Data fusion method and system for fuel cell engine fault prediction

Abstract

The invention discloses a data fusion method and system for fuel cell engine fault prediction, and the method comprises the following steps: S1, obtaining an actual measurement fault sample through atest; S2, obtaining a simulation fault sample; S3, performing feature-level fusion on the actual measurement fault sample and the simulation fault sample to obtain a first fusion sample; S4, performing data-level fusion on the first fusion sample and an actual measurement fault sample to obtain a second fusion sample; and S5, jointly taking the second fusion sample and the actually measured faultsample as a training sample of a fuel cell engine fault prediction model. According to the invention, the problem of insufficient experimental data of fuel cell engine fault prediction can be solved.

Description

technical field [0001] The invention relates to the technical field of fuel cell vehicle engine fault diagnosis, in particular to a data fusion method and system for fuel cell engine fault prediction. Background technique [0002] With the continuous development of fuel cell vehicle technology, fuel cell vehicles have gradually become practical. However, in the prior art, because the fuel cell vehicle has not been put into practical use in large quantities, the fault diagnosis of the engine of the fuel cell vehicle is still seldom involved. [0003] However, there are very few studies on fuel cell engine failure prediction. The difficulty of fuel cell engine failure prediction mainly lies in: because fuel cell vehicles have not been widely used, and the failure data of fuel cell engines is too small, it is difficult for existing prediction models to make predictions based on too little data. In addition, even through large-scale trials, it is difficult to obtain a sufficie...

AI Technical Summary

Problems solved by technology

The difficulty of fuel cell engine failure prediction mainly lies in: because fuel cell vehicles have not been widely used, the failure data of fuel cell engines is too small, and the existing prediction models are difficult to predict based on too little data

In addition, even through large-scale trials, it is difficult to obtain a sufficient amount of data

Images