Estimating State of Charge for xEV batteries using 1D Convolutional\n Neural Networks and Transfer Learning

2020/11/02 by Arnab Bhattacharjee, Bhattacharjee, Arnab, Ashu Verma +5
Engineering · #Advanced Battery Technologies Research #FOS: Computer and information sciences #FOS: Electrical engineering #Fault Detection and Control Systems #Fuel Cells and Related Materials #Machine Learning (cs.LG) #Signal Processing (eess.SP) #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2011.00841

openalex publication_date 2020/11/02 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28

Abstract

In this paper we propose a one-dimensional convolutional neural network\n(CNN)-based state of charge estimation algorithm for electric vehicles. The CNN\nis trained using two publicly available battery datasets. The influence of\ndifferent types of noises on the estimation capabilities of the CNN model has\nbeen studied. Moreover, a transfer learning mechanism is proposed in order to\nmake the developed algorithm generalize better and estimate with an acceptable\naccuracy when a battery with different chemical characteristics than the one\nused for training the model, is used. It has been observed that using transfer\nlearning, the model can learn sufficiently well with significantly less amount\nof battery data. The proposed method fares well in terms of estimation\naccuracy, learning speed and generalization capability.\n

Related

• Predictive Modeling of Charge Levels for Battery Electric Vehicles using CNN EfficientNet and IGTD Algorithm
• Analysis of NARXNN for State of Charge Estimation for Li-ion Batteries\n on various Drive Cycles
• A Temporal Convolution Network Approach to State-of-Charge Estimation in\n Li-ion Batteries
• State-of-charge Estimation of a Li-ion Battery using Deep Learning and Stochastic Optimization
• State-of-Charge Estimation of a Li-Ion Battery using Deep Forward Neural\n Networks