Search Results

As CAFE requirements increase, automotive OEMs are pursuing innovative methods to lightweight their Body In Whites (BIWs). Within FCA US, this lightweighting research and development activity often occurs through Decoupled Innovation projects. A Decoupled Innovation team comprised of engineers from the BIW Structures Group, in collaboration with Tier 1 supplier Magna Exteriors, sought to re-design a loadbearing component on the BIW that would offer significant weight savings when the current steel component was replaced with a carbon fiber composite. This paper describes the design, development, physical validation and partnership that resulted in a composite Rear Package Shelf Assembly solution for a high-volume production vehicle. As the CAFE requirements loom closer and closer, these innovation-driven engineering activities are imperative to the successful lightweighting of FCA US vehicles.

This research proposes a control system for Spark Ignition (SI) engines with external Exhaust Gas Recirculation (EGR) based on model predictive control and a disturbance observer. The proposed Economic Nonlinear Model Predictive Controller (E-NMPC) tries to minimize fuel consumption for a number of engine cycles into the future given an Indicated Mean Effective Pressure (IMEP) tracking reference and abnormal combustion constraints like knock and combustion variability. A nonlinear optimization problem is formulated and solved in real time using Sequential Quadratic Programming (SQP) to obtain the desired control actuator set-points. An Extended Kalman Filter (EKF) based observer is applied to estimate engine states, combining both air path and cylinder dynamics. The EKF engine state(s) observer is augmented with disturbance estimation to account for modeling errors and/or sensor/actuator offset.

Within a hybrid electric vehicle, given a power request initiated by pedal actuation, a portion of overall power may be generated by fuel within an internal combustion engine, and a portion of power may be taken from or stored within a battery via an e-machine. Generally speaking, power taken from a vehicle battery must eventually be recharged at a later time. Recharge energy typically comes ultimately from engine generated power (and hence from fuel), or from recovered braking energy. A hybrid electric vehicle control system attempts to identify when to use each type of power, i.e., battery or engine power, in order to minimize overall fuel consumption. In order to most efficiently utilize battery and fuel generated power, many HEV control strategies utilize a concept wherein battery power is converted to a scaled fueling rate.

As the move to decrease physical prototyping increases the need to virtually prototype vehicles become more critical. Assessing NVH vehicle targets and making critical component level decisions is becoming a larger part of the NVH engineer’s job. To make decisions earlier in the process when prototypes are not available companies need to leverage more both their historical and simulation results. Today this is possible by utilizing a hybrid modelling approach in an NVH Simulator using measured on road, CAE, and test bench data. By starting with measured on road data from a previous generation or comparable vehicle, engineers can build virtual prototypes by using a hybrid modeling approach incorporating CAE and/or test bench data to create the desired NVH characteristics. This enables the creation of a virtual drivable model to assess subjectively the vehicles acoustic targets virtually before a prototype vehicle is available.

Cost benefit and teraflop restrictions imposed by racing sanctioning bodies make steady-state RANS CFD simulation a widely accepted first approximation tool for aerodynamics evaluations in motorsports, in spite of its limitations. Research involving generic and simplified vehicle bodies has shown that the veracity of aerodynamic CFD predictions strongly depends on the turbulence model being used. Also, the ability of a turbulence model to accurately predict aerodynamic characteristics can be vehicle shape dependent as well. Modifications to the turbulence model coefficients in some of the models have the potential to improve the predictive capability for a particular vehicle shape. This paper presents a systematic study of turbulence modeling effects on the prediction of aerodynamic characteristics of a NASCAR Gen-6 Cup racecar. Steady-state RANS simulations are completed using a commercial CFD package, STAR-CCM+, from CD-Adapco.

Accurate battery state of charge (SOC) estimation is essential for safe and reliable performance of electric vehicles (EVs). Lithium-ion batteries, commonly used for EV applications, have strong time-varying and non-linear behaviour, making SOC estimation challenging. In this paper, a processor in the loop (PIL) platform is used to assess the execution time and memory use of different SOC estimation algorithms. Four different SOC estimation algorithms are presented and benchmarked, including an extended Kalman filter (EKF), EKF with recursive least squares filter (EKF-RLS) feedforward neural network (FNN), and a recurrent neural network with long short-term memory (LSTM). The algorithms are deployed to two different NXP S32Kx microprocessors and executed in real-time to assess the algorithms' computational load. The algorithms are benchmarked in terms of accuracy, execution time, flash memory, and random access memory (RAM) use.

Vibration shaker testing is a great tool of validating the vibration fatigue performance of automotive components & systems. However, the representative vibration schedule requires a pre-knowledge of the acceleration history for the test object, which usually is not available until the later development phase of a vehicle program when physical properties are available. Sometimes, a generic vibration schedule developed from the worst-case loading profiles are used with risk of lacking correlation with later full vehicle durability test such as Road Test Simulator (RTS) or Proving Ground (PG) road test due to the higher loading amplitude. This paper proposes a virtual accelerometer approach to collect acceleration responses of a component from a virtual vehicle model. First, a multiple body dynamic model will be produced for virtual load calculation over a series of digitalized virtual proving ground road profiles.

