Search Results

This paper reports on the injuries to the head, neck and thorax of fifteen child surrogates, subjected to varying levels of sudden acceleration. Measured response data in the child surrogate tests and in matched tests with a three-year-old child test dummy are compared to the observed child surrogates injury levels to develop preliminary tolerance data for the child surrogate. The data are compared with already published data in the literature.

A mathematical model is developed which describes the operation of an A/F control system containing an engine, an exhaust gas oxygen sensor, and a feedback controller. The dependence of model parameters on engine operating conditions is discussed, and the model is used to compare integral and proportional/integral control algorithms. Actual data obtained on an engine-dynamometer are presented to test the validity of the model.

Vehicle restraint system design is a difficult optimization problem to solve because (1) the nature of the problem is highly nonlinear, non-convex, noisy, and discontinuous; (2) there are large numbers of discrete and continuous design variables; (3) a design has to meet safety performance requirements for multiple crash modes simultaneously, hence there are a large number of design constraints. Based on the above knowledge of the problem, it is understandable why design of experiment (DOE) does not produce a high-percentage of feasible solutions, and it is difficult for response surface methods (RSM) to capture the true landscape of the problem. Furthermore, in order to keep the restraint system more robust, the complexity of restraint system content needs to be minimized in addition to minimizing the relative risk score to achieve New Car Assessment Program (NCAP) 5-star rating.

Frequency domain testing has had limited use in the past for durability evaluations of automotive components. Recent advances and new perspectives now make it a viable option. Using frequency domain testing for components, test times can be greatly reduced, resulting in considerable savings of time, money, and resources. Quality can be built into the component, thus making real-time subsystem and full vehicle testing and development more meaningful. Time domain testing historically started with block cycle histogram tests. Improved capabilities of computers, controllers, math procedures, and algorithms have led to real time simulation in the laboratory. Real time simulation is a time domain technique for duplicating real world environments using computer controlled multi-axial load inputs. It contains all phase information as in the recorded proving ground data. However, normal equipment limitations prevent the operation at higher frequencies.

This paper reports the analytical procedure developed for a simulation of knee impact during a barrier crash using a hybrid III dummy lower torso. A finite element model of the instrument panel was generated. The dummy was seated in mid-seat position and was imparted an initial velocity so that the knee velocity at impact corresponded to the secondary impact velocity during a barrier crash. The procedure provided a reasonably accurate simulation of the dummy kinematics. This simulation can be used for understanding the knee bolster energy management system. The methodology developed has been used to simulate impact on knee for an occupant belted or unbelted in a frontal crash. The influence of the vehicle interior on both the dummy kinematics and the impact locations was incorporated into the model. No assumptions have been made for the knee impact locations, eliminating the need to assume knee velocity vectors.

A cycle counting algorithm that will reduce a complex history into a series of discrete cycles is presented. The cycles determined by this technique are defined as closed stress-strain hysteresis loops of the type obtained from constant amplitude tests. Using the computer cycle counting algorithm, life predictions were made and compared with experimental results. These predictions were found to be typically within a ±3 factor of error. Also, the computer counting method was found to yield more accurate life predictions when compared to the histogram and range counting methods.

In this paper, a soft computing approach to a model-free vehicle stability control (VSC) algorithm is presented. The objective is to create a fuzzy inference system (FIS) that is robust enough to operate in a multitude of vehicle conditions (load, tire wear, alignment), and road conditions while at the same time providing optimal vehicle stability by detecting and minimizing loss of traction. In this approach, an adaptive neuro-fuzzy inference system (ANFIS) is generated using previously collected data to train and optimize the performance of the fuzzy logic VSC algorithm. This paper outlines the FIS detection algorithm and its benefits over a model-based approach. The performance of the FIS-based VSC is evaluated via a co-simulation of MATLAB/Simulink and CarSim model of the vehicle under various road and load conditions. The results showed that the proposed algorithm is capable of accurately indicating unstable vehicle behavior for two different types of vehicles (SUV and Sedan).

