An online fault diagnosis method for fuel cell system

Abstract

The present invention disclosed an online fault diagnosis method for the fuel cell system, including obtaining the first fault diagnosis model. The first fault diagnosis model is obtained by real -time data and machine learning methods based on electric heaps. The real -time data of the electric heap is passed throughObtained electric heap benchmark testing experiments; during the operation of the fuel cell system, obtaining real -time measured electric stack operation data, using the input parameters of the first fault diagnosis model, and obtaining the result of the fault diagnosis of the electric heap;The second fault diagnosis model contains standard data corresponding to each auxiliary component in the fuel cell system; obtains the real -time running data of each auxiliary component in the fuel cell system, and uses it as the input parameter of the second fault diagnostic model to obtain eachThe fault diagnosis of the auxiliary components; the present invention is based on the existing fuel cell system. It does not need to increase the sensor and test equipment without additional increases, and will not interfere with the operation of the fuel cell system. The implementation cost is low and easy to promote.

Description

technical field [0001] The invention belongs to the technical field of fuel cells, and in particular relates to an online fault diagnosis method for a fuel cell system. Background technique [0002] A fuel cell is a device that directly converts the chemical energy of fuel into electrical energy. It has the advantages of high operating efficiency, cleanliness, no pollution, and low noise. It is expected to replace the traditional heat engine to solve the environmental pollution problem of the energy system. At present, fuel cells have been popularized and applied in the fields of automobiles, drones, and stationary power generation, and have broad application prospects in the future. [0003] The fuel cell system includes a stack, an air supply system, a hydrogen supply system, a thermal management system, etc. It is a complex nonlinear system. During operation, faults such as water flooding, membrane drying, gas shortage, short circuit, catalyst poisoning, etc. may occur, ...

AI Technical Summary

Problems solved by technology

[0004] Although a variety of fault diagnosis methods for fuel cell systems have been proposed in the prior art, they still cannot meet the needs of practical applications. There are mainly the following problems: 1) Most of the diagnostic methods are only proposed for one type of fault and cannot be directly used in practice. 2) Most of the fault diagnosis methods are only for fuel cell stacks, but in fact the fault detection of key components such as air compressors and hydrogen circulation pumps is also indispensable; 3) The diagnostic methods are complicated and costly to implement. Cannot be used for online troubleshooting

Due to the above reasons, a standardized and unified fault diagnosis method that can be popularized and used has not yet been formed in the field of fuel cell technology.

Images