Battery health state estimation method based on charging data and LSTM neural network

Abstract

The invention relates to a battery health state estimation method based on charging data and an LSTM neural network. The method comprises the following steps: constructing an original data set; preprocessing the data set; performing feature extraction on the input data in the data set; performing correlation coefficient analysis on the input features and the target values; constructing a neural network model structure; training a neural network model; optimizing the neural network model; evaluating the neural network model and embedding the neural network model into the battery management system; and performing battery state-of-health online estimation. According to the method, the battery SOH is estimated by using the relatively stable charging data, dimension reduction processing is performed on the input data, and correlation analysis is performed on the input data and the output data, so that the operation speed of a battery SOH estimation network model is improved, and the battery SOH estimation precision is also improved.

Description

Technical field [0001] The invention relates to the field of battery technology, and specifically relates to a battery health state estimation method based on charging data and LSTM neural network. Background technique [0002] The proposal of the "double carbon" goal will certainly accelerate the leap-forward development of new energy power generation. The high proportion of fluctuating and intermittent new energy has increased the uncertainty of power system dispatching operations and pressure on peak and frequency regulation, affecting the consumption of new energy and the efficient and optimized operation of the power system, becoming an important challenge facing new power systems. Lithium-ion batteries have the advantages of high energy density, long cycle life, and no memory effect, and are one of the important technical routes for new energy storage systems. As the number of charge and discharge cycles of lithium-ion batteries increases, the state of health (SOH) of...

AI Technical Summary

Problems solved by technology

The existing data-driven battery SOH estimation method generally uses the voltage, current, and temperature of the battery during charging and discharging as the input of the neural network to estimate the remaining power of the battery, but this method has the following problems: ① The amount of input data If it is too large, the calculation speed of the battery SOH estimation is very slow; ②The correlation analysis between the input features and the output target is not considered, and the input features with poor correlation will be retained, which will reduce the estimation accuracy; ③The timing dependence is serious, and the battery SOH Estimation needs to consider the correlation problem between information with a sufficiently large time interval

Images