (homogenen) Materialien (Stahl, Aluminium uvm.) die Eigenschaft, dass  sie je nach Anwendungsgebiet infolge der Mehrphasigkeit physikalische Eigenschaften verstärken bzw. verringern (Gewichtsreduzierung, Steifigkeitserhöhung). Zudem ist man in der Lage, diese Eigenschaften durch gezielte Phasenanordnungen innerhalb der Kompositstruktur (unidirektionale Faserverbundstoffe, Laminate) orientierungsabhängig zu verändern. Aufgrund dieser Merkmale werden diese Materialien vor allem für Leichtbauanwendungen in der Luft- und Raumfahrtindustrie, dem Automobilbau, dem Schienenfahrzegbau, der Energietechnik (Windräder) und anderen Indsutriezweigen zunehmend eingesetzt. Daher ist es von großer Bedeutung, die physikalischen Materialgrößen möglichst genau zu kennen, wozu experimentelle und auch numerische Methoden genutzt werden können. Im Promotionsprojekt sollen die die Materialeigenschaften mittels des numerischen Verfahrens der finiten Elemente (FEM) berechnet werden (virtuelles Prüflabor). Aufgrund der komplexen Geometrie wird ein mikroskopischer Ausschnitt ausgewählt, der für die Kompositbeschreibung repräsentativ ist. Unter Verwendung von Homogenisierungsalgorithmen werden anhand eines solchen repräsentativen Volumenelementes (RVE) makroskopisch homogene Steifigkeitseigenschaften abgeleitet, wozu das RVE speziellen Verformungsbedingungen unterworfen wird. Eine wichtige Bedeutung im physikalischen Modell, speziell bei der Betrachtung von Kompositen, hat der Kontaktbereich der Materialphasen. Stetigkeits- bzw. Unstetigkeitsübergänge von physikalischen Größen auf mikroskopischer Ebene können das Verhalten des Komposits (Zwei-Phasen-Komposit) auf Makroebene entscheidend verändern. Des Weiteren kann es infolge von Herstellungsprozessen dazu kommen, dass sich zwischen den Hauptphasen eine dünne Zwischenphase bildet, die das Gesamtverhalten des Komposits wesentlich beeinflussen kann. Die Berücksichtigung derartiger Phasenübergänge ist ebenfalls Teil des Promotionsprojektes.

View project in the research portal

Die zunehmende Anwendung virtueller Methoden in der Produktentwicklung erfordert eine enge Verknüpfung von Berechnungsmethoden mit Methoden zur interaktiven und möglichst auch echtzeitfähigen graphischen Repräsentation der Ergebnisse. Wünschenswert wäre eine direkte Interaktion in Echtzeit mit den Modellen. Dadurch würde es in der virtuellen Realität beispielsweise möglich, in der virtuellen Szene Parameteränderungen vorzunehmen und deren Auswirkung unmittelbar darstellen zu können. Im Promotionsprojekt werden neue Möglichkeiten für die Integration, Darstellung und Exploration von Modell- und Ergebnisdaten in der Virtuellen Realität entwickelt, um die Auswertungsqualität zu steigern und den Auswertungsprozeß zu beschleunigen. Eine Einbettung und Kombination verschiedener Berechnungssysteme kann genutzt werden, um aufwendige numerische Simulationen gezielt zu unterstützen. Unter Nutzung einer VR-Umgebung soll es möglich werden, die berechneten Simulationsdaten kombiniert mit der Visualisierung individueller Analysestrategien für den Aufbau einer virtuellen Szene zu nutzen, wobei es vorteilhaft ist, zusätzliches Wissen über das hinterlegte Simulationsmodell zu integrieren. Ein solches erweitertes virtuelles Modell bietet sehr gute Möglichkeiten für einen echtzeitfähigen interaktiven Eingriff in das  Modell, wodurch beispielsweise Optimierungsaufgaben sehr effizient gelöst werden könnten.