An accurate bushing model is vital for vehicle dynamic simulation regarding fatigue life prediction. This paper introduces the Advanced Bushing Model (ABM) that was developed in MATLAB® environment, which gives high precision and fast simulation. The ABM is a time-domain model targeting for vehicle durability simulation. It dynamically captures bushing nonlinearities that occur on stiffness, damping and hysteresis, through a time-history-based fitting technique, compensated with frequency dependency functionality. Among the simulated and test-collected bushing loads, good correlations have been achieved for elastomer bushings and hydraulic engine mounts and validated with a random excitation signal. This ABM model has been integrated into a virtual shaker table (from a parallel project) as the engine mount model to simulate the mount load, and has shown acceptable prediction on fatigue damage.

Several powertrain noise phenomena have been studied over the years. Sound quality metrics, like loudness, sharpness, modulation, and tonality, among others, have been developed to characterize powertrain noises. While these readily available metrics work well on steady state and some transient noises, they do not correlate directly with subjective impressions. Moreover, it is difficult to assign a meaningful single rating for time varying noises that may also be associated with simultaneous variations in frequency content. This paper summarizes the process of creating a vehicle level objective metric and its application to blind noise samples to verify correlation with subjective impressions, particularly in association with clatter noise at moderate engine speeds (2000-3500 rpm) with light to moderate throttle tip-ins.

SUV Aerodynamics has received increased attention as the stake this segments holds in the automotive market keeps growing year after year, as well as its direct impact on fuel economy. Understanding the key physics in order to accomplish both fuel efficient and aesthetic products is paramount, which indeed gave origin to a major initiative to foster collaborative aerodynamic research across academia and industry, the so-called DrivAer model. In addition to this sedan-based model, a new dedicated SUV generic model, called AeroSUV [1], has been introduced in 2019, also intended to provide a common framework for aerodynamic research for both experimental work and numerical simulation validation. The present paper provides an area of common ground for SUV bodywork design focused on aerodynamic drag reduction by investigating both Estate and Fast back configurations of the generic AeroSUV model.

Artificial intelligence (AI) is dramatically changing multiple industries. AI’s potential to transform Computer-Aided Engineering (CAE) cannot be overlooked. Conventionally, Finite Element Analysis (FEA) is the simulation of any given physical phenomenon to obtain an approximate solution to a group of problems governed by Partial Differential Equations (PDE). Implementation of AI methods in this area combines human intelligence with numerical solutions to make them more efficient. This paper attempts to develop a Deep Neural Network (DNN) model to solve an elasto-plastic impact problem of a symmetric short crush tube made of three materials impacted by a moving wall. A structured learning database was established to train and validate the model using finite element simulations. Tube size, gauge and elasto-plastic material properties were used as input attributes or features. The maximum axial displacement of the tube is the target label to predict.

Lithium-ion battery (LIB) terminal voltage models are investigated using two modelling approaches. The first model is a third-order Thevenin equivalent circuit model (ECM), which consists of an open-circuit voltage in series with a nonlinear resistance and three parallel RC pairs. The parameters of the ECM are obtained by fitting the model to hybrid pulse power characterization (HPPC) test data. The parametrization of the ECM is performed through quadratic-based programming. The second is a novel modelling approach based on long short-term memory (LSTM) recurrent neural networks to estimate the battery terminal voltage. The LSTM is trained on multiple vehicle drive cycles at six different temperatures, including −20°C, without the necessity of battery characterization tests. The performance of both models is evaluated with four automotive drive cycles at each temperature. The results show that both models achieve acceptable performance at all temperatures.

Paper withdrawn by author.

Cooling airflow is an essential factor when it comes to vehicle performance and operating safety. In recent years, significant efforts have been made to maximize the flow efficiency through the heat exchangers in the under-hood compartment. Grille shutters, new fan shapes, better sealings are only some examples of innovations in this field of work. Underhood cooling airflow simulations are an integral part of the vehicle development process. Especially in the early development phase, where no test data is available to verify the cooling performance of the vehicle, computational fluid dynamics simulations (CFD) can be a valuable tool to identify the lack of fan performance and to develop the appropriate strategy to achieve airflow goals through the heat exchangers. For vehicles with heat exchangers in the underhood section the airflow through those components is of particular interest.

The objective of this study is to analytically model the fatality risk in frontal vehicle-to-vehicle crashes of the current vehicle fleet, and its sensitivity to vehicle mass change. A model is built upon an empirical risk ratio-mass ratio relationship from field data and a theoretical mass ratio-velocity change ratio relationship dictated by conservation of momentum. The fatality risk of each vehicle is averaged over the closing velocity distribution to arrive at the mean fatality risks. The risks of the two vehicles are summed and averaged over all possible crash partners to find the societal mean fatality risk associated with a subject vehicle of a given mass from a fleet specified by a mass distribution function. Based on risk exponent and mass distribution from a recent fleet, the subject vehicle mean fatality risk is shown to increase, while at the same time that for the partner vehicles decreases, as the mass of the subject vehicle decreases.

Battery state-of-charge (SOC) is critical information for the vehicle energy management system and must be accurately estimated to ensure reliable and affordable electrified vehicles (xEV). However, due to the nonlinear temperature, health, and SOC dependent behaviour of Li-ion batteries, SOC estimation is still a significant automotive engineering challenge. Traditional approaches to this problem, such as electrochemical models, usually require precise parameters and knowledge from the battery composition as well as its physical response. In contrast, neural networks are a data-driven approach that requires minimal knowledge of the battery or its nonlinear behaviour. The objective of this work is to present the design process of an SOC estimator using a deep feedforward neural network (FNN) approach. The method includes a description of data acquisition, data preparation, development of an FNN, FNN tuning, and robust validation of the FNN to sensor noise.