In order to accurately predict the V2G charging and discharging load of electric vehicles, so as to regulate the peak to valley difference of power grid load and ensure power supply stability, this paper proposed a spatiotemporal distribution prediction method for V2G charging and discharging loads of electric vehicles based on collaborative optimization of supply and demand sides. A collaborative optimization objective model for both supply and demand sides was built, the Whale Optimization Algorithm was used for iterative solution to obtain the optimal charging and discharging load curve, and the optimal charging and discharging period was determined. The influencing indicators of charging and discharging loads within the optimal time periods in different spatial regions were collected, serving as inputs for constructing a prediction model based on multiple linear regression, thus achieving the prediction of spatial-temporal distribution of electric vehicle V2G charging and discharging loads. The experimental results show that the load prediction model obtained with the proposed method has a relatively large coefficient of determination, indicating that the prediction results of this research method are closer to the actual load, and have high prediction accuracy.

To address the issue that eliminating a peak value of sound pressure level in the noise transfer function during the optimization of vehicle body noise transfer function for vehicle preparation can easily lead to new peak values of noise sound pressure level, an optimization method for vehicle body noise transfer function with the introduction of control variables is proposed. Taking the cab of a certain model as the research object, an optimization model is constructed, with the thickness of the structural panels in the cab as the variable. The proposed algorithm is used to iteratively optimize the relevant parameters. The results show that this method effectively reduces the peak value of sound pressure level in the noise transfer function within the target frequency band.

The current available capacity of the battery is difficult to obtain, and the health status of the battery is difficult to estimate accurately during the operation of the vehicle. Therefore, this paper proposed to use the parking and charging segment data of the vehicle to correct the battery capacity obtained by ampere-hour integration method through box diagram and Kalman filter algorithm. The support vector regression model was constructed for battery degradation prediction. The effective model input parameters were determined by Pearson correlation analysis. The model parameters were optimized by genetic algorithm. Results show that the fitting accuracy of the optimized model reaches 88%, which is 12% higher than that before optimization, can accurately predict the SOH of vehicle battery.

To restrain the fluctuation of high frequency noise in the vehicle caused by The uncertainty of the parameters of the acoustic package, this paper proposed an uncertainty analysis and optimization method based on sub-interval modified perturbation method for high frequency noise of automobiles. Firstly, the basic equation of statistical energy for the analysis of high frequency noise was established. Secondly, the uncertainty parameters were divided into several sub-intervals, and the perturbation radius of each sub-interval was calculated by using the modified interval perturbation method to obtain the center value and the perturbation radius of the high-frequency noise performance. Finally, the high frequency noise performance and parts quality of the vehicle were optimized by the cross-interval uncertainty optimization method. This method was applied to analyze and optimize the interior front circumference and carpet of a vehicle. After optimization, the weight of acoustic package parts decreases by 3%, the noise fluctuation range decreases by more than 35%, and the robustness of the system is significantly improved.

A driver fatigue detection method based on parallel short-term facial features is proposed to achieve faster and more accurate fatigue warning. The method utilizes the YOLOv7-MCW object detection network, which incorporates the MicroNet module, CA attention mechanism, and Wise-IoU loss function, to extract short-term facial features of the driver’s face. The parallel Informer temporal prediction network is then used to integrate the spatiotemporal information obtained from the YOLOv7-MCW object detection network, enabling the detection and warning of driver fatigue. The results demonstrate that the YOLOv7-MCW-Informer model achieves accuracy rates of 97.50% and 94.48% on the publicly available datasets UTA-RLDD and NTHU-DDD, respectively, with a single-frame detection time reduced to 28 ms, proving the excellent real-time fatigue detection performance of the model.

