Biodiesel is a good substitute for petroleum diesel, because of its good environmental performance, good engine starting performance, good fuel performance, wide source of raw materials, renewable and so on, it has a wide application prospect. However, the fat composition and content of different raw materials are different, and the performance and emission characteristics of the prepared biodiesel also show different trends during combustion. Based on the types of raw materials for biodiesel production, this paper classifies biodiesel and summarizes the characteristics of different kinds of biodiesel. Firstly, the physical and chemical properties and the preparation process of biodiesel was introduced in detail. Secondly, this paper is different from the classification of biodiesel based on the technological iteration of biodiesel production, but classifies biodiesel according to the source of raw materials, which provides a basis for the summary of the table and the horizontal comparison of biodiesel produced from various raw materials. Then, the use characteristics and changing trends of biodiesel prepared from different kinds of raw materials are summarized and sorted out in the table, and the summarized experimental results are visually presented, the characteristics of biodiesel produced from different raw materials were evaluated, in all the studies on biodiesel, it has been shown that in terms of pollution emissions, the engine found that the emission of CO and carbon oxides was reduced when using most biodiesel, and most studies showed that the emission of hydrocarbons was also reduced, but the emission of NO_x was increased, which is almost a common feature of all different types of biodiesel. Finally, according to the limitations and shortcomings of different types of biodiesel in the use process, the future direction of biodiesel use characteristics optimization was proposed.

Traffic accident mortality, as one of the most important safety indicators, is rarely considered in current AEB performance evaluation studies. Given that the pedestrians have the highest fatality rate in traffic accidents, the paper proposes an evaluation method for pedestrian AEB performance field testing. The pedestrian AEB test scenarios are constructed according to the C-NCAP and I-VISTA protocols and a performance evaluation model is established by using the analytic hierarchy process. The proposed test and evaluation model is verified through real vehicle testing. The results show that the pedestrian AEB performance score of the test vehicle is 7.23, which is much lower than the benchmark score of 34.05. This indicates that the AEB performance of the vehicle is only average.

At present, the vehicle side impact safety evaluations rarely consider the effect of pre-crash braking on occupant posture. In order to study the occupant displacement induced by braking during the pre-crash phase and its impact on occupant kinematics and injury outcomes under side impact conditions, the paper combines volunteer experiments with CAE simulations. Three side impact models, including a standard posture model, a muscle-tensed model, and a muscle-relaxed model, were developed to compare differences in occupant kinematics and key injury indicators across the models. The results show that, during braking, the volunteers experienced greater displacement under the relaxed muscle state, with the maximum displacements of the head and first thoracic vertebra (T1) reaching 225 mm and 145 mm, respectively. This displacement significantly changed the contact between the upper body of the dummy and the side restraint system during the side impact, creating a risk that the dummy's chest could move outside the effective protection zone of side airbag. The peak Y-acceleration of the head was increased by 163.98 m/s²(79.4%), and the maximum abdominal compression was increased by 13.53 mm (64.0%) during the side impact. These results provide valuable insights for the development of advanced restraint systems and integrated safety testing methods.

This paper investigates the impact characteristics, damage evolution in residual performance, and the damage repair methods of PET foam sandwich structures through a combination of experimental and numerical approaches. The objective is to clarify the mechanisms of impact-induced deformation and damage, the principles for determining damage tolerance and the fundamental structural repair strategies, thereby providing systematic guidance for the application of PET foam sandwich structures. In addition, the successive orthogonal design optimization method is used to perform discrete multi-objective optimization of the structural repair scheme, and a set of optimal repair parameters is obtained for the sandwich structure. These results provide a valuable reference for the effective repair of PET foam sandwich structures.

As simulation testing plays a more important role in the development process of autonomous driving technology, the confidence level of simulation testing has become a key focus in the industry. However, there is no unified metric or evaluation framework for measuring simulation test confidence. To address this gap, this paper proposes a confidence evaluation method based on two dimensions: timing and trend. Firstly, an objective characterization index set was constructed to describe the vehicle's control timing and the variation trends of the control parameters. Secondly, four typical scenarios covering both longitudinal and lateral control were designed, with six sets of experiments conducted for each scenario. A software-in-the-loop (SIL) simulation platform was developed to perform the simulation tests. Finally, the Pearson correlation coefficient was used to evaluate the consistency of parameter trends, and the relative error was used to assess the consistency of control timing. The results show that the correlation coefficients of the variation curves for both lateral and longitudinal control parameters are all above 0.95, and the relative errors of the control timing indicators are below 5%, which is within the acceptable range.