This paper focuses on the role and significance of linear momentum and kinetic energy in controlling air bags aboard vehicles. Among the results of the study are analytic and geometric models that characterize crash behavior and control algorithms that quantify crash severity and identify crash configurations. These results constitute an effective basis for crash-data design and air-bag control.

Various algorithms such as a direct computation approach, maximization requirement criteria method established by Chou and Nyquist, and a partitioning technique, for computing HIC are reviewed in this paper. An evaluation has been conducted considering both the accuracy and efficiency of these algorithms using theoretical pulses and experimental resultant head accelerations of a dummy obtained from the literature, Hyge sled and frontal barrier impact tests. Using results obtained from direct computations as “exact” values for comparison, all the algorithms evaluated provide HIC estimates in close agreement with the “exact” values. The CPU times, which are used as a measure for the assessment of computational efficiency, vary from algorithm to algorithm. Methods using a partitioning logic developed by Mentzer and a faster algorithm developed by Holstein and Alem are found to be very efficient, and are recommended for use in the computation of HIC.

This paper describes a comprehensive methodology for the simulation of vehicle body panel buckling in an electrophoretic coat (electro-coat or e-coat) and/or paint oven environment. The simulation couples computational heat transfer analysis and structural analysis. Heat transfer analysis is used to predict temperature distribution throughout a vehicle body in curing ovens. The vehicle body temperature profile from the heat transfer analysis is applied as an input for a structural analysis to predict buckling. This study is focused on the radiant section of the curing ovens. The radiant section of the oven has the largest temperature gradients within the body structure. This methodology couples a fully transient thermal analysis to simulate the structure through the electro-coat and paint curing environments with a structural, buckling analysis.

This paper proposes a novel algorithm. This algorithm is called Self-Organizing Fuzzy Neural Network (SOFNN). SOFNN revolutionizes how researchers apply control theories, image/signal processing on control systems and other applications. In general, SOFNN is an identification technique that automatically initiates, builds and fine-tunes the required network parameters. SOFNN evaluates required structures without predefined parameters or expressions regarding systems. SOFNN sets out to learn and configure a system's characteristics. Self-constructing and self-tuning features enable SOFNN to handle complex, non-linear, and time-varying systems with higher accuracy, making systems identification easier. SOFNN constructs and fine-tunes the system parameter through two phases. The two phases are the construction and the parameter-tuning phase. The two phases run concurrently allowing SOFNN to identify systems on-line.

Model characteristics of a finite element deformable featureless headform with one to four layers of solid elements for the headform skin are studied using both the LS-DYNA3D and FCRASH codes. The models use a viscoelastic material law whose constitutive parameters are established through comparisons of drop test simulations at various impact velocities with the test data. Results indicate that the one-layer model has a significant distinct characteristic from the other (2-to-4-layer) models, thus requiring different parametric values. Similar observation is also noticed in simulating drop tests with one and two layers of solid elements for the headform skin using PAM-CRASH. When using the same parametric values for the viscoelastic material, both the LS-DYNA3D and FCRASH simulations yield the same results under identical impact conditions and, thereby, exhibit a “functional equivalency” between these two codes.

Various methods are available to predict future cylinder air charge for improved air/fuel control. However, there can never be perfect prediction. This paper presents an algorithm to correct for imperfect cylinder charge prediction. This is done by expanding the air/fuel control boundary to include the catalyst, and correcting prediction errors as soon as possible using small corrective changes to later cylinder fuel inputs. The method was experimentally tested and showed improved air/fuel control as indicated by reduced variability of catalyst downstream air/fuel ratio. Additional vehicle testing showed potential to further reduce emissions.

An analytical method for determining engine torque harmonics is presented. The approach employs an engine cycle simulation model to calculate instantaneous cylinder pressure for each operating condition based on engine characteristics that can be determined experimentally and/or analytically. Cylinder pressure is converted to instantaneous torque from which harmonics are determined using an FFT algorithm. A description of the cycle simulation model, including required data, is presented. The method is validated by presenting correlation results at a number of operating conditions (i.e. engine speeds and loads) comparing analytical versus test driveline torque harmonics. Finally, limitations in the method as well as possible extensions to the method are discussed.

