Crash Simulation Research Articles

Numerical and Experimental Evaluation of a Battery Cell under Impact Load

Impact damage is one of the most critical scenarios for the lithium-ion battery pack of an electrical vehicle, as it involves mechanical abusive loads with serious consequences on electrical and thermal stability. The development of a numerical model for an explicit dynamic simulation of a Li-ion battery pack under impact implies a significant computational effort if detailed models of a single battery cell are employed. The present paper presents a homogenized finite element model of a battery cell, validated by experimental tests of individual materials and an impact test of an entire cell. The macro model is composed of shell elements representing outside casing and elements with a homogenized and isotropic material for the jelly roll. The displacements and deformed shape of the numerical model of the battery cell were compared with those measured on real test specimens; full-field optical scanning was employed to reconstruct the 3D shape of the deformed battery. The overall deformation of the simulation and experimental results are comparable with a deviation of the maximum intrusion of 14.8% for impact direction and 19.5% for the perpendicular direction considering the cumulative effects of simplifying hypotheses of the numerical model and experimental side effects. The results are a starting point for future analyses of a battery pack and its protection systems under impact. The model presented in this paper, considering the low computing power needed for calculation and acceptable mesh size for crash, should be able to be used in bigger resources consuming crash simulation models. In this way, the cells’ deformation and behavior can be tracked more easily for safety management and diagnosis of the crashworthiness of the packs or car batteries.

Robust design assessment of automotive crash box structures through model order reduction

The research work is focused on a comprehensive study of crashworthiness parameters associated with widely used crash box components of passenger cars. The finite-element models representing the automotive crash box structure were procured from full-car digital models issued by NHTSA. From the full car model, the crash box geometry is extracted for the crash analysis and further design optimizations. The crashworthiness parameters such as energy absorption, crush force efficiency, specific energy absorption is studied in detail with analytical and finite-element solutions for each variant and are further enhanced using multi-objective design optimization through RSM-based DOE methods. For guiding engineers in real- time design exploration and other parametric assessments, a reduced-order model for each of the variants was developed by using the sPGD (sparse Proper Generalised Decomposition) method. The developed reduced order models of the crash boxes were able to accurately capture the non-linearities associated with them and delivered accuracy of 90%. On undergoing robust design assessment with crash box designs, taking into account of all variants, an average of 34.33% in energy absorption capacity was achieved. The heavier crash box variants which had good SEA capacity are performed for successful mass reduction of 18.91% without affecting their crash performance. For the explicit crash simulations, the virtual performance solution finite-element solver from the ESI group is utilized.

Method for automated detection of outliers in crash simulations

Stricter legal requirements in crash safety lead to more complex development processes in computer-aided engineering and result in an increasing number of simulations. Both, the construction of the simulation models as well as their evaluation are costly and time-consuming. Therefore, an automated workflow is required that significantly facilitates the analysis of the results by the engineer and increases the quality of the evaluation. In this study an automated evaluation process is proposed that detects anomalous crash behaviour in a bundle of crash simulations. The individual states from the simulation are analysed separately from each other and an outlier score is calculated using a kth-nearest-neighbour approach. Subsequently, these results are averaged into one score for each simulation. With the help of different statistical methods, a threshold value is calculated, from which a simulation can be identified as an outlier. The evaluation is carried out on 5 datasets. On average, the precision and recall of the presented method are 1.0 and 0.91, respectively.

Estimating Automobile Crash Characteristics from Images using Deep Learning


 
 
 Crash characteristics such as crash velocity (Delta-V) and location of collision (LOC) are important determinants of the severity of the injury sustained by an occupant of an accident vehicle. Based on the predicted severity levels of injury, insurance companies can estimate the claim’s cost and better plan their financial reserves. We present a promising approach for accurately predicting Delta-V and LOC using deep learning methods, without the need for a forensic crash reconstruction. We constrain the study to small passenger vehicles and to front and rear collisions with crash velocities under 96 kph. We first develop and refine our image processing and deep CNN architectures using images created by using vehicle crash simulation software. Using a k-fold cross-validation approach, our methods are able to predict the crash velocity of simulated collisions (108 images) with a MAE of 3.41 kph (MAPE of 8.2%). Similarly, a multiple task learning CNN is able to predict Delta-V of real-world collisions (310 images) with a MAE of 4.19 kph (MAPE of 16.2%) and classify the LOC with 92% accuracy.
 