The rapid development of autonomous driving technology has increased the demand for the authentic and diverse simulation test scenarios. However, traditional methods for constructing autonomous driving simulation scenarios heavily rely on manual editing, which is not only costly but also limited by the combination and complexity of scene elements, making it difficult to meet the comprehensive testing and validation needs of autonomous driving systems. To address this issue, this paper proposes a method for generating autonomous driving simulation test scenarios based on Large Language Models (LLMs). This approach utilizes a pre-trained LLM, enhanced through LoRA fine-tuning, and integrates a scenario language parser to produce a structured interpretive language, which is used to generate scenario files. The generated text is processed by a parser to convert it into usable scenario files, effectively addressing the issues of overly long texts and model hallucinations, while also achieving the specialization of a general model's capabilities.

To meet the safety requirements for driver fatigue detection and warning in the most common L2 level intelligent driving scenarios, a four-class classification of driver states is achieved using EEG signals collected by a multi-channel wireless EEG cap. A convolutional recurrent neural network is used to train models using different combinations of frequency-domain, time-domain and nonlinear features. The results show that the best recognition performance is achieved when combining the differential entropy from nonlinear features with the average absolute value from time-domain features. Furthermore, three integration strategies are proposed to fuse base classifiers trained on different feature combinations. The method achieves accurate multi-class classification of driver fatigue states in a cost-effective and user-friendly manner, and promotes the application of wearable devices in driving scenarios to improve driving safety.

A monocular vehicle distance estimation based on attention mechanisms is proposed to improve estimation accuracy on uneven road surfaces. Channel and spatial attention mechanisms are incorporated into the ImVoxelNet neural network to enhance contour perception and feature discrimination, thereby reducing missed vehicle detections. Redundant information in inverse perspective mapping is eliminated through region-of-interest corner calibration, mitigating image distortion. To address variations in vehicle pose, a pose-interference-aware camera extrinsic parameter matrix is proposed, and a coordinate transformation model for uneven surfaces is established. Finally, the proportional relationship between real-world and inverse perspective images is used to construct a distance estimation model, achieving accurate estimation of the longitudinal and lateral distances. Experimental results show that the proposed method maintains a relative error below 3% within a longitudinal range of 80 meters and a lateral range of 4 meters, validating its effectiveness and accuracy.

Amid the accelerating electrification of the global automotive industry, power batteries, as the core components of new energy vehicles (NEVs), have become a strategic focal point in the global competition to achieve green and low-carbon automotive development. Currently, China leads the world in power battery technology, industrial scale, and supporting ecosystem. This paper investigates the carbon footprint accounting methods for power batteries, comprehensively studies the current development status and challenges of China's power battery carbon footprint, and deeply analyzes the experiences and insights gained from the EU's carbon footprint management. It is proposed to accelerate the improvement of the top-level design for carbon footprint management, establish a comprehensive carbon footprint accounting framework, facilitate international mutual recognition of accounting systems, explore the market-oriented operation of power battery carbon footprints, and promote mutual recognition of international standards and regulations.

The significant reduction in electric vehicle driving range at low temperatures has limited their widespread adoption in extremely cold regions. To address this industry challenge, this paper proposes an indirect heat pump system for light commercial electric vehicles operating in severe cold climates. The five-way valve design used in the system enables a high level of system integration, and satisfies the thermal demands of various vehicle subsystems under low-temperature conditions. A 1-D simulation model of the vehicle thermal management system was established and validated through bench testing. The low-temperature performance of the system was assessed using climate chamber experiments on the actual vehicle. The heating performance and energy consumption of the proposed system were compared with those of the traditional Positive Temperature Coefficient (PTC) heating mode. The results show that the proposed thermal management system can meet the heating demands at low temperatures, with the average foot outlet temperature reaching 32.3 ℃ at an ambient temperature of -5 ℃. Compared with the traditional PTC heating, the heat pump system proves superior energy-saving performance, reducing system energy consumption by more than 50% and extending the driving range by approximately 15%.

