Deep Learning in the State of Charge Estimation for Li-Ion Batteries of Electric Vehicles: A Review

Abstract

As one of the critical state parameters of the battery management system, the state of charge (SOC) of lithium batteries can provide an essential reference for battery safety management, charge/discharge control, and the energy management of electric vehicles (EVs). To analyze the application of deep learning in electric vehicles’ power battery SOC estimation, this study reviewed the technical process, common public datasets, and the neural networks used, as well as the structural characteristics and advantages and disadvantages of lithium battery SOC estimation in deep learning methods. First, the specific technical processes of the deep learning method for SOC estimation were analyzed, including data collection, data preprocessing, feature engineering, model training, and model evaluation. Second, the current commonly and publicly used lithium battery dataset was summarized. Then, the input variables, data sets, errors, and advantages and disadvantages of three types of deep learning methods were obtained using the structure of the neural network used for training as the classification criterion; further, the selection of the deep learning structure for SOC estimation was discussed. Finally, the challenges and future development directions of lithium battery SOC estimation using the deep learning method were explained. Over all, this review provides insights into deep learning for EVs’ Li-ion battery SOC estimation in the future.

Advancing State of Charge Management in Electric Vehicles With Machine Learning: A Technological Review

“…Some current methods, designed to handle complex battery behaviors, may introduce overcomplication when applied to simpler battery systems. For EVs with straightforward battery designs, the added complexity may not necessarily translate into proportional benefits, leading to inefficiencies in terms of computational resources and implementation costs [70][71][72]. In simple battery systems, the added complexity resulted in an 11.17% increase in computational resources without significant improvement in accuracy, highlighting the inefficiency of applying complex models [51,54].…”

State of Charge Estimation Model Based on Genetic Algorithms and Multivariate Linear Regression with Applications in Electric Vehicles

“…On the other hand, AI algorithms and machine learning techniques have also been considered to estimate the SOC in several applications; the most used and relevant techniques are stochastic Fuzzy Neural Networks, Adaptive Neuro-Fuzzy Inference System (ANFIS), Neural Networks such as Recurrent Neural Networks (RNN) and Fed-Forward Neural Networks (FNN), and Support Vector Machines (SVM), among others [ 22 , 23 ]. More recently, Deep Learning (DL) techniques have also been considered during the SOC estimation; in fact, Convolutional Neural Networks (CNN) [ 19 ] and autoencoders [ 24 ] are the most preferred. Nevertheless, although DL techniques can lead to the achievement of accurate results, their implementation can be associated with a high computational burden because a significant amount of data needs to be processed (i.e., CNN) and a priori knowledge is also mandatory to set specific values in the hyperparameters (i.e., autoencoder).…”

High‐precision state of charge estimation of lithium‐ion batteries based on improved particle swarm optimization‐backpropagation neural network‐dual extended Kalman filtering

“…Meanwhile, the development of deep learning provides an emerging solution for SOC estimation 21 . As a data‐driven method to solve the SOC estimation problem of lithium batteries, deep learning has the advantages of high precision and short modeling time 22 . The author proposes a stacked bidirectional long and short‐term memory (SBLSTM) neural network for SOC estimation, which makes full use of battery time information to estimate SOC value 23 .…”