Air Compressor Remote Fault Diagnosis And Prediction Technology

The air compressor is an important machine for generating compressed air.Its working condition is complicated,leading to a high probability of failure.At present,most domestic enterprises is in a laggard level about air compressor monitoring and condition analysis.In this regard,this paper proposes a remote monitoring system for air compressors.While improving the monitoring level,research and design of air compressor fault diagnosis and prediction methods are used to analyze the operation status of air compressor.This article takes screw air compressors as the research object and studies its mechanical structure and working principle,analyzes the fault on rotor system failures and other common failures,and selects BP neural network after comparing and analyzing common fault diagnosis methods in this paper.Secondly,this paper studies and designs a remote monitoring system for air compressors.On the one hand the system chooses electrical parameters and thermodynamic parameters to monitor in case of the energy perspective,on the other hand the system chooses vibration signals to monitor in case of the machine itself.After analyzing three kinds of parameters measurement principles and methods,the system scheme consist of local data acquisition,wireless data transmission,and remote data processing.In the end the hardware and software are designed to achieve system function.Finally,this paper studies the fault diagnosis and prediction algorithm based on BP neural network.The principle of BP neural network is analyzed,then the article clarifies the specific steps of BP neural network in fault diagnosis application.The design of the rotor system test bench in order to obtain sample data,the BP neural network of vibration signals,studying and training BP neural network of vibration signals are executed in order.On the other hand,a BP neural network with performance parameters including electrical parameters and thermodynamic parameters is designed to assist the fault diagnosis and prediction of the air compressor and improve the comprehensiveness and reliability of the research method.At the end of this article,the system test analysis is executed.The functional test analysis of the air compressor remote monitoring system designed in this article,and the application test of the air compressor fault diagnosis and prediction method designed by BP neural network,both perform well.The feasibility of the research design content.is verified.