A state of health estimation method for electric vehicle Li-ion batteries using GA-PSO-SVR

Abstract

State of health (SOH) is the ratio of the currently available maximum capacity of the battery to the rated capacity. It is an important index to describe the degradation state of a pure electric vehicle battery and has an important reference value in evaluating the health level of the retired battery and estimating the driving range. In this study, the random forest algorithm is first used to find the most important health factors to lithium-ion batteries based on the dataset released by National Aeronautics and Space Administration (NASA). Then the support vector regression (SVR) algorithm is developed to predict the SOH of a lithium-ion battery. The genetic algorithm-particle swarm optimization (GA-PSO) algorithm is brought forward to optimize the parameter values of the SVR, which could improve the estimation accuracy and convergence speed. The proposed SOH estimation method is applied to four batteries and gets a root mean square error (RMSE) of 0.40% and an average absolute percentage error (MAPE) of 0.56%. In addition, the method is also compared with genetic algorithm-support vector regression (GA-SVR) and particle swarm optimization-support vector regression (PSO-SVR), respectively. The results show that (i) compared with the PSO-SVR method, the proposed method can decrease the average RMSE by 0.10%, and the average MAPE by 0.17%; (ii) compared with the GA-PSO method, number of iterations under the proposed method can be reduced by 7 generations.

Research on the stress magnetic coupling effect of a permalloy based on an optimized Jiles Atherton model in magnetic shielding devices

“…In this paper, the GA-PSO is used to identify the JA model parameters of permalloy. GA-PSO adds the selection and crossover operations of GAs to PSO, which has the characteristics of fast optimization and less tendency to fall into local optima [33][34][35]. Generally, when identifying the parameters of the JA model, the root mean square error of the measured and analyzed magnetic field intensity values is used to construct the objective function, and the fitness value is written as:…”

“…In this paper, the number of iterations is 200, the cross-probability r is 0.5, the inertia coefficient ω is 0.5, the value of the acceleration factor is c 1 = c 2 = 2. The parameters are set according to the rules of GA-PSO algorithm [33][34][35] and adjusted according to the calculation results to get better results.…”

A State-of-Health Estimation Method for Lithium Batteries Based on Fennec Fox Optimization Algorithm–Mixed Extreme Learning Machine

“…Based on the charging curves trend, 12 health indicators were extracted [32], categorized into four types: (1) time and time ratio of different charging stages: CC charging stage time F1, CV charging stage time F2, their ratio F3, and total charging time F4; (2) the integration of current curves in different charging stages over time: CC charging stage…”

“…Based on the charging curves trend, 12 health indicators were extracted [32], categorized into four types: (1) time and time ratio of different charging stages: CC charging stage time F1, CV charging stage time F2, their ratio F3, and total charging time F4; (2) the integration of current curves in different charging stages over time: CC charging stage current integration F5, CV charging stage current integration F6, and total charging stage current integration F7; (3) the integration of temperature curves in different charging stages over time: CC charging stage temperature integration F8, CV charging stage temperature integration F9, and total charging stage temperature integration F10; (4) the maximum slope of the curve: the maximum slope of the charging voltage curve F11 and the maximum slope of the charging current curve F12. Taking the B07 battery as an example, these features were standardized (as shown in Equation ( 16) in Section 3.4) to conform to a normal distribution.…”

Comprehensive Review of Machine Learning, Deep Learning, and Digital Twin Data-Driven Approaches in Battery Health Prediction of Electric Vehicles

“…However, many of the current methods are developed under ideal laboratory conditions and do not account for the complex and dynamic operational environments of electric vehicles. Therefore, researchers have proposed various methods to estimate the battery SOH under realistic EV conditions, such as data-driven models, machine learning algorithms, and regional capacity analysis [4][5][6].…”