View project in the research portal

The aim of the interdisciplinary project is the application of virtual methods for the holistic development and optimization of electric vehicles. The sub-project being worked on by the Chair of Computational Mechanics is aimed at hybrid modeling, simulation (numerical and analytical methods) and the holistic presentation of a wide range of physical fields in virtual reality (VR). Work is currently underway to transfer multiphysical data from finite element analyses (mechanical stresses, deformations, temperatures, velocities, sound pressures, magnetic fields, electric fields, etc.) into virtual reality.
This text was translated with DeepL

View project in the research portal

By using multifunctional materials, such as piezoelectric ceramics, intelligent lightweight structures can be developed that have the ability to adapt independently to changing operating conditions and external influences in order to positively influence the vibration and radiation behavior. The aim of the research project is to develop a numerical calculation method for the effective design and complete simulation of intelligent lightweight structures, including electromechanical and vibroacoustic coupling. For this purpose, FE-based modeling approaches that describe the behavior of passive basic structures, piezoelectric energy converters and surrounding acoustic interior and exterior spaces are to be linked within the framework of a coupled overall system. The focus of the research project is on the numerical mapping of infinitely extended fluid areas. For this purpose, the properties of the Perfectly Matched Layer methods in free-field modeling are to be investigated and further developed with regard to the inclusion of control methods. In order to analyze the mode of action and possible applications of the Perfectly Matched Layer methods in model-based controller design, the electromechanically-acoustically coupled overall model of the controlled system is prepared in such a way that it is included as a subsystem in the mathematical description of the control loop. This results in a generally valid approach with which both feedback control and feedforward control can be simulated efficiently.This text was translated with DeepL

View project in the research portal

The objective of the subproject is the modelling and analysis of the of Lamb wave generation  in the piezoelektric actor array, the propagation of the Lamb waves through the structure (undamaged and damaged) and the receiving of the Lamb waves in the sensor array.  Methods are under progress to support an optimal design of such a structural health monitoring systems as well as the signal processing.

View project in the research portal

Guided ultrasonic waves in thin fiber composite structures - such waves are also known as Lamb waves after their discoverer Horace Lamb - propagate over greater distances and, due to their short wavelengths, also interact with minor structural damage. Such waves can therefore be used advantageously for structural health monitoring (SHM). In order to calculate the propagation of such high-frequency waves, an extremely large number of conventional linear finite elements (at least 20 elements per wavelength) are required in the context of explicit time integration, so that such calculations for real structural components quickly reach the performance limits of current computing technology. Therefore, the focus of the dissertation project is on the development and investigation of various higher-order finite elements (polynomial degree p ≥ 3). Specifically, these are p-elements based on the normalized integrals of Legendre polynomials, spectral elements based on Lagrange polynomials and isogeometric finite elements based on non-uniform rational B-splines (NURBS).
This text was translated with DeepL

View project in the research portal

The dissertation project is aimed to a Lamb wave based structural health monitoring of sandwich structures with a cellular core layer. Such structures are frequently used as lightweight structures in many applications due to their high strength and stiffness behavior. Because of their complex build-up the understanding of Lamb wave generation and propagation is much more difficult than in the case of conventional composite structures. In the project the finite element method is applied to analyze the Lamb wave propagation at different frequency ranges and in sandwich panels with different build-ups to better understand the details of wave propagation. Different modeling approaches are applied, compared and evaluated, such as the application of a homogenized core layer in connection with a 3D finite element model of the cover layers, or the application of a discrete modeling of the core layer (honeycomb, foam) with shell type finite elements, where the cover layers are again modeled with 3D finite elements, etc. The objective of the project is to develop an effective modeling approach which can be used to design SHM systems for sandwich types of structure with different core materials.

