Search Results

In the early stages of vehicle development, it is critical to establish performance goals for the major systems. The fundamental modes of body and chassis frames are typically assessed using FE models that are discretized using shell elements. However, the use of the shell-based FE method is problematic in terms of fast analysis and quick decision-making, especially during the concept phase of a vehicle design because it takes much time and effort for detailed modeling. To overcome this weakness, a one-dimensional (1D) method based on beam elements has been extensively studied over several decades, but it was not successful because of low accuracy for thin-walled beam structures. This investigation proposes a 1D method based on thin-walled beam theory with comparable accuracy to shell models. Most body pillars and chassis frame members are composed of thin-walled beam structures because of the high stiffness-to-mass ratio of thin-walled cross sections.

Recently, papers have been published purporting to study the effect of rear axle tramp during tread separation events, and its effect on vehicle handling [1, 2]. Based on analysis and physical testing, one paper [1] has put forth a mathematical model which the authors claim allows vehicle designers to select shock damping values during the development process of a vehicle in order to assure that a vehicle will not experience axle tramp during tread separations. In the course of their work, “lumpy” tires (tires with rubber blocks adhered to the tire's tread) were employed to excite the axle tramp resonance, even though this method has been shown not to duplicate the physical mechanisms behind an actual tread belt separation. This paper evaluates the theories postulated in [1] by first analyzing the equations behind the mathematical model presented. The model is then tested to see if it agrees with observed physical testing.

Current CAE modeling and simulation techniques in the time domain allow, by now, very accurate prediction of many ride-comfort performances of the cars. Nevertheless, the prediction of the steering wheel rotation vibration excited by, for instance, wheel unbalance or asymmetric obstacle impact, often runs into the difficulty of modeling the steering line with sufficient accuracy. For a classic rack and pinion, hydraulic assisted steering line, one of the challenges is to model the complex and non linear properties - stiffness, friction and damping - of the rack-rack case system. This paper proposes a rack model, thought for easy implementation in complex multi-body models, and an identification procedure of its parameters, based on measurements, in the operational range of the wheel unbalance excitation. The measurements have been gathered by specific tests on the components and the test set-up is also shown here.

Due to complexities of interaction among gears and crescent-shaped island, a crescent oil pump is one of the most difficult auto components to model using three dimensional Computational Fluid Dynamics(CFD) method. This paper will present a novel approach to address the challenges inherent in crescent oil pump modeling. The new approach is incorporated into the commercial pump design tool PumpLinx from Simerics, Inc.. The new method is applied to simulate a production crescent oil pump with inlet/outlet ports, inner/outer gears, irregular shaped crescent island and tip leakages. The pump performance curve, cavitation effects and pressure ripples are studied using this tool and will be presented in this paper. The results from the simulations are compared to the experiment data with excellent agreement. The present study shows that the proposed computational model is very accurate and robust and can be used as a reliable crescent pump design tool.

This paper describes a transient, thermodynamic, crank angle-based engine model in Modelica that can be used to simulate a range of advanced engine technologies. A single cylinder model is initially presented and described, along with its validation against steady-state dynamometer test data. Issues related to this single cylinder validation are discussed, including the appropriate conservation of hot residual gases under very early intake valve opening (IVO) conditions. From there, the extension from a single cylinder to a multi-cylinder V8 engine model is explained and simulation results are presented for a transient cylinder-deactivation scenario on a V8 engine.

In this paper, an application process is studied at which the insertion loss (IL) test data of sound insulating parts or noise control treatments are utilized for the sound transmission loss (STL) simulation of the trimmed dash structure. The considered sound barrier assemblies were composed of a felt layer, a mass layer, and a decoupler layer. Flat samples of sound barrier assemblies with several different thicknesses were prepared, and ILs of them were measured by using a sound transmission loss facility. Flat samples were assumed to have mass-spring-mass resonance frequencies. The mass was set as the area mass of the sound barrier layer of the felt layer and the mass layer. The spring constant of the decoupler layer was assumed as the multiplication of that of an air spring and a spring correction factor.

The high-frequency whining noise produced by motors in modern electric vehicles causes a significant issue, leading to annoyance among passengers. This noise becomes even more noticeable due to the quiet nature of electric vehicles, which lack other noises to mask the high-frequency whining noise. To improve the noise caused by motors, it is essential to optimize various motor design parameters. However, this task requires expert knowledge and a considerable time investment. In this study, we explored the application of artificial intelligence to optimize the NVH performance of motors during the design phase. Firstly, we selected and modeled three benchmark motor types using Motor-CAD. Machine learning models were trained using Design of Experiment methods to simulate batch runs of Motor-CAD inputs and outputs.

