Improved bat algorithm optimization ELM-based fault diagnosis method of engine fuel system

Abstract

The invention relates to an improved bat algorithm optimization ELM-based fault diagnosis method of an engine fuel system, and belongs to the technical field of mechanical engineering automation. Thefault diagnosis method comprises the steps of acquiring a vibration signal of the fuel system in real time by a sensor, and extracting a characteristic vector of the sensor; optimizing a weight and athreshold of an extreme learning machine by employing a bat algorithm, and building a network structure of the optimized extreme learning machine; and inputting the characteristic vector obtained by extraction to the optimized extreme learning machine for training to obtain a fault diagnosis model of the engine fuel system, and performing fault diagnosis of the engine fuel system. The fault diagnosis method has the advantages that the classification speed is rapid, and the fault diagnosis accuracy of the engine fuel system is improved.

Description

technical field [0001] The invention relates to a fault diagnosis method of an engine fuel system based on an improved bat algorithm optimization ELM, belonging to the technical field of mechanical engineering automation. Background technique [0002] The engine is the "heart" of the entire mechanical equipment, and the fuel system of the engine determines the working performance of the engine to a large extent. Due to the complex structure of the fuel system, it is also a frequent component of failure. According to statistics from authoritative organizations, among the various failures of diesel engines, the failure of the fuel system accounts for as high as 27% of the total. It can be seen that once the fuel system breaks down, the economic and other losses it brings are relatively serious. [0003] Early fault diagnosis mostly relied on personal experience, and sometimes some relatively simple instruments were used to assist judgment. However, this diagnostic method is...

AI Technical Summary

Problems solved by technology

However, this diagnostic method is not only affected by uncertain human factors, but also has a relatively low accuracy rate of diagnostic results and takes a long time.

Images