View project in the research portal

Im Mittelpunkt des Projektes steht Modellierung der Lambwellenausbreitung in dünnen, aus mehreren orthotropen Schichten bestehende Faserverbundstrukturen. Die Wellen sind dispersiv und werden durch die Mikrostruktur (Faser- und Lagenaufbau) und Strukturschäden beeinflusst, wodurch Modekonversionen und Wellenreflektionen verursacht werden. Zur Modellierung der Wellenausbreitung werden analytische und numerische Methoden entwickelt, getestet und bewertet.

View project in the research portal

Mit dem Projekt wird ein interdisziplinärer wissenschaftlicher Beitrag zur Weiterentwicklung von numerischen und experimentellen Methoden der Mechanik zur Schwingungs- und Geräuschreduktion von PKW-Komponenten (Motoren, Karosserie, Einbauteile) geleistet. Der Fokus des Projektes liegt auf aktiven Maßnahmen zur Reduktion der Schallabstrahlung von flächigen Komponenten, wie z.B. der Ölwanne.
Projektpartner sind Prof. H. Tschöke und Prof. R. Kasper vom Institut für Mobile Systeme der OvGU.

View project in the research portal

Ziel des interdisziplinären Projektes ist die Entwicklung von durchgängigen Modellierungskonzepten zur Simulation komplexer mechatronischer Systeme aus dem Bereich Automotive unter Einbeziehung von VE und VR Techniken. Projektpartner sind Dr. U. Schmucker als Projektkoordinator vom Fraunhofer-Institut für Fabrikbetrieb und -automatisierung (IFF) Magdeburg, Prof. R. Kasper vom Institut für Mobile Systeme der OvGU Magdeburg sowie Prof. G. Saake vom Institut für Technische und Betriebliche Informationssysteme der OvGU Magdeburg.

View project in the research portal

Das Ziel des Projektes ist die Entwicklung von Methoden und Verfahren, mit denen aus CAD-Daten sowie aus Simulationsergebnissen eine Visualisierung und Interaktion im Rahmen der VR (virtuelle Realität) möglich ist. Ein Schwerpunkt ist die Aufbereitung und Visualisierung nichtgeometrischer Daten für den Nutzer. Das Teilprojekt befaßt sich zunächst mit der Übertragung multiphysikalischer Daten aus Finite-Element-Analysen (mechanische Spannungen, Verformungen, Temperaturen, Geschwindigkeiten, Schalldrücke, Magnetfelder, elektrischer Felder u.ä.)  in die virtuelle Realität.

View project in the research portal

Das Ziel des Dissertationsprojektes ist die Entwicklung eines semi-analytischen Verfahrens (SAFE) zur Simulation der Lambwellenausbreitung in Faserverbundstrukturen sowie die Berechnung von Dispersionsdiagrammen Reflektions- und Transmissionskoeffizienten.

View project in the research portal

The winding of thin multilayer films is simulated using FEM.
This text was translated with DeepL

View project in the research portal

Das Teilprojekt leistet einen Beitrag zur Entwicklung von kostengünstigen Hohlprofilen aus Faserverbundmaterial für die Massenproduktion. Der Schwerpunkt liegt auf der Entwicklung und Anwendung von numerischen Methoden (Finite-Element-Methode) für einen zuverlässigen und sicheren Entwurf der Hohlprofile. .

View project in the research portal

Das Ziel des Projektes ist es, gemeinsam mit dem industriellen Projektpartner SYMACON eine optimale metallische Endverbindung für Glasfaserstäbe der Firma Polystal zu entwickeln. Neben experimentellen Methoden werden begleitende Simulationen unter Nutzung der Finite-Element-Methode (Abaqus) genutzt, um zu einer optimalen Gestaltung der Verbindungelemente zu gelangen.

