Power battery pack fault fusion diagnosis method and system based on improved CNN

Abstract

The invention discloses a power battery pack fault fusion diagnosis method and system based on an improved CNN, and the method comprises the steps: carrying out the wavelet packet decomposition of a voltage change signal and an SOC change signal of a lithium battery, and obtaining an energy value to form an input feature vector; enabling the diagnosis network to perform preliminary diagnosis on the power battery pack fault; judging whether the preliminary diagnosis result meets a definite diagnosis condition or not; if yes, obtaining a diagnosis result of the power battery pack; if the diagnosis condition is not met, performing auxiliary diagnosis on the power battery pack fault by adopting a CNN network; performing fusion diagnosis on the preliminary diagnosis result and the auxiliary diagnosis result through a D-S evidence theory method; and judging the fusion diagnosis result to obtain a final diagnosis result. According to the method and system, the structure of the CNN network isimproved, the optimal convolution kernel size in the convolution layer is determined through the BIC criterion, the definite diagnosis condition is judged, and the auxiliary diagnosis network is further adopted for auxiliary and fusion diagnosis, so that the fault diagnosis accuracy of the power battery pack is improved.

Description

technical field [0001] The present invention relates to the technical field of battery fault diagnosis, in particular to an improved CNN-based power battery pack fault fusion diagnosis method and system. Background technique [0002] In recent years, lithium-ion batteries have become an important part of power batteries due to their advantages such as high cell voltage, high cycle times, high energy density, and no pollution. Ensuring the safe and reliable operation of lithium-ion batteries is the core of its development. At present, my country's lithium-ion battery technology is not fully mature, and the initial characteristics of inconsistent failures of battery packs are not easy to be found. The inconsistency of lithium-ion battery packs means that there are large deviations in parameters such as capacity, internal resistance, and voltage of batteries of the same model and specification, which will lead to overcharge and overdischarge of the battery during charging and d...

AI Technical Summary

Problems solved by technology

However, as the number of CNN layers increases, the size of the final feature vectors continues to decrease. If only the features with the highest complexity in the last layer are used for classification, there is a risk of overfitting, which leads to a decrease in the accuracy of diagnosis.

Images