Battery system fault diagnosis method and system

Abstract

The invention discloses a battery system fault diagnosis method and a system. According to the method, on the basis of acquired temperature signals, humidity signals, battery current signals and battery voltage signals of the battery system, a BP neural network in a multi-input and multi-output structure is used to complete battery system fault diagnosis. The system comprises a data acquisition module, and a neural network fault diagnosis module, wherein the data acquisition module is used for acquiring temperature signals, humidity signals, battery current signals and battery voltage signals of the battery system; and the neural network fault diagnosis module is used for using the BP neural network in the multi-input and multi-output structure to complete battery system fault diagnosis on the basis of the acquired temperature signals, the humidity signals, the battery current signals and the battery voltage signals of the battery system. Signals of multiple elements of the battery system are fused, fault diagnosis on a complicated battery system is realized, the result has a high accuracy rate, the diagnosis efficiency is high, an adaptive correction function is provided, and the method and the system of the invention can be used for fault diagnosis of a power battery system and an energy storage battery system of an electric vehicle.

Description

technical field [0001] The invention relates to a battery system fault diagnosis method and system based on BP neural network, belonging to the field of battery system fault diagnosis. Background technique [0002] As the development trend of the future automobile industry, the era of electric vehicles is coming. [0003] For electric vehicles, the battery system is the energy source for powering it, therefore, the normal power supply of the battery system is very important. Generally, a battery system is composed of a battery, a cooling circuit, a heating circuit, a battery management circuit, various sensors, a battery box, a high-voltage circuit, and signal lines for connecting them. The battery system is a system with a very complex structure. Therefore, its fault phenomena are diverse, and the causes of many complex faults often present the characteristics of ambiguity, randomness, and combination. [0004] At present, the diagnosis method adopted by the battery syste...

AI Technical Summary

Problems solved by technology

[0004] At present, the diagnosis method adopted in the battery system is a component-oriented diagnosis method, which draws local conclusions based on local information, and it is difficult to make an accurate judgment on the root cause of the failure.

However, if the fault diagnosis is carried out by establishing an accurate mathematical model based on the diagnostic object, it will be difficult to achieve practical results due to the complexity and nonlinear characteristics of the battery system.

Images