View project in the research portal

Das Ziel des Forschungsverbundes aus universitären und industriellen Partnern ist die Entwicklung eines VR basierten Operationssimulators für die laparoskopische Chirurgie. Mit dem Projekt sollen die Voraussetzungen für die Entwicklung und Anwendung interaktiver, digitaler Visualisierungs- und Simulationstechniken im medizinischen Bereich zur besseren Behandlung von Patienten geschaffen werden. Der Schwerpunkt des Teilprojektes des Lehrstuhls für Numerische Mechanik der Otto-von-Guericke-Universität Magdeburg liegt auf der Entwicklung echtzeitfähiger Organmodelle, die das Verhalten beim operativen Eingriff im virtuellen Raum realitätsnah abbilden.

View project in the research portal

Das Ziel des Projektes ist die Entwicklung numerischer Berechnungsmodelle für die Simulation des Verformungsverhaltens von faserverstärkten Kunststoffschläuchen und Hohlprofilen. Die Modellentwicklungen basiert auf der Finite-Element-Methode. Die theoretischen Arbeiten werden begleitet durch experimentelle Untersuchungen zum Materialverhalten, zum globalen Verformungsverhalten und zur Verifikation der Berechnungsmodelle. Für die Entwicklung von Materialmodellen werden Homogenisierungsmethoden auf der Grundlage von repräsentativen Volumenmodellen (RVE) eingesetzt.

View project in the research portal

Neue Hochleistungsverbundwerkstoffsystem eröffnen interessante neue Möglichkeiten für die Gestaltung extrem leichter und dabei hochfester Bauteile und Tragwerke. Allerdings erfordert die Ausweitung des Einsatzes derartiger Materialien auf neue Anwendungsbereiche im Maschinenbau, in der Fördertechnik, in der Robotertechnik, in der Medizintechnik usw. zuverlässige Richtlinien für den Entwurf und die Berechnung, die bisher nicht zur Verfügung stehen. Neben der Vielzahl unterschiedlicher Fasermaterialien und Harz-Härter-Systeme gibt es durch die Wahl des Lagenaufbaus und der Faserorientierungen einen großen Freiheitsgrad bei der Gestaltung derartiger Strukturbauteile, wodurch sich der Entwurf und die zuverlässige Dimensionierung als ungleich komplizierter darstellen als beispielsweise die Auslegung einer vergleichbaren metallischen Struktur. Ziel des Projektes ist es, einen Beitrag zur Entwicklung zuverlässiger Richtlinien für den Entwurf und die Berechnung neuer Hochleistungsverbundwerkstoffe aus glasfaserverstärkten Kunststoffen zu leisten, die durch diskrete Glasfaserliner, wie sie beispielsweise von der Firma Polystal produziert werden, verstärkt sind. Die sich daraus ergebenden Konsequenzen für das Tragverhalten von Faserverbundstrukturen und die erforderlichen Berechnungsgrundlagen sollen untersucht werden, wobei auch das Schädigungsverhalten und die Entwicklung schadenstoleranter Strukturen in die Untersuchungen einbezogen werden sollen.

View project in the research portal

Das Ziel des Projektes ist es, numerische Berechnungsmethoden zu entwickeln, die es unter Nutzung der Methode des repräsentativen Volumenelementes (RVE) und der Finite-Element-Methode weitgehend automatisch ermöglichen, homogenisierte Materialeigenschaften für faser- und partikelverstärkte Werkstoffsysteme zu gewinnen. Es wurden Homogenisierungsmethoden für piezoelektrische Langfasersysteme sowie für kurzfaser- und partikelverstärkte Kunststoffe entwickelt, wobei insbesondere Materialien mit zufällig verteilten Naturfasern und Hohlkugeln betrachtet wurden. Für die automatische Durchführung der Homogenisierung wurde eine Reihe von Softwaretools unter Nutzung von ANSYS entwickelt und erfolgreich getestet.

