Title: Based on IAGA-BP neural network internal temperature prediction of solar car

Authors: Tingting Zhang

Addresses: Department of Civil Engineering, Jinan Engineering Polytechnic, 6088 East Jingshi Road, Jinan, Shandong, 250200, China

Abstract: The instability of the interior temperature of the solar car during winter heating leads to a lot of unnecessary energy consumption. This paper proposes to predict internal temperature based on the improved adaptive genetic algorithm-back propagation neural network (IAGA-BP). Firstly, the traditional adaptive genetic algorithm's crossover and mutation probability are improved to get the improved adaptive genetic algorithm. Then, an internal temperature prediction model based on IAGA-BP neural network is established. Finally, the results of IAGA-BP are compared with results based on the particle swarm optimisation-back propagation neural network model (PSO-BP). The experimental results show that the mean absolute and square errors of IAGA-BP temperature prediction are 0.2810 and 0.1070. IAGA-BP network has better prediction accuracy than PSO-BP network. Therefore, IAGA-BP neural network temperature prediction model can reasonably predict the temperature to achieve the purpose of energy-saving.

Keywords: new energy vehicle; economic and environmental protection; adaptive genetic algorithm; BP neural network; temperature prediction.

DOI: 10.1504/IJVD.2023.134742

International Journal of Vehicle Design, 2023 Vol.92 No.2/3/4, pp.300 - 315

Received: 24 Apr 2022
Received in revised form: 11 Sep 2022
Accepted: 17 Nov 2022
Published online: 09 Nov 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article