A battery state of health evaluation method and device based on an improved adaptive fuzzy neural network

Abstract

The application discloses a battery state of health evaluation method and device based on an improved adaptive fuzzy neural network, and belongs to the technical field of battery management. The method comprises the following steps: firstly, a lithium battery capacity aging model is constructed, a plurality of groups of health indexes are extracted from charging and discharging data and are subjected to median-quartile range normalization processing; then, key features are screened out by comprehensively considering feature accuracy, correlation and extraction complexity through a random forest-AHP method; subsequently, Bayesian optimization is adopted to determine hyperparameters such as the number of membership functions and initial standard deviation, a rule activation mechanism of a dynamic parameter optimization system driven by working conditions is introduced, model parameters are iteratively updated in combination with a recursive least square method and a gradient descent method, and high-precision estimation is realized; finally, a full life cycle evaluation system covering four stages of normal, attention, abnormal and serious is constructed, and health indexes are dynamically adjusted. Through the introduction of dynamic parameters driven by working condition features, high-precision SOH estimation under complex working conditions is realized.

Images