View project in the research portal

Das Projekt zielt auf die Anwendung der Finite-Element-Methode (FEM) für die Modellierung und Berechnung dünnwandiger Leichtbaustrukturen aus faserverstärkten Kunststoffen mit applizierten piezoelektrischen Patchen als Aktoren und Sensoren für die Formkontrolle und die Schwingungsdämpfung. Dünnwandige Strukturen reagieren empfindlich auf äußeren Störungen, wobei häufig große elastische Verformungen verursacht und die Grenzen der Theorie kleiner Verformungen überschritten werden. Das wesentliche Ziel des Projektes ist es, eine neues finites Schalenelement zu entwickeln, das den Einfluß moderat großer Verformungen (von Karman Theorie) auf das aktive und passive Verhalten adaptiver Strukturen bei statischen und dynamischen Anwendungen berücksichtigt.

View project in the research portal

Das Ziel des Kooperationsprojektes mit der Siemens AG ist es, einen Beitrag zur Entwicklung von Entwurfsmethoden für die aktive Schwings- and Lärmreduktion von technischen Sytemen zu leisten, die sich durch eine ausreichende Robustheit und hohe Zuverlässigkeit auszeichnen. Die Forschungsarbeit erfolgt in enger Verbindung von theoretischer Entwicklung und experimenteller Erprobung.

View project in the research portal

Im Rahmen des Forschungsprojektes erfolgt die Entwicklung einer Optimierungssoftware auf der Grundlage von Evolutionsstrategien zur optimalen Auslegung von mechanischen Strukturen, die aus Materialien mit Mikrostruktur bestehen. Das Ziel besteht darin, Designparameter auf der Mikroebene (Materialsystem) so zu verändern, daß auf der Makroebene (Struktur) ein gewünschtes optimales Verhalten erreicht wird. Die Wahl der Designparameter und der Zielfunktion soll weitgehend problemunabhängig erfolgen können. Ein dafür geeigneter schneller Algorithmus läßt sich unter Nutzung der Methode des repräsentativen Volumenelementes entwickeln. Grundlage der Optimierung ist die Finite-Element-Methode, die für die numerische Berechnung der Zielfunktion eingesetzt wird.

View project in the research portal

Das Ziel des Projektes besteht in der Entwicklung eines numerischen Verfahrens zum Entwurf und zur Simulation intelligenter Leichtbaustrukturen unter Einbeziehung vibroakustischer Kopplungen. Als aktive Materialien werden piezokeramische Patche auf die Struktur appliziert. Die Grundlage für den Entwurf ist ein virtuelles Gesamtmodell, das alle wesentlichen Teilkomponenten erfaßt. Dies sind die Struktur, die piezokeramischen Sensoren und Aktoren, das akustische Fluid (Unterscheidung zwischen innerem und äußerem Abstrahlproblem), die vibroakustische Kopplung und die Regelung. Für die Entwicklung eines derartigen virtuellen Gesamtmodells wird die Finite-Element-Methode benutzt. Die theoretische Lösungsansätze werden numerisch getestet und experimentell verifiziert.

View project in the research portal

Als Basis einer Neuentwicklung ist ein geeignetes virtuelles Computermodell erforderlich, das neben der mechanischen Struktur zusätzlich auch die Aktoren, Sensoren und die Regler umfassen muß. Unter Nutzung der Finite-Element-Methode lassen sich geeignete virtuelle Modelle entwickeln. Auf der Basis derartiger Modelle wurden optimale LQ-Regler mit zusätzlicher Dynamik entwickelt und getestet. Existiert ein Prototyp, kann ein Modell mittels der experimentellen Systemidentifikation gewonnen werden. Alternativ lassen sich auch adaptive Reglerkonzepte nutzen. Die Anwendung und Testung der Reglerkonzepte erfolgte an Hand einfacher Teststrukturen sowie an der trichterförmigen Schalenstruktur eines Kernspintomographen mittels einer "Hardware-in-the-Loop" Realisierung unter Nutzung von dSPACE.

