Reciprocating compressor sensitive characteristic extracting and fault diagnosis method based on internet of things

Abstract

The invention relates to a reciprocating compressor sensitive characteristic extracting and fault diagnosis method based on the internet of things. Aiming at the current situation that actual early warning parameters and fault diagnosis of an existing reciprocating compressor are short of effective association, on the basis of a reciprocating compressor online monitoring and diagnosis system based on the internet of things technology, the internal correspondence of faults and features is found out through the typical fault mechanism research, and the reciprocating compressor fault diagnosis method adopting fault sensitive feature parameter extraction is provided. Sensitive feature parameters corresponding to different faults are extracted aiming at fault case data of the reciprocating compressor, and a fault sensitive feature parameter set is formed; by the adoption of different intelligent classification algorithms, an automatic fault classifier is built on the basis of the fault sensitive feature parameter set, and automatic diagnosis of unit faults is achieved.

Description

technical field [0001] The invention relates to fault diagnosis technology for reciprocating compressors, and is a sensitive feature extraction and fault diagnosis method for reciprocating compressors based on the Internet of Things. Background technique [0002] Oil refining, chemical industry, oil extraction, gas extraction and gas pipeline production are typical process industries, and reciprocating compressors are large-scale and key equipment widely used in process industry production. Due to the high pressure of reciprocating compressors, dangerous compression medium, and many faulty parts, dozens of reciprocating compressor accidents occur in China every year, including explosions, fires, broken piston rods, and cylinder collisions. The loss is immeasurable. [0003] At present, although domestic reciprocating compressors have gradually installed online monitoring systems, the unit failure analysis and diagnosis of the existing online monitoring systems rely more on ...

AI Technical Summary

Problems solved by technology

[0008] Integrating the research results of reciprocating compressor fault monitoring and diagnosis at home and abroad, there is no research work on fault diagnosis by extracting reciprocating compressor fault sensitive characteristic parameters proposed in the present invention

Images