Search Results

It is essential to include uncertainties in the simulation process in order to perform reliable vibroacoustic predictions in the early design phase. In this contribution, uncertainties are quantified using the generalized Polynomial Chaos (gPC) expansion in combination with a Finite Element (FE) model of a vehicle body in white. It is the objective to particularly investigate the applicability of the gPC method in the industrial context with a high number of uncertain parameters and computationally expensive models. A non-intrusive gPC expansion of first and second order is implemented and the approximation of a stochastic response process is compared to a Latin Hypercube sampling based reference solution with special regard to accuracy and computational efficiency. Furthermore, the method is examined for other input distributions and transferred to another FE model in order to verify the applicability of the gPC method in practical applications.

During the last decades, big steps have been taken towards a realistic simulation of NVH (Noise Vibration Harshness) behavior of vehicles using the Finite Element (FE) method. The quality of these computation models has been substantially increased and the accessible frequency range has been widened. Nevertheless, to perform a reliable prediction of the vehicle vibroacoustic behavior, the consideration of uncertainties is crucial. With this approach there are many challenges on the way to valid and useful simulation models and they can be divided into three areas: the input uncertainties, the propagation of uncertainties through the FE model and finally the statistical output quantities. Each of them must be investigated to choose sufficient methods for a valid and fast prediction of vehicle body vibroacoustics. It can be shown by rough estimation that dimensionality of the corresponding random space for different types of uncertainty is tremendously high.

This paper describes a novel approach for modeling an automotive HVAC unit. The model consists of black-box models trained with experimental data from a self-developed measurement setup. It is capable of predicting the temperature and mass flow of the air entering the vehicle cabin at the various air vents. A combination of temperature and velocity sensors is the basis of the measurement setup. A measurement fault analysis is conducted to validate the accuracy of the measurement system. As the data collection is done under fluctuating ambient conditions, a review of the impact of various ambient conditions on the HVAC unit is performed. Correction models that account for the different ambient conditions incorporate these results. Numerous types of black-box models are compared to identify the best-suited type for this approach. Moreover, the accuracy of the model is validated using test drive data.

Highly automated driving (HAD) is under rapid development and will be available for customers within the next years. However the evidence that HAD is at least as safe as human driving has still not been produced. The challenge is to drive hundreds of millions of test kilometers without incidents to show that statistically HAD is significantly safer. One approach is to let a HAD function run in parallel with human drivers in customer cars to utilize a fraction of the billions of kilometers driven every year. To guarantee safety, the function under test (FUT) has access to sensors but its output is not executed, which results in an open loop problem. To overcome this shortcoming, the proposed method consists of four steps to close the loop for the FUT. First, sensor data from real driving scenarios is fused in a world model and enhanced by incorporating future time steps into original measurements.

In the scope of a research collaboration, ITT Automotive Europe and Darmstadt University of Technology are developing control strategies for a low-cost Brake-by-Wire system. However, since there is a wide range of variation in the efficiency of the gear units used in electromechanical brakes, this becomes a demanding task. The paper first describes the assembly and operation of ITT's early generation brake actuator. It introduces a model of the electromechanical brake with its structure and subsystems as a major tool in the development process. A detailed analysis of the signals, already available from the brake and the vehicle, is discussed for their advantages and disadvantages with regard to a possible use in the controller design. Different approaches for clamping-force, peripheral-force and brake-torque sensing are compared. An integrated clamping force sensor for feedback control of prototype actuators was developed.

Faced with the need to reduce development time and cost, the hardware-in-the-loop (HIL) simulation increasingly proves to be an efficient tool in the automotive industry. It offers the possibility to investigate new engine control systems with fewer expensive engine dynamometer experiments and test drives. In the scope of a research collaboration, Daimler Benz and Darmstadt University of Technology are developing a hardware-in-the loop simulator for the investigation of the electronic engine management of the new Mercedes Benz truck engine series 500 and 900. This paper first describes the necessary models for real-time simulation of the subsystems Diesel engine, turbo charger and vehicle. Then the setup of the simulator test bench is introduced and the performance of the simulator is demonstrated by several experimental results.