 

Deep learning methodology for predicting time history of head angular kinematics from simulated crash videos

Head kinematics information is important as it is used to measure brain injury risk. Currently, head kinematics are measured using wearable devices or instrumentation mounted on the head. This paper evaluates the deep learning approach in predicting time history of head angular kinematics directly from videos without any instrumentation. To prove the concept, a deep learning model was developed for predicting time history of head angular velocities using finite element (FE) based crash simulation videos. This FE dataset was split into training, validation, and test datasets. A combined convolutional neural network and recurrent neural network based deep learning model was developed using the training and validations sets. The test (unseen) dataset was used to evaluate the predictive capability of the deep learning model. On the test dataset, correlation coefficient obtained between the actual and predicted peak angular velocities was 0.73, 0.85, and 0.92 for X, Y, and Z components respectively.

A CRASH simulation of the contribution of binary stars to the epoch of reionization

ABSTRACT We use a set of three-dimensional (3D) radiative transfer simulations to study the effect that a large fraction of binary stars in galaxies during the epoch of reionization has on the physical properties of the intergalactic medium (i.e. the gas temperature and the ionization state of hydrogen and helium), on the topology of the ionized bubbles and on the 21 cm power spectra. Consistently to previous literature, we find that the inclusion of binary stars can speed up the reionization process of H i and He i, while He ii reionization is still dominated by more energetic sources, especially accreting black holes. The earlier ionization attained with binary stars allows for more time for cooling and recombination so that gas fully ionized by binary stars is typically colder than gas ionized by single stars at any given redshift. With the same volume averaged ionization fraction, the inclusion of binary stars results in fewer small ionized bubbles and more large ones, with visible effects also on the large scales of the 21 cm power spectrum.

True Fracture Strain Measurement and Derivation for Advanced High-Strength Steel Sheets

<div class="section abstract"><div class="htmlview paragraph">The importance of true fracture strain was initially highlighted in the context of local versus global formability considerations used in material selection among advanced high strength steels (AHSSs) of similar tensile strength. Inspired by the relative studies, a precedent work compared the fracture strain results via either digital image correlation (DIC) based method or optical fracture surface measurement on different AHSS samples. It concluded that the DIC-based testing results generally underestimated the fracture strain. As a continued study, the present work further analyzed the DIC-based testing procedure and attributed such an underestimation mainly to the volume constancy assumption. Furthermore, this work pointed out that also because of the same assumption, the optical fracture surface measurement to some extent overestimated the fracture strain. Nevertheless, it was also observed that different AHSS grades were affected discrepantly by the two methods. Therefore, scanning electron microscope (SEM) was applied to inspect the morphology of various micro-voids and dimples on different fracture surfaces to explain the discrepancy. To bypass the volume constancy assumption, this work also proposed two alternative methods, including a DIC-based thinning measurement method and a hybrid method, and discussed their limitations. In addition, the effects of DIC measurement frame rate and anisotropic plasticity based on the Hill-1948 yield function were also studied in this work. Last but not the least, by substituting the different fracture strain results based on all the introduced methods into an early stage of calibration of the Generalized Incremental Stress State dependent damage Model (GISSMO) for forming and crash simulations, the importance of the fracture strain accuracy was further highlighted.</div></div>

Using the Instantaneous Center of Rotation to Examine the Influence of Yaw Rate on Occupant Kinematics in Eccentric Planar Collisions

<div class="section abstract"><div class="htmlview paragraph">The biomechanical injury assessment for an occupant in a planar vehicle-to-vehicle collision often requires a kinematic analysis of impact-related occupant motion. This analysis becomes more complex when the collision force is eccentric to the center of gravity on a struck vehicle because the vehicle kinematics include both translation and potentially significant yaw rotational rates. This study examines the significance of vehicle yaw on occupant kinematics in eccentric (off-center) planar collisions. The paper describes the calculation of the instantaneous center of rotation (ICR) in a yawing vehicle post-impact and explores how mapping this quantity may inform an occupant’s trajectory when using a free particle “occupant” analysis. The study initially analyzed the impact-related occupant motion for all the outboard seat positions in a minivan using several hypothetical examples of eccentric vehicle-to-vehicle crash configurations with varying PDOF, delta-V, and yaw rate. The ICR and free particle occupant trajectories were calculated for six different simulated crash examples to illustrate which seating positions were most influenced by post-impact vehicle yaw. The process was repeated for all the outboard seat positions in a sedan using the vehicle kinematics from a staged two-vehicle crash test. It was found that the ICR can provide the crash analyst or the biomechanist a useful mechanism to visualize the relationship between vehicle and occupant kinematics in an eccentric planar vehicle collision.</div></div>