To study the sound quality of vehicle in rainy day driving, objective parameter calculation and subjective evaluation were conducted for the sound quality of the vehicle cockpit aerodynamic noise signals in the wind-rain field, with a correlation analysis between the two parameters. Based on the Improved Whale Optimization Algorithm-Back Propagation (IWOA-BP) algorithm, six objective parameters including loudness, roughness, jitter, speech intelligibility, speech interference and sound pressure level as input, and subjective scoring as output were used to establish a prediction model, which was compared with the traditional BP neural network prediction model and the WOA-BP prediction model. The results indicate that the mean absolute percentage error of BP, WOA-BP and IWOA-BP algorithms are 28.33%, 6.35% and 2.82% respectively, proving that the sound quality prediction model of automobile cockpit aerodynamic noise in wind-rain field established based on IWOA-BP algorithm has a higher accuracy and a better effect.

This study provided a review of classification methods and quantitative evaluation of driving behavior, which firstly expounded the meaning and representation methods of driving behavior, and divided the driving behavior classification methods into three categories: statistics based classification method, machine learning based classification method, and the hybrid (combination, integration) classification method. Different driving behavior classification methods were summarized from the aspects of representative algorithms, advantages and limitations. Secondly, the quantitative evaluation research of driving behavior was systematically described from multiple dimensions. Finally, the application status and prospect of driving behavior classification and quantitative evaluation results in many fields were introduced.

To balance braking safety and braking energy recovery of distributed drive electric vehicles, this article explored the vehicle braking torque distribution control strategy based on NSGA-II multi-objective optimization algorithm. An optimization module based on fuzzy control was established, and the optimal torque distribution coefficient was determined from the Pareto frontier optimization according to vehicle speed and demand braking torque. With an electric passenger vehicle as the research object, a brake torque distribution control strategy model was built based on Matlab/Simulink and VPAT and simulated. A hardware-in-loop simulation platform was built, and real-time performance and validity of this algorithm were verified. As was shown in the result, the brake torque distribution coefficient of the control strategy based on NSGA-II was closer to the distribution coefficient of ideal I curve, and the operation points of the motor braking efficiency zones were increased by 9.51 percentage point, and the energy recovery rate of vehicle regenerative braking was increased by 4.71 percentage point.

To improve the thermal management efficiency of vehicular lithium-ion batteries, Nano-Graphite (NG) and Nano-Aluminum (NA) were added to Paraffin (PA) to prepare Nano-Composite Phase Change Materials (NCPCM). Based on combinatorial mathematics modeling, the influence of different mass ratios of NCPCM’s physical parameters such as thermal conductivity and latent heat of phase change on battery thermal management efficiency was studied. Experimental analysis of thermal management of car lithium-ion batteries was conducted using RGO/BN/PA CPCM, RT44HC/CF CPCM and NCPCM3 under slow charge to fast charge conditions. Results show that at a 3.5 C discharge rate, NCPCM3 with a thermal conductivity of 2.1 W/(m·K) and a latent heat of 206.18 J/g could reduce the maximum temperature and maximum temperature difference of the car lithium-ion battery pack to 44.2 ℃ and 4.4 ℃ respectively. Further improvement of the thermal performance parameters of NCPCM did not significantly improve its thermal management performance. NCPCM3 shows significantly better battery thermal management efficiency than RGO/BN/PA CPCM and RT44HC/CF CPCM, ensuring normal operating temperature for the car lithium-ion battery pack.

In order to improve the ability of the collision warning system to perceive the surrounding environment, this paper proposed a collision warning system based on YOLOv5 and hazardous area judgment. Firstly, the discriminative ability and accuracy of the model were improved by the channel attention module, then, the extraction ability of the model for multi-size features was improved by using path aggregation network and spatial pyramid pooling, and finally, the warning accuracy of the warning system was improved by filtering relatively safe targets through the introduction of warning activation regions. The results show that the introduction of warning activation regions improves the accuracy, precision and recall of the warning system by 20%, 50% and 26.7%, respectively, the running speed is increased by 49.1%, which further proves the effectiveness of the method.