Energy storage lithium battery charge state evaluation method and system fusing physical model and neural network

By fusion of physical models and neural networks, the lumped parameter equivalent circuit model and long and short-term memory neural network are used to solve the problem of insufficient accuracy of lithium battery SOC estimation under complex conditions, achieving higher prediction accuracy and robustness.

CN120352779AActive Publication Date: 2025-07-22SOUTHWEAT UNIV OF SCI & TECH
View PDF 4 Cites 2 Cited by

Patent Information

Application Number
CN202510813079.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-18
Publication Date
2025-07-22
Estimated Expiration
2045-06-18

AI Technical Summary

Technical Problem

The existing lithium battery state of charge (SOC) estimation methods are insufficient in complex dynamic changes and multi-temperature conditions. Traditional methods require accurate equivalent circuit models and adaptive filtering algorithms. Data-driven neural network methods require a large amount of label data and optimization algorithms.

Method used

A method of fusing physical models and neural networks is used to extract the charge characteristic information of lithium batteries using the lumped parameter equivalent circuit model, and a new lumped parameter equivalent circuit model-long and short-term memory neural network model is constructed, combining the extended Kalman filtering and gating mechanism for SOC prediction.

Benefits of technology

Improves the accuracy and robustness of SOC prediction, simplifies the model construction process, and enhances the prediction performance in multi-temperature environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120352779A_ABST
    Figure CN120352779A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of lithium battery charge state evaluation, and discloses an energy storage lithium battery charge state evaluation method and system fusing a physical model and a neural network, and the method comprises the steps: employing a lumped parameter equivalent circuit model battery system characterization method, and extracting the charge characteristic information of a power lithium ion battery; taking the charge characteristic information of the power lithium ion battery as input characteristics of a neural network, and constructing a novel lumped parameter equivalent circuit model-long and short-term memory neural network model; and utilizing the novel lumped parameter equivalent circuit model-long short-term memory neural network model to realize state-of-charge estimation of the power lithium ion battery to be tested. According to the method, an innovative and effective solution is provided for SOC estimation of the power lithium ion battery, the model construction process is simplified, the prediction performance of the SOC in the multi-temperature environment is improved, and a new thought is provided for research and application in related fields.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Battery energy state evaluation method based on forgetting factor adaptive feedback correction

    CN115598541A

  • SOH estimation method and system based on battery electromechanical characteristic parameter fusion

    CN115902674A

  • Battery SOC estimation method fusing AEKF algorithm and LSTM neural network

    CN119986396A

  • Data-driven lithium ion battery state prediction method fusing physical information

    CN120121993A

Cited By

  • Lithium battery SOH estimation method and system based on improved TCN-GRU

    CN120891396A

  • Battery charge state estimation method and system based on physical gating neural network

    CN121348098A