This paper deals with physical modelling of a turbocharger with variable turbine geometry (VTG) and its real-time simulation based on dynamic artificial neural networks (ANN). Thermodynamic und fluiddynamic equations, describing the basic functionality and relations between pressure, mass flow and temperature at the inlet and outlet ports of compressor and turbine, build up a multiple input multiple output model (MIMO). A special kind of ANN, namely the LoLiMoT algorithm, is used for real-time simulation. Training the network using measurement and simulated data, the dynamic behaviour can be simulated with less computational effort than the physical model. The neural network may be used in engine control systems as observer for non measurable signals, like rotor speed or turbine and compressor torque, figure 1.

In the present work we investigated experimentally and computationally the unsteady flow around a BMW car model including wheels*. This simulation yields mean flow and turbulence fields, enabling the study aerodynamic coefficients (drag and lift coefficients, three-dimensional/spatial wall-pressure distribution) as well as some unsteady flow phenomena in the car wake (analysis of the vortex shedding frequency). Comparisons with experimental findings are presented. The computational approach used is based on solving the complete transient Reynolds-Averaged Navier-Stokes (TRANS) equations. Special attention is devoted to turbulence modelling and the near-wall treatment of turbulence. The flow calculations were performed using a robust, eddy-viscosity-based ζ - ƒ turbulence model in the framework of the elliptic relaxation concept and in conjunction with the universal wall treatment, combining integration up to the wall and wall functions.

This paper describes a practical approach to extract the global static stiffness of a body in white (BIW) from dynamic measurements in free-free conditions. Based on a limited set of measured frequency response functions (FRF), the torsional and bending stiffness values are calculated using an FRF based substructuring approach in combination with inverse force identification. A second approach consists of a modal approach whereby the static car body stiffness is deduced from a full free-free modal identification including residual stiffness estimation at the clamping and load positions. As an extra important result this approach allows for evaluating the modal contribution of the flexible car body modes to the global static stiffness values. The methods have been extensively investigated using finite element modeling data and verified on a series of body in white measurements.

For the development of sophisticated digital (e.g., Finite-Element-models like CASIMIR) or physical (e.g.,ASPECT-Dummy) models of the mechanisms of human-seat-interaction it is very important to know the shape of the contact surface between the human buttocks and back and the seat cushion and backrest, respectively. Currently, these surfaces are usually determined by purely experimental procedures, that require complicated and expensive measuring equipment. This paper presents an alternative hybrid approach of standard experimental investigations of the pressure distributions between human body and seat (cushion and backrest) and proceeding numerical simulations with the Finite-Element-Method (FEM). Pressure distributions are measured with standard measuring equipment for individual persons or defined percentile groups. Due to the simplicity of these measurements, they can be performed for a larger number of individuals at low cost.

The aim of the project is to reliably identify the set of constructive features responsible for the highest noise levels in the interior of motor vehicles. A simulation environment based on artificial intelligence techniques such as neural networks and genetic algorithms has been implemented. We used a system identification approach in order to approximate the functional relationship between the target noise series and the sets of constructive parameters corresponding to the cars. The noise levels were measured with a microphone positioned on the driver''s chair, and corresponded to a variation of the engine rotation of 600-900 rot/min. The database includes 45 different cars, each described by vectors of 67 constructive features.

A joint study was conducted between BMW and Goodyear with the objective of analysing the cause and identifying methods to reduce the structure-borne interior noise in a vehicle driving on rough road surfaces. A vibro-acoustic characterization of the car was performed by measuring the car vibro-acoustic transfer functions and by using a transfer path analysis technique to identify the main suspension parts affecting the interior noise at target frequencies. The vibration transmissibility characteristics of the tire were measured and also simulated by Finite Element in [1-200Hz] frequency range. The vibro-acoustic interaction between the tire and car sub-systems was examined. A Finite Element sensitivity analysis was used to define and build new prototype tires. A 3dB(A) interior noise improvement was obtained with these new tires at target frequencies.

The use of model based approaches in areas such as simulation, control design, optimization, etc. is crucial for the development of highly sophisticated systems. This is especially true for typically very tight time-to-market frames. Physical modeling of IC engine emissions based on first principles is extremely complex and still requires by far too much calculation time. However, special fast neural networks represent a promising alternative for an accurate modeling of the emission behavior, even for dynamic conditions. This paper first describes the process of developing dynamic neural emission models. The required data is collected by a specially designed dynamic measurement strategy. The models themselves are then used for the optimization of the dynamic engine behavior concerning consumption, emissions and drivability.