A systematic review and meta-analysis of the impact of curbs on crash outcomes

Introduction Road traffic crashes involving vertical curbs are commonly reported to occur on highways and expressways in India. We found a gap in terms of systematically assessing the evidence of the impact of curbs on road safety outcomes in the real world. Method We conducted a systematic review and meta-analysis of the impact of curbs on the risk of road traffic injuries. We used keywords in a database of records prepared by an earlier evidence gap map (EGM). The EGM used a comprehensive search strategy including 6 academic database, 17 organizational websites, hand searching, contacting experts and back referencing. Results We found 4 studies that evaluated impact of a curbed median or a curbed shoulder. We found that the presence of a curb on a median increases the risk for all crashes, all single-vehicle crashes, all median-related crashes and median-related injury crashes. The data also indicate that the severity of accidents reduces for curbs on median while it increases for curbs on shoulder, though the latter effect is not statistically significant. All the epidemiological studies were conducted on rural highways and did not report effects for different traffic speeds or vehicle types. However, our review of crash tests and simulation studies indicates that the impact of a curb design may be highly sensitive to speed and vehicle types. Conclusions The safety impacts of a curb depend on the context of the road. In an urban road, a curb should ensure safety of pedestrians from an errant vehicle. On high-speed rural roads, curbs should be avoided and treatments should facilitate safe departure of the vehicle from the roadway.

Characterization of high-strength bolts and the numerical representation method for an efficient crash analysis

This paper presents novel insights of a numerical bolt representation approach taking into account damage and failure, suitable for large-scale models as in full vehicle crash simulations. Custom-built experimental equipment combined with digital image correlation enables observation of bolt failure behavior and its physical characteristics under tension, shear as well as combined loading. A fine-meshed model demonstrates the transferability of the physical properties to numerics including the associated restrictions. The experimental and numerical findings are used to develop an efficient substitute bolt model with high prediction accuracy that takes into account the bolt kinematics due to the thread via volume elements.An isotropic constitutive law with linear elastic and multilinear plastic progression supplemented by a phenomenological damage model depicts the bolt material behavior numerically.The observation of the bolt models under different tension and shear ratios provides new insights to the models accuracy which is assessed by an analytically failure envelope.

Influence of electrolyte presence on the vibration behavior of a Li-ion pouch cell and on the mechanical characteristics of its incorporated [formula omitted]-coated polyolefin separator

Quality improvement of battery crash simulation models is essential for realistic mechanically induced short-circuit prediction in modern electric vehicles, where the battery is part of the load path. In this paper, the electrolyte influence on the mechanical behavior of a commercial pouch cell and its incorporated Al2O3-coated separator is investigated. To mechanically characterize the separator, both compression and tensile tests were performed. Modal analyses were performed to determine the damping behavior and dynamic stiffness of a normally conditioned cell (wet) and a cell without electrolyte (dry). In the tensile tests, a slight anisotropy and an increase in stiffness, triggered by higher test speeds, were observed. The wet specimens soaked in substitute electrolyte mixture showed longer elongation to failure with a stiffer response than the dry specimens under tensile loading. In the compression tests, exactly the opposite mechanical response of wet and dry specimens was measured. From the modal analyses, a stiffening of the dry cell could be determined. Based on the component tests, a finite element model of the separator was developed in LS-DYNA, enabling prediction of the deformation behavior of the separator under mechanical loads — this can be further used in high-fidelity cell simulation models.