It is an important factor in electric vehicles to show customers how much they can drive with the energy of the remaining battery. If the remaining mileage is not accurate, electric vehicle drivers will have no choice but have to feel anxious about the mileage. Additionally, the potential customers have range anxiety when they consider Electric Vehicles. If the remaining mileage to drive is wrong, drivers may not be able to get to the charging station and may not be able to drive because the battery runs out. It is important to show the remaining available driving range exactly for drivers. The previous study proposed an advanced model by predicting the remaining mileage based on actual driving data and based on reflecting the pattern of customers who drive regularly. The Bayesian linear regression model was right model in previous study.

Automated driving is currently one of the most active areas of research worldwide. While the general progress in developing specific algorithms for perception, planning and control tasks is very advanced, testing and validation of the resulting functions is still challenging due to the large number of possible scenarios and generation of ground-truth. Currently, real world testing and simulations are used in combination to overcome some of these challenges. While real world testing does not suffer from imperfect sensor models and environments, it is expensive, slow and not accurately repeatable and therefore unable to capture all possible scenarios. However, simulation models are not sophisticated enough to fully replace real world testing. In this paper, we propose a workflow that is capable of augmenting real sensor-level data with simulated sensor data.

Aerodynamic simulation results are most of the time compared to wind tunnel results. It is too often simplistically believed that it suffice to take the CAD geometry of a car, prepare and run a CFD simulation to obtain results that should be comparable. With the industry requesting accuracies of a few drag counts when comparing CFD to wind tunnel results, a careful analysis of the element susceptible of creating a difference in the results is in order. In this project a detailed 1:4 scale model of the Hyundai Genesis was tested in the model wind tunnel of the FKFS. Five different underbody panel configurations of the car were tested going from a fully paneled car to a car without panels. The impact of the moving versus static ground was also tested, providing over all ten different experimental results for this car model.

This paper describes the issues encountered in using conventionally acquired tire test data for dynamic total vehicle handling simulations and the need for improved methodology. It describes the new test procedure that was used to acquire all three forces and three moments in a transient mode for a matrix of loads, slip and camber angles. A study of the test data supports the premises that the overturning moment, Mx, should not be neglected in dynamic simulations, and that the effects of camber should not be treated as only an independent, linearly additive, camber thrust. Instead of the conventional application of a bi-cubic regression fit to a six region data division, a new algorithm is applied. The data is divided differently into five regions in the α - Fz plane, and a variable format regression equation is applied as appropriate. The resulting regression coefficients matrix is readily usable in dynamic simulations, and is shown to have a superior curve fit to the test data.

This paper investigates the use of an adaptive proportional-integral-derivative (APID) controller to reduce a combustion engine crankshaft speed pulsation. Both computer simulations and engine test rig experiments are used to validate the proposed control scheme. The starter/alternator (S/A) is used as the actuator for engine speed control. The S/A is an induction machine. It produces a supplemental torque source to cancel out the fast engine torque variation. This machine is placed on the engine crankshaft. The impact of the slowly varying changes in engine operating conditions is accounted for by adjusting the APID controller parameters on-line. The APID control scheme tunes the PID controller parameters by using the theory of adaptive interaction. The tuning algorithm determines a set of PID parameters by minimizing an error function. The error function is a weighted combination of the plant states and the required control effort.

This research aims to develop an inverse control method capable of adaptively simulating dynamic models of car subsystems in the rig-test condition. Accurate simulation of the actual vibration conditions is one of the most crucial factors in realizing reliable rig-test platforms. However, most typical rig tests are conducted under simple random or harmonic sweep conditions. Moreover, the conventional test methods are hard to directly adapt to the actual vibration conditions when switching the dynamic characteristics of the subsystem in the rig test. In the present work, we developed an inverse controller to adaptively control the vibration exciter referring to the target vibration signal. An adaptive LMS filter, employed for the control algorithm, updated the filter weights in real time by referring to the target and the measured acceleration signals.

