Lithium battery temperature prediction method and device, equipment and storage medium

Abstract

The invention discloses a lithium battery temperature prediction method, device and equipment and a storage medium, and the method comprises the steps: obtaining the measurement data of a lithium battery, inputting first spatio-temporal data into an ML-ELM model for model order reduction processing, obtaining dimension-reduced second spatio-temporal data, inputting circuit parameters and the second spatio-temporal data into the first ELM model, and obtaining the temperature of the lithium battery. The space-time dynamic is implicitly learned through the measured data, so that the step of reconstructing a constitutive equation in the thermal process of the battery is avoided, the analysis difficulty is reduced, and the application range is wide; and inputting the low-dimensional time coefficient into a K-ELM model for reconstruction processing to obtain dimension-raised third spatio-temporal data, performing error compensation processing according to the circuit parameters, the first spatio-temporal data and the third spatio-temporal data to obtain a temperature prediction value of the lithium battery, and compensating errors of order reduction processing and reconstruction processing to improve the accuracy of the temperature prediction value. The invention can be widely applied to the technical field of lithium batteries.

Description

technical field [0001] The invention relates to the field of lithium batteries, in particular to a lithium battery temperature prediction method, device, equipment and storage medium. Background technique [0002] Energy crisis and environmental pollution are two major problems faced by countries in the world today. In order to effectively alleviate these two problems, lithium-ion batteries have the advantages of high voltage platform, high energy density, high power density, high conversion efficiency and no pollution. become a focus of attention today. However, the temperature of the lithium battery has a great impact on its performance. Excessive temperature not only reduces the performance and life of the battery, but also induces thermal runaway or even explosion of the battery, resulting in catastrophic consequences. An accurate lithium battery temperature prediction model can effectively improve the battery. Thermal management performance while reducing accident risk...

AI Technical Summary

Problems solved by technology

However, the temperature of a lithium battery has a great impact on its performance. Overheating not only reduces the performance and life of the battery, but also induces thermal runaway and even explosion of the battery, resulting in disastrous consequences. An accurate lithium battery temperature prediction model can effectively improve the performance of the battery. Thermal management performance, while reducing the risk of accidents, so the establishment of a suitable and efficient lithium-ion battery temperature prediction model is very important

[0003] Some existing temperature prediction methods, for example: 1) Based on Newman’s pseudo-two-dimensional model, establish a transient temperature model of a lithium battery cell under different discharge rate conditions, design a constant current charge and discharge experimental device, collect and calculate And simulate the thermal behavior data of the single battery, and analyze the results of the transient temperature parameters and the voltage distribution curve, but this method needs to establish an accurate constitutive equation of the battery thermal process, and the charging and discharging of lithium batteries involves a series of physical / chemical reactions , resulting in constitutive equations that are often difficult to construct, difficult and narrow in scope of application; 2) Lithium-ion battery equivalent low-order thermal model learning method based on extreme learning machine: use the Karhunene-Loeve method to conduct time / Space decoupling, and then integrate the extreme learning machine and the Galerkin method to predict the battery temperature, but the process of reducing the order may lead to the loss of nonlinear information and reduce the prediction accuracy

Images