Fault detection is increasingly an essential part of vehicle development. Integrating such fault detection subsystems raises the reliability, maintainability, and safety of automobile components. Weak shock absorbers can lead to significantly longer braking distances (up to 20%) and furthermore worsen the driving handling. Reduced tire pressure increases the wear of the tire dramatically and may lead to punctures due to an overheating of the tire. Recent studies show that 40% of all drivers have set wrong tire pressures (Wachter, 1994). Therefore, this paper presents fault detection algorithms for the suspension system implemented on a Hitachi SH7055 microcontroller. Real measurements of a vehicle are made to proof the algorithms.

The Advanced Lighting Simulation (ALS) is a development tool for systematically investigating and optimizing the Adaptive Light Control (ALC) system to provide the driver with improved headlamps and light distributions. ALS is based on advanced CA-techniques and modern validation facilities. To improve night time traffic safety the BMW lighting system ALC has been developed and optimized with the help of ALS. ALC improves the headlamp illumination by means of continuous adaptation of the headlamps according to the current driving situation and current environment. BMW has already implemented ALC prototypes in real vehicles to demonstrate the advantages on the real road.

Due to more severe exhaust gas regulations with sharper exhaust gas limitations and rising requirements for on-board diagnosis in this contribution a method for injection mass supervision in diesel engines using a fast broadband oxygen sensor will be presented. Based on a physical model the injected fuel mass can be determined by evaluating the measured air mass and oxygen concentration in the exhaust gas. Cylinder individual injection mass calculation becomes possible using an inverse model of the oxygen sensor dynamic. Thereby the sensor dynamic is specified by evaluating step responses of the oxygen concentration at jumps of the injection mass. For cylinder assignment the runtimes of the exhaust gas in the exhaust pipe have to be determined. They result from the calculation of the cross correlation function of the reconstructed fuel mass and measured mean indicated cylinder pressure.

Methods for model-based fault detection are presented which detect a wide range of faults using only common production sensors, namely air mass sensor, manifold pressure sensor, manifold temperature sensor and engine speed. Five suitable reference models for fault detection are set up and identified at the test stand. The developed fault detection algorithms use the dependencies of the four sensor signals based on the reference models. Thereby five residuals and five symptoms are calculated. The model-based fault detection algorithms are implemented with a dSPACE Rapid Control Prototyping system and verified at the test stand. Measurements of online fault detection are shown.

The Energy and Environment Test Centre (EVZ) is a complex comprising three large climatic wind tunnels, two smaller test chambers, nine soak rooms and support infrastructure. The capabilities of the wind tunnels and chambers are varied, and as a whole give BMW the ability to test at practically all conditions experienced by their vehicles, worldwide. The three wind tunnels have been designed for differing test capabilities, but share the same air circuit design, which has been optimized for energy consumption yet is compact for its large, 8.4 m₂, nozzle cross-section. The wind tunnel test section was designed to meet demanding aerodynamic specifications, including a limit on the axial static pressure gradient and low frequency static pressure fluctuations - design parameters previously reserved for larger aerodynamic or aero-acoustic wind tunnels. The aerodynamic design was achieved, in-part, by use of computational fluid dynamics and a purpose-built model wind tunnel.

Over the past decades, interior noise from wind noise or engine noise have been significantly reduced by leveraging improvements of both the overall vehicle design and of sound package. Consequently, noise sources originating from HVAC systems (Heat Ventilation and Air Conditioning), fans or exhaust systems are becoming more relevant for perceived quality and passenger comfort. This study focuses on HVAC systems and discusses a Flow-Induced Noise Detection Contributions (FIND Contributions) numerical method enabling the identification of the flow-induced noise sources inside and around HVAC systems. This methodology is based on the post-processing of unsteady flow results obtained using Lattice Boltzmann based Method (LBM) Computational Fluid Dynamics (CFD) simulations combined with LBM-simulated Acoustic Transfer Functions (ATF) between the position of the sources inside the system and the passenger’s ears.

A local Gaussian process regression approach is presented, which allows to model nonlinearities of internal combustion engines more accurate than global Gaussian process regression. By building smaller models, the prediction of local system behavior improves significantly. In order to predict a value, the algorithm chooses the nearest training points. The number of chosen training points depends on the intensity of estimated nonlinearity. After determining the training points, a model is built, the prediction performed and the model discarded. The approach is demonstrated with a benchmark system and air charge test bed measurements. The measurements are taken from a turbocharged SI gasoline engine with both variable inlet valve lift and variable inlet and exhaust valve opening angle. The results show how local Gaussian process regression outmatches global Gaussian process regression concerning model quality and nonlinearities in particular.