The state of charge (SOC) of battery is an important indicator of battery condition assessment. Accurate SOC estimation can reflect the real-time capacity of battery and effectively extend the service life of battery. In this paper, aiming at the problem that the error function of BP neural network is easy to fall into local optimum and the effect of initial weight and threshold reverse iterative updating is limited, a neural network optimized by genetic algorithm is proposed to improve the estimation accuracy and global fast optimization. By studying the equivalent first-order RC model of the battery and building the simscape simulation model to simulate the charging and discharging process of the battery, the parameters of the battery terminal voltage, current, internal resistance, temperature and battery SOC are obtained. After the data is processed, the voltage, current, internal resistance and temperature are taken as the factors of battery SOC estimation. The experimental results show that the neural network optimized by genetic algorithm can obtain the optimal initial threshold and weight globally, thus establishing a high-precision SOC estimation model of battery, avoiding local optimization and improving the estimation accuracy of battery.
Battery SOC Estimation Method Based on BP Neural Network Optimized by Genetic Algorithm
Lect. Notes Electrical Eng.
International Conference on Electrical and Information Technologies for Rail Transportation ; 2021 October 21, 2021 - October 23, 2021
Proceedings of the 5th International Conference on Electrical Engineering and Information Technologies for Rail Transportation (EITRT) 2021 ; Chapter : 6 ; 45-56
2022-02-23
12 pages
Article/Chapter (Book)
Electronic Resource
English
On Using Genetic Algorithm Optimized Activation Functions to Increase Neural Network Accuracy
British Library Conference Proceedings | 2014
|British Library Conference Proceedings | 2021
|