The mixed Weibull distribution is widely used for modeling failure distributions and predicting durability. In practical engineering development, accurate parameter estimation for the model is critically important. Therefore, improving the estimation accuracy of the mixed Weibull distribution has become an urgent and challenging issue in the field. Based on the original mixed Weibull distribution, this paper proposes an optimized parameter estimation approach using a novel B&R-SSA algorithm. Firstly, this method establishes an iterative optimization model to estimate the location, scale, and shape parameters based on the method of successive approximation. To address the low efficiency and tendency of the original Salp Swarm Algorithm (SSA) to become trapped in local optima, a novel B&R-SSA algorithm is proposed by introducing a “betrayal” behavior mechanism and an adaptive inertia weight strategy. This improved algorithm is then applied to solve the iterative model. Finally, Monte Carlo simulations and engineering case studies are conducted. Both the simulation and experimental results demonstrate that the proposed method achieves good accuracy and computational efficiency in estimating the parameters of the mixed Weibull distribution.

This paper investigates the mesh independence of wheel fatigue life simulations. To address issues such as long fatigue assessment cycles, high testing costs, and the strong influence of mesh size on the finite element simulation accuracy, fatigue life predictions were carried out using both the nominal stress method and the local stress-strain method. The mesh independence of the simulation results was verified based on the grid convergence index (GCI) theory, leading to the recommendation of an optimal mesh configuration for fatigue life simulation. A fatigue life simulation method is proposed with high-quality meshing and low error range, which provides a theoretical foundation for establishing mesh independence in fatigue life prediction. Taking an aluminum alloy wheel as the research object, the wheel was meshed with different grid sizes according to the GCI theory. Radial and bending static simulations of the wheel, as well as fatigue life prediction simulations, were then conducted. The simulation results under different grid sizes were analyzed and a criterion was proposed to assess the rationality of tetrahedral meshing for aluminum alloy passenger car wheel hubs. Corresponding verification analysis was conducted. The results show that the GCI can effectively supports the mesh independence analysis and validation in wheel fatigue life simulations. The results in this paper provide a reliable basis for evaluating the suitability of tetrahedral mesh sizes in such applications.

To understand female consumers' perceptual needs for electric vehicle styling and to solve the problem of matching these needs with electric vehicle designs, a research method based on perceptual analysis for styling electric vehicles for women is proposed. This study analyzes the styling differences between electric vehicles and traditional fuel vehicles to determine the direction of electric vehicle styling. Additionally, morphological analysis is used to deconstruct the design of electric vehicles. Based on the theoretical framework of perceptual engineering, perceptual vocabulary and representative samples for electric vehicles are collected and selected. Factor analysis is used to extract the main factors affecting the styling of electric vehicles for women. The judgment matrices are constructed using the hierarchical analysis to calculate the relative weights of the perceptual terms. Using quality function deployment tools, a mapping model between perceptual imagery and automobile design elements is built to transform users' perceptual imagery into design elements for women's electric vehicle styling. It is concluded that female users prefer EV styling designs with a wheelbase ≤ 60% of the vehicle length, a larger A-pillar tilt angle, a grille-less design, closed wheels, rounded or fused headlights, slim taillights, pop-up door handles, rounded shoulder lines, and floating roof windows. The research framework based on perceptual analysis can effectively extract female consumers' needs for electric vehicle styling. This framework provides some guidance for the design of electric vehicles for women.

Taking the zero-flush door system of a certain vehicle model as the research object, a finite element model of the glass guide seal is developed. Using MSC.Marc, the compression load of the seal is calculated under varying gaps between the sliding and fixed glass. Simulation results show that when the gap value decreases by 1.2 mm from the design value, the compression will sharply increases from 6.7 N/dm to 28.9 N/dm, which leads to jamming during glass lifting and lowering. Therefore, the critical gap tolerance threshold is determined to be ±1.2 mm. A tolerance model of the zero-flush door system is constructed using the tolerance analysis software 3DCS, and 20 000 virtual assembly simulations are conducted using the Monte Carlo method. It is found that the tolerance fluctuation range of the gap between the sliding glass and the fixed glass is ±1.39 mm, with a 0.89% probability of exceeding the critical value of ±1.2 mm. After optimizing the structure of the zero-flush door system and re-evaluating the tolerance, the fluctuation range is reduced to ±1.1 mm, and the probability of exceeding the critical value is reduced to zero. During the small-batch trial production phase of a specific vehicle model, data were collected from 30 randomly selected samples. The results showed that the gap fluctuation range was ±1.08 mm, and no jamming or sticking occurred during the lifting or lowering of the glass, indicating that the design objectives were achieved. The deep integration of finite element analysis and dimensional tolerance evaluation provides important guidance for the structural design and performance simulation of flexible systems, ensuring a high level of consistency between simulation and real-world performance.