A practical modeling methodology for adhesively bonded structures using discrete springs has been developed for crash simulation. As a first step, a series of coupon tests with adhesively bonded substrates have been conducted under tension, peel and shearing. Both deformable and rigid substrates have been used in these tests. The resulting data has been used to determine the properties of the adhesive springs. A set of numerical simulations of the coupon tests have been conducted to verify that the adhesive spring properties derived earlier do indeed represent the mechanical properties of the physical adhesives in the coupon tests.

Today numerical models are a major part of the diesel engine development. They are applied during several stages of the development process to perform extensive parameter studies and to investigate flow and combustion phenomena in detail. The models are divided by complexity and computational costs since one has to decide what the best choice for the task is. 0D models are suitable for problems with large parameter spaces and multiple operating points, e.g. engine map simulation and parameter sweeps. Therefore, it is necessary to incorporate physical models to improve the predictive capability of these models. This work focuses on turbulence and mixing modeling within a 0D direct injection stochastic reactor model. The model is based on a probability density function approach and incorporates submodels for direct fuel injection, vaporization, heat transfer, turbulent mixing and detailed chemistry.

Cooling drag is a metric that measures the influence of air flow travelling through the open grille of a ground vehicle on overall vehicle drag, both internally (engine air flow) and externally (interference air flow). With the interference effects considered, a vehicles cooling drag can be influenced by various air flow fields around the vehicle, not just the air flow directly entering or leaving the engine bay. For this reason, computational fluid dynamics (CFD) simulations are particularly difficult. With insights gained from a previously conducted set of experimental studies, a CFD validation effort was undergone to understand which air flow field characteristics contribute to CFD/test discrepancies. A Lattice-Boltzmann Large Eddy Simulation (LES) method was used to validate several test points. Comparison using integral force values, surface pressures, and cooling pack air mass flows was presented.

One important design goal for spark-ignited engines is to minimize cyclic variability. A small amount of cyclic variability (slow burns) can produce undesirable engine vibrations. A larger amount of cyclic variability (incomplete burns) leads to increased hydrocarbon consumption/emissions. Recent studies have reported deterministic patterns in cyclic variability under extremely lean (misfiring) operating conditions. The present work is directed toward more realistic non-misfiring conditions. Production engine test results suggest that deterministic patterns in cyclic variability are the consequence of incomplete combustion, hence control algorithms based on the occurrence of these patterns are not expected to be of significant practical value.

In the analysis of automotive vibration signals, it is often desirable to precisely track individual tonal components of a signal over time. This kind of analysis is useful in defining root cause of cyclic vibrations, allowing the engineer to localize a vibration problem to a specific subsystem. The most commonplace analysis methods are block-oriented Fourier Transform (FT) approaches. However, these methods invariably lead to time localization uncertainty and imprecise frequency resolution. Other analysis methods are available that can estimate the precise frequency of content of a signal at every time step, given certain signal assumptions. This paper studies several of these analysis methods. Narrow Band filters, Kalman filtering, Auto-Regressive signal modeling and the Phase Locked Loop are examined as methods of signal decomposition and order extraction.

Ford designed and built a midsize family sedan for the PNGV (Partnership for a New Generation of Vehicle). The side impact performance of the aluminum vehicle and the current CAE capability was studied. The vehicle was tested according to the specifications of FMVSS 214. The results show the vehicle meet the federal safety requirements. The impact performances of the front and rear dummies were comparable to those of the steel counterpart. CAE analysis was conducted to develop the body component design and to predict the structural and dummy responses. The results show that without modeling of the joint (rivet and weld) separation, the accuracy of the CAE crash analysis for this aluminum vehicle was inadequate. When empirical separation criteria were incorporated to model the joint, analysis results correlated with the test. Further development of robust modeling methods for joint separation is needed to improve the prediction of aluminum structure crash responses.

A successful piston design requires eliminate the following failure modes: structure failure, skirt scuffing and piston unusual noise. It also needs to deliver least friction to improve engine fuel economy and performance. Traditional approach of using hardware tests to validate piston design is technically difficult, costly and time consuming. This paper presents an up-front CAE tool and an analytical process that can systematically address these issues in a timely and cost-effectively way. This paper first describes this newly developed CAE process, the 3D virtual modeling and simulation tools used in Ford Motor Company, as well as the piston design factors and boundary conditions. Furthermore, following the definition of the piston design assessment criteria, several piston design studies and applications are discussed, which were used to eliminate skirt scuffing, reduce piston structure dynamic stresses, minimize skirt friction and piston slapping noise.