Research On Fault Diagnosis Method Of Mine Fan Based On Time Series And Blind Source Separation

Mine fan is the key equipment in the production of coal mine,must ensure the normal operation of the fan,when the wind turbine failure occurs,it will cause damage to the normal production of coal and the staff.In order to realize the safe production of coal mine,it is necessary to require the ventilator to operate safely,reliably and efficiently.However,the mine fan is working in a continuous long term,and the working conditions are very complex and unpredictable.In order to ensure the fan to operate in a safe condition,when the fault occurs,the timely detection and prevention,and to analyze and determine the fault,this paper presents a fault diagnosis system of mine fan the time series and the blind source separation.The research object of this paper is to study the explosion proof counter rotating axial flