Energy storage battery capacity prediction method, server and computer readable storage medium

Abstract

The invention provides an energy storage battery capacity prediction method, a server side and a computer readable storage medium, and the method comprises the following steps: obtaining battery data, and carrying out the preprocessing of the battery data; selecting a battery capacity degradation characteristic quantity and a battery capacity from the battery data; initializing cuckoo search algorithm parameters and generating an initial solution, and calculating an adaptive value of the initial solution; generating a new solution according to a step length formula and a nest position formula, comparing adaptive values of the initial solution and the new solution, and iterating to obtain an optimal solution; endowing the weight and threshold information of the optimal solution to a parameter space of the Elman neural network, and performing parameter fine tuning training; and inputting the test set into the trained Elman neural network, and outputting a prediction result of the battery capacity. Through the method, the prediction speed and the prediction precision of the residual capacity of the battery can be improved.

Description

technical field [0001] The present disclosure generally relates to the technical field of battery energy storage, and specifically relates to a method for predicting the capacity of an energy storage battery, a server, and a computer-readable storage medium. Background technique [0002] Energy storage batteries are widely used in new energy vehicles, power station energy storage and other fields, and their reliability and service life are the focus of attention. Accurately predicting the remaining battery capacity of energy storage batteries is of great significance and value to ensure the healthy use of batteries and prolong the service life of batteries. The research on methods for predicting the remaining battery capacity of energy storage batteries has received extensive attention and attention. [0003] At present, the prediction methods for the remaining capacity of the battery are mainly the model method and the data-driven method. Because the model method cannot sim...

AI Technical Summary

Problems solved by technology

[0003] At present, the prediction methods for the remaining capacity of the battery are mainly the model method and the data-driven method. Because the model method cannot simultaneously consider the influence of the environment and load characteristics on the battery capacity degradation, it is often impossible to achieve accurate modeling; the data-driven method has difficulty in initial parameters. Defects such as determination, slow prediction speed, etc.;

Images