View project in the research portal

Um das Verformungsverhalten von Kunststoffschläuchen, die durch Gewebeeinlagen verstärkt sind, berechnen zu können, wurde ein erster Ansatz für die Modellbildung entwickelt. Für den Kunststoff wurde ein Mooney-Rivlin-Materialmodell identifiziert und die Gewebeeinlagen wurden durch ein anisotropes Materialmodell beschrieben. Die Materialparameter wurden sowohl experimentell gewonnen als auch mit Hilfe eines geeignet gewählten RVE (repräsentatives Volumenelement) auf numerischem Wege durch Homogenisierung ermittelt.

View project in the research portal

Das untersuchte Bauteil besteht aus einem Kunststoff-Verbundmaterial mit angefügten Befestigungselementen zur Integration in ein übergeordnetes Objekt und zur Aufnahme von weiteren Anbauteilen. Aufgrund des komplizierten geometrischen und werkstoffseitigen Aufbaus sind derartige Verbundbauteile besonders sensibel hinsichtlich ihres dynamischen und akustischen Verhaltens. Als Grundlage für die Entwicklung und die spätere Verifikation von Simulationsmodellen zur numerischen Analyse wurde das dynamischen Verhalten zunächst mit Hilfe der experimentellen Modalanalyse untersucht und dabei das Eigenschwingverhalten und das Verhalten bei verschiedenen Anregungen mit Hilfe moderner Laservibrometer-Messtechnik aufgenommen und ausgewertet.

View project in the research portal

Das untersuchte komplexe Bauteil besteht aus einem Kunststoff-Verbundmaterial mit angefügten Befestigungselementen zur Integration in ein übergeordnetes Objekt und zur Aufnahme von weiteren Anbauteilen. Aufgrund des komplizierten geometrischen und werkstoffseitigen Aufbaus sind derartige Verbundbauteile besonders sensibel hinsichtlich ihres dynamischen und akustischen Verhaltens. Im Rahmen des Projektes wurde ein komplexes numerisches Modell (Finite-Element-Modell) hinsichtlich seiner Eignung für die Schwingungsberechnung untersucht. Vor besonderer Bedeutung war dabei die Einbeziehung realistischer Materialparameter und hier insbesonderte die Berücksichtigung der Dämpfungseigenschaften.

View project in the research portal

Die Erfindung betrifft die Integration und die ganzheitliche regelungstechnische Verknüpfung unterschiedlicher sowohl lokal als auch global wirkender Aktoren zum hochgenauen Positionieren, Halten oder Unterstützen von Objekten. Die Erfindung zielt insbesondere auf die Schwingungsreduktion und die Erhöhung der Fertigungsgenauigkeit von hochpräzisen, magnetgelagerten Werkzeugmaschinen ab.

View project in the research portal

Die wesentlichen wissenschaftlichen und technischen Ergebnisse des Teilvorhabens "Modellierung" des BMBF-Leitprojektes ADAPTRONIK sind:

• Entwicklung neuer numerischer Methoden auf der Basis der FEM zur ganzheitlichen Simulation von adaptiven geregelten Verbundstrukturen mit strukturkonform integrierten Faser und Folien,
• Koppelung von Finite-Element-Analyse (FEA-Software COSAR) und Regelung (Matlab/Simulink),
• Entwicklung von neuen Optimierungsmethoden für die Vorausberechnung der Form, der Anzahl und des Ortes der Plazierung von piezoelektrischen Sensoren und Aktoren unter Einbeziehung der Regelung,
• Anwendung der entwickelten Softwaretools für den optimalen Entwurf industrieller Prototypen: (i) adaptives Dachblech eines VW-Bora, (i) Trichter eines Siemens MRT, (iii) Leichtbauzylinder aus Kohlefaser und (iv) vieler weiterer Teststrukturen.

View project in the research portal

[img]http://www.ttz.uni-magdeburg.de/fodb/bilder/uni_md_adames.jpg[/img]

View project in the research portal