A Computational Framework for Human-Centric Vehicular Crashworthiness Design and Decision-Making Under Uncertainty

Abstract Government agencies, globally, strive to minimize the likelihood and frequency of human death and severe injury on road transport systems. From an engineering design standpoint, the minimization of these road accident effects on occupants becomes a critical design goal. This necessitates the quantification and management of injury risks on the human body in response to several vehicular impact variables and their associated uncertainties for different crash scenarios. In this paper, we present a decision-based, robust design (RD) framework to quantify and manage the impact-based injury risks on occupants for different computational model-based car crash scenarios. The key functionality offered is the designer's capability to conduct robust concept exploration focused on managing the selected impact variables and associated uncertainties, such that injury risks are controlled within acceptable levels. The framework's efficacy is tested for near-side impact scenarios with impact velocity and angle of impact as the critical variables of interest. Two injury criteria, namely, head injury criterion (HIC) and lateral neck injury criteria (Lateral Nij) are selected to quantitatively measure the head and neck injury risks in each crash simulation. Using the framework, a RD problem is formulated to determine the combination of impact variables that best satisfies the injury goals defined. The framework and associated design constructs are generic and support the formulation and decision-based robust concept exploration of similar problems involving models under uncertainty. Our focus in this paper is on the framework rather than the results per se.

Deformation and damage behavior of lightweight steels at high rate multiaxial loading

Abstract For the characterization of the deformation and damage behavior of the automotive lightweight steels ZStE340 and DP1000, high speed tension tests for various stress triaxialities with shear tensile specimens, smooth tensile specimens and notched specimens and static and dynamic Nakajima tests were performed. The deformation up to failure was recorded using high speed video cameras with high resolution in the necking zone and evaluated with 2D and in part 3D digital image correlation (DIC). The maximal achieved effective strain at the location of failure, determined by video analysis and fractography, was specified as failure strain. The results for various specimen geometries were plotted versus stress triaxiality as determined by FE simulation in order to receive failure diagrams. Overall, both materials show descending failure curves from shear loading up to tensile loading of notched specimens. Up to equibiaxial tension, failure curves are rising again. For uniaxial and equibiaxial tension, with increasing strain rate larger failure strains occured for DP1000, while no significant influence of strain rate on failure strain was observed for ZStE340. However, under shear loading with increasing strain rate, decreasing failure strains occur for both materials. This is probably due to an adiabatic temperature rise in a small localized zone which promotes shear failure. For shear crash tests a temperature rise of more than 200 K was measured by a high speed infrared camera. These results show the necessity of dynamic tests under different stress triaxialities to provide reliable input data for crash simulations and increase the ability to accurately predict failure. Especially shear failure must be investigated very carefully under crash loading, because here failure occurs at smaller failure strains than it would be expected under quasistatic loading.

Consideration of manufacturing induced adhesive damage in automotive simulations

This work contributes to the understanding of damage of adhesive layers generated during the manufacturing process and its effect on the subsequent load carrying capacity of the joint. A temperature dependent cohesive zone model is used to predict damage during the manufacturing process. Therefore, parameters for a temperature-dependent as well as a strain rate-dependent model were determined on various specimens. It is shown, that it is possible to predict the manufacturing induced damage within the adhesive layer with sufficient accuracy using a cohesive zone model. Furthermore, a continuous process chain for numerical simulations is developed, which allows to determine the adhesive layer damage by means of a process simulation and to consider the changed mechanical properties in a subsequent crash simulation. To simplify the procedure, a software tool is presented that automatically performs the described steps. The developed method is applied to a car body simulation to investigate the effects caused by the manufacturing process.

자율주행 차량의 충돌 조건을 고려한 운전자 거동 시뮬레이션

In this study, the behavior of the driver was derived by conducting a crash simulation considering automated vehicle accident conditions using autonomous emergency braking (AEB) and a human body model (HBM). Based on car-to-car intersection accident conditions in the OSCCAR project and the actual accident report, a crash accident case was selected. The base crash scenario was reconstructed by conducting a driving simulation with reference to the selected accident cases. Additional simulations applying AEB are performed. Based on the boundary conditions, a car-to-car crash simulation was performed to derive a crash pulse. This crash pulse and HBM were applied to a simple cabin model for conducting driver behavior analysis. The results confirmed that the head behavior of the driver of the opposing vehicle increased in the lateral direction.

