Method for SoC Estimation in Lithium-Ion Batteries Based on Multiple Linear Regression and Particle Swarm Optimization

Castanho, Diego Solak; Guerreiro, Márcio; Silva, Ludmila C. A.; Eckert, Jony Javorski; Alves, Thiago Antonini; Tadano, Yara de Souza; Stevan, Sérgio Luiz; Siqueira, Hugo Valadares; Corrêa, Fernanda Cristina

doi:10.3390/en15196881

Cited by 18 publications

(11 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is imperative to acknowledge that, although linear regression is a clear and understandable methodology, the association between SOC and influential parameters within a battery system may not consistently adhere to a strictly linear pattern. In instances where the relationship exhibits non-linearity, more sophisticated machine learning methodologies, including but not limited to dual linear regression, random forest regression, support vector machines, and deep learning techniques such as ANNs, could be contemplated to effectively capture the intricate non-linear associations present in the dataset [59][60][61][62][63][64][65].…”

Section: Linear Regression (Lr) Modelsmentioning

confidence: 99%

Comparative Study-Based Data-Driven Models for Lithium-Ion Battery State-of-Charge Estimation

Hussein,

Esoofally,

Donekal

et al. 2024

Batteries

View full text Add to dashboard Cite

Batteries have been considered a key element in several applications, ranging from grid-scale storage systems through electric vehicles to daily-use small-scale electronic devices. However, excessive charging and discharging will impair their capabilities and could cause their applications to fail catastrophically. Among several diagnostic indices, state-of-charge estimation is essential for evaluating a battery’s capabilities. Various approaches have been introduced to reach this target, including white, gray, and black box or data-driven battery models. The main objective of this work is to provide an extensive comparison of currently highly utilized machine learning-based estimation techniques. The paper thoroughly investigates these models’ architectures, computational burdens, advantages, drawbacks, and robustness validation. The evaluation’s main criteria were based on measurements recorded under various operating conditions at the Energy Systems Research Laboratory (ESRL) at FIU for the eFlex 52.8 V/5.4 kWh lithium iron phosphate battery pack. The primary outcome of this research is that, while the random forest regression (RFR) model emerges as the most effective tool for SoC estimation in lithium-ion batteries, there is potential to enhance the performance of simpler models through strategic adjustments and optimizations. Additionally, the choice of model ultimately depends on the specific requirements of the task at hand, balancing the need for accuracy with the complexity and computational resources available and how it can be merged with other SoC estimation approaches to achieve high precision.

show abstract

Section: Linear Regression (Lr) Modelsmentioning

confidence: 99%

Comparative Study-Based Data-Driven Models for Lithium-Ion Battery State-of-Charge Estimation

Hussein,

Esoofally,

Donekal

et al. 2024

Batteries

View full text Add to dashboard Cite

show abstract

“…In the paper presented by Castanho et al [35], the authors reported that the literature points out several approaches to accomplish SoC estimation, presenting high error, low reliability, and high computational costs. From this perspective, they proposed a model to perform SoC estimation in lithium-ion batteries for electric vehicles in four different scenarios.…”

Section: Battery Performance Estimationmentioning

confidence: 99%

Second Life of Lithium-Ion Batteries of Electric Vehicles: A Short Review and Perspectives

et al. 2023

Self Cite

View full text Add to dashboard Cite

Technological advancement in storage systems has currently stimulated their use in miscellaneous applications. The devices have gained prominence due to their increased performance and efficiency, together with the recent global appeal for reducing the environmental impacts caused by generating power or by combustion vehicles. Many technologies have been developed to allow these devices to be reused or recycled. In this sense, the use of lithium-ion batteries, especially in electric vehicles, has been the central investigative theme. However, a drawback of this process is discarding used batteries. This work provides a short review of the techniques used for the second-life batteries of electric vehicles and presents the current positioning of the field, the steps involved in the process of reuse and a discussion on important references. In conclusion, some directions and perspectives of the field are shown.

show abstract

“…Following the input of the optimal parameters, LS-SVM outputs the results of the prediction. The particle swarm optimization algorithm produces the most optimal values for relevant parameters (Castanho et al 2022). In Figure 4, the blue curve depicts how rapidly the PSO method converges, indicating that the PSO is highly quick at quickly and precisely determining the optimal solution, which aids in determining the best LS-SVM parameters and increases the model's validity and accuracy.…”

Section: Pso Program Parameters Selectionmentioning

confidence: 99%

An improved particle swarm optimization-least squares support vector machine-unscented Kalman filtering algorithm on SOC estimation of lithium-ion battery

Zhou

Wang

Xie

et al. 2023

International Journal of Green Energy

View full text Add to dashboard Cite

For real-time monitoring and safe control of electrical vehicles, it is important to accurately estimate the state of charge of lithium-ion batteries. A combined data-driven modeling approach based on Least squares support vector machine based on particle swarm optimization and unscented Kalman filter is proposed to obtain a better state of charge estimation accuracy. In this article, least squares support vector machine is used to establish the nonlinear connection between current, voltage, and SOC, and the parameters of least squares support vector machine are optimized by particle swarm optimization to improve the accuracy of voltage estimation, and the state and measurement equations are established by Least squares support vector machine in unscented Kalman filter for SOC estimation. The experimental results show that the maximum voltage error for the voltage prediction made with the PSO optimized model is 0.5 V. The maximum SOC error under various working situations is similarly kept to 0.5%, which is a significant improvement compared to the traditional algorithm. The above data show that the PSO considerably increases the precision of the Least squares support vector machine, as well as the estimation accuracy of the voltage and SOC, demonstrating the effectiveness of the model.

show abstract

Method for SoC Estimation in Lithium-Ion Batteries Based on Multiple Linear Regression and Particle Swarm Optimization

Cited by 18 publications

References 79 publications

Comparative Study-Based Data-Driven Models for Lithium-Ion Battery State-of-Charge Estimation

Comparative Study-Based Data-Driven Models for Lithium-Ion Battery State-of-Charge Estimation

Second Life of Lithium-Ion Batteries of Electric Vehicles: A Short Review and Perspectives

An improved particle swarm optimization-least squares support vector machine-unscented Kalman filtering algorithm on SOC estimation of lithium-ion battery

Contact Info

Product

Resources

About