Object detection has been one of the critical technologies in autonomous driving. To improve the detection precision, a novel optimization algorithm is presented to enhance the performance of the YOLOv5 model. First, by improving the hunting behavior of the grey wolf algorithm(GWO) and incorporating it into the whale optimization algorithm(WOA), a modified whale optimization algorithm(MWOA) is proposed. The MWOA leverages the population’s concentration ratio to calculate $$p_h$$p h for selecting the hunting branch of GWO or WOA. Tested by six benchmark functions, MWOA is proven to possess better global search ability and stability. Second, the C3 module in YOLOv5 is substituted by G-C3, and an extra detection head is added, thus a highly optimizable detection G-YOLO network is constructed. Based on the self-built dataset, 12 initial hyperparameters in the G-YOLO model are optimized by MWOA using a score fitness function of compound indicators, thus the final hyperparameters are optimized and the whale optimization G-YOLO (WOG-YOLO) model is obtained. In comparison with the YOLOv5s model, the overall mAP increases by 1.7 $$\%$$% , the mAP of pedestrians increases by 2.6 $$\%$$% and the mAP of cyclists increases by 2.3 $$\%$$% .

The thermal management system is one of the important assemblies that ensure the secure operation of electric vehicles (EVs). Using intelligent algorithms to optimize the control strategy of the thermal management system can reduce energy consumption under the premise of effective heat dissipation of EVs. This paper attempts to construct the control strategy of EV thermal management system by coupling the modified genetic algorithm (MGA) and support vector regression (SVR). Firstly, the double-population adaptive mutation method and a novel optimization process are adopted to obtain MGA. Afterward, the performance of MGA is verified by four benchmark functions compared with three typical algorithms, which are genetic algorithm (GA), double-population genetic algorithm (DPGA), and quantum genetic algorithm (QGA). The results demonstrate that the accuracy and stability of MGA are obviously better than the other three algorithms. Secondly, MGA is applied to modify parameters of SVR kernel function, and the accuracy of MGA-SVR algorithm is verified by the Auto-MPG and Computer Hardware data sets. The mean square deviations of the SVR algorithm test set are 0.0186 and 0.0806, respectively, and the mean square deviations of the MGA-SVR algorithm test set are 0.0099 and 0.0054, respectively, which fully shows that MGA-SVR have more accurate forecasting capabilities. Finally, the thermal management system model of EV is built by the one-dimensional simulation software KULI. Under the Chinese working condition, fan speed which meets the cooling requirements of the motor and controller is obtained from the KULI model, and the database is constructed. Then, MGA-SVR is trained by database and employed to predict fan speed under the Chinese working condition and obtain control strategy of the thermal management system. Compared with traditional control strategy, the thermal management system based on MGA-SVR control strategy can not only meet the radiating requirements, but also effectively reduce the power consumption of fans.

In axial fan simulation using multiple reference frame (MRF) method, an inappropriate size of rotating zone may affect the accuracy of results. In order to study the influence of axial and radial sizes in a wide flow rate range, five rotating zone sizes were investigated at seven flow rate points from 0 to 11.3 m³/s in a simplified fan setup configuration. Comparing the fan characteristic curves, it is found that the accuracy of simulation results is sensitive both to axial size and radial size, and the sensitivity varies with flow rates. In order to accurately predict the fan performance, it is recommended that the MRF rotating zone be long enough to cover the rotating backflow and pre-rotation area upstream the fan for prediction at low flow rates, and the radial size of MRF zone be large enough to include the trailing vortexes for prediction at middle to high flow rates. While the other sizes are not very influential, as long as the integrity of the blades is included.

Proton exchange membrane fuel cell (PEMFC) has become one of the new energy vehicle powertrains due to its special advantages, such as no pollution to the environment, high energy efficiency and power density. In order to improve the performance, a three-dimensional simulation model of the actual PEMFC is constructed. Combined with the mathematical models such as the electrochemical model and current conservation model, the model is calibrated by experiments. When the working voltage is 0.64 V, the error of the simulation results is 0.73%, compared with the experimental data. Then, the Taguchi method is used to design a multi-factor and multi-level orthogonal experimental scheme of PEMFC. Based on the orthogonal experimental table, the effects of different anode wave channel distortion, gas diffusion layer thickness and gas diffusion layer porosity on the current density are studied with a simulation experiment. The influence of the above factors on the orthogonal experiment results is analyzed by the signal-to-noise ratio. The regression equation is obtained by calculating the orthogonal experimental data. The t-test results are greater than 3.49, which indicates that each independent variable in the regression equation is important. R² test is 0.915, and the F test is 53.508, indicating that the regression equation is significant and the optimal and worst structural parameter combinations are predicted. The current density reaches 14,190.18 $\mathrm{A}/{\mathrm{m}}^2$ under the optimal structure combination, which is 6.14% higher than the calibrated model. Single factor experiments are carried out on these three different structural parameters to verify the effectiveness of the Taguchi method, and the best combination of structural parameters is obtained.

The 3D fan zone model is a recently developed variant of the actuator disk models, based on an input fan curve to simulate the fan effect. However, the fan curves are typically obtained in experimental situations under common standards, which could be different from the operating scenario in simulation, specifically, the definitions of flowrate and the fan appurtenances effect. In order to adapt the typical fan curve for the 3D fan zone model, an iterative correction method to modify the flowrate and exclude the fan appurtenances effect was proposed. The embedded evaluation method for fan appurtenances effect was improved by taking the fan effect into account. To verify, the iterative correction method was applied to an axial cooling fan simulation. With corrected fan curve, the maximum deviation of fan total pressure reduced from 282.6 to 0.7 Pa. The iterative correction method of fan curve is highly recommended.