This paper examines the theoretical worst case of normal headform impact on an infinitely rigid surface with the help of a dynamic spring-mass model. It is pointed out that the current approach is not an actual representation of any vehicle upper interior but is useful in gaining insight into the headform impact phenomenon and determining how to further enhance design. After considering force-deflection characteristics of a variety of commonly used headform impact protection countermeasures, a mathematical model is set up with spring properties that approximate those of physical countermeasures. Closed-form solutions are derived for various dynamic elasto-plastic phases including elastic unloading and contact. A parametric study is then carried out with HIC(d) as the dependent variable, and spring stiffness, yield force and spring length (representing countermeasure crush space) as the design variables.

With the increased market share of electric vehicles, the demand for energy-efficient routing algorithms specifically optimized for electric vehicles has increased. Traditional routing algorithms are focused on optimizing the shortest distance or the shortest time in finding a path from point A to point B. These traditional methods have been working well for fossil fueled vehicles. Electric vehicles, on the other hand, require different route optimization techniques. Negative edge costs, battery power limits, battery capacity limits, and vehicle parameters that are only available at query time, make the task of electric vehicle routing a challenging problem. In this paper, we present an ant colony based, energy-efficient routing algorithm that is optimized and designed for electric vehicles. Simulation results show improvements in the energy consumption of electric vehicles when applied to a start-to-destination routing problem.

The viscous criterion (V*C) has been proposed by biomechanics researchers as a generic biomechanical index for potential soft tissue injury. It is defined by the product of the velocity of deformation and the instantaneous compression of torso and abdomen. This criterion requires calculation and differentiation of measured torso/abdomen compression data. Various computational algorithms for calculating viscous criterion are reviewed and evaluated in this paper. These include methods developed by Wayne State University (WSU), NHTSA (DOT) and Ford. An evaluation has been conducted considering the accuracy of these algorithms with both theoretical and experimental data from dummy rib compressions obtained during a crash test. Based on these results, it is found that: V*C results depend on the scheme used in the computation process, the sampling rate and filtering of original raw data. The NHTSA method yields the lowest V*C value.

The SAE Large Male and Small Female Dummy Task Group has recommended a change to the Hybrid III dummy hip joint. This change was made because of a non-biofidelic interference in the current design that can influence chest accelerations. The modifications include a new femur casting shaft design and the addition of an elastomeric stop to the top of the casting. Static testing and Hyge sled tests were done to evaluate the modifications. Based on the results, the new design satisfied the requirements set by the SAE task group and reduced the influence of hip joint characteristics on chest accelerations.

Fatigue analysis in the time domain using the rainflow cycle counting algorithm is considered the most accurate method for estimating damage. Dirlik's method has been found to be very accurate for damage estimation in the frequency domain. Previous studies have demonstrated the usefulness of Dirlik's method for ocean engineering and wind turbines but few have shown how well Dirlik performs in automotive applications. This study compares Dirlik's method with the rainflow cycle counting and with other frequency domain methods. The study analyzes measured data for an automotive component subjected to five test track load conditions. In addition, fourteen of Dirlik's original spectra and seven additional spectra which combine sine and random spectra are studied. It was found that Dirlik's method predicts more damage than the rainflow cycle counting method when applied to the original data used in creating the method.

This paper discusses steps for identifying, evaluating and recommending a quantifiable design metric or metrics for Side Airbag (SAB) development. Three functionally related and desirable attributes of a SAB are assumed at the onset, namely, effective SAB coverage, load distribution and efficient energy management at a controlled force level. The third attribute however contradicts the “banana shaped” force-displacement response that characterizes the ineffective energy management reality of most production SAB. In this study, an estimated ATD to SAB interaction energy is used to size and recommend desired force-deformation characteristic of a robust energy management SAB. The study was conducted in the following three phases and corresponding objectives: Phase 1 is a SAB assessment metric identification and estimation, using a uniform block attached to a horizontal impact machine.