Strain rate-dependent crash simulation of woven glass fabric thermoplastic composites

Woven fabric thermoplastic composites possess high specific strength and stiffness along with thermoformability. To utilize the full potential of these materials to achieve better crash-safe designs in automotive structural parts, their crash behavior must be predicted accurately. For reliable crash simulations, strain rate-dependent material data and equally capable material modeling are required. In this study, quasi-static and high-speed tests are carried out to measure tensile and in-plane shear properties. A strain rate-dependent continuum damage mechanics model is formulated to describe the deformation and damage behavior of woven glass fabric composites. Tensile and in-plane shear tests on a lab scale are used to calibrate the material parameters of the model. The model was implemented as a user-defined material subroutine (VUMAT) for Abaqus. Experimental results from coupon tests were used to verify the results of a single-element simulation. Finally, a structural level dynamic crash test of a u-profile on a drop tower was used to validate the predictions of the user material model.

Proposing an Uncertainty Management Framework to Implement the Evidence Theory for Vehicle Crash Applications

Abstract The purpose of this work is to enable the use of the Dempster–Shafer evidence theory (ET) for uncertainty propagation on computationally expensive automotive crash simulations. This is necessary as the results of these simulations are influenced by multiple possibly uncertain aspects. To avoid negative effects, it is important to detect these factors and their consequences. The challenge when pursuing this effort is the prohibitively high computational cost of the ET. To this end, we present a framework of existing methods that is specifically designed to reduce the necessary number of full model evaluations and parameters. An initial screening removes clearly irrelevant parameters to mitigate the curse of dimensionality. Next, we approximate the full-scale simulation using metamodels to accelerate output generation and thus enable the calculation of global sensitivity indices. These indicate effects of the parameters on the considered output and more profoundly sort out irrelevant parameters. After these steps, the ET can be performed rapidly and feasibly due to fast-responding metamodel and reduced input dimension. It yields bounds for the cumulative distribution function of the considered quantity of interest. We apply the proposed framework to a simplified crash test dummy model. The elementary effects method is used for screening, a kriging metamodel emulates the finite element simulation, and Sobol' sensitivity indices are determined before the ET is applied. The outcome of the framework provides engineers with information about the uncertainties they may face in hardware testing and that should be addressed in future vehicle design.

Development of impact attenuator analysis tools in crash scenario using Euler method and finite element analysis

The problem considered in this work is the development of simulation method for simulating car crash which utilizes simple car—impact attenuator model developed in MATLAB. Usually, car crash simulation is done using full finite element simulation which could take hours or days depending on the model size. The purpose of proposed method is to achieve quick results on the car crash simulation. Past works which utilizes simple car—impact attenuator model to simulate car crash use continuous time model and the impact attenuator parameter is obtained from the experimental results. Different from the related works, this work uses discrete time model, and the impact attenuator parameter is obtained from finite element simulation. Therefore, the proposed simulation method is not only achieving quick simulation results but also minimizing the cost and time in obtaining the impact attenuator parameter. The proposed method is suitable for parametric study of impact attenuator.

A novel approach to incorporating previous fatigue damage into a failure model for short-fibre reinforced plastics

This paper describes how previous material damage due to cyclic loading can influence the failure properties of short-fibre reinforced thermoplastics under crash conditions. To enable these effects to be incorporated into numerical crash simulations, we present a mathematical failure model that takes fatigue damage into account. We perform an experimental study on the thermoplastic PBT GF30: We initially characterize the material under cyclic loading conditions. Subsequently, the behaviour of pre-damaged material specimens in highly dynamic loading experiments is compared to that of virgin specimens to determine the loss of strength due to fatigue damage. Based on the aforementioned experimental results, we develop, calibrate and implement an analytical model for use in the commercial software LS-DYNA.

Study on the Crash Simulation Model for the Protective Member of Flexible Structure

The protective member of flexible structure such as the protective net has been developed against tornado missiles in nuclear power plants. However, modeling methods/knowledge of flexible structure were not much reported as compared with stiff structure like concrete walls. In order to obtain the modeling methods, a comparative study has been carried out about simulation models where a sheet of protective rope net subjected to impact loading with a free drop weight is modeled using various materials/conditions. Here, ropes of the net are modeled with beam elements, while a steel flame supporting the net is modeled with shell elements. The simulation results show that it is better not to consider compressive stiffness about the material model of the ropes. As for the frame using an elasto-plastic material, it is important to consider strain rate dependency of flow stress. It is also found that the turnbuckle which connects the ropes and outside of the frame should be modeled with rigid body and that boundary conditions considered here for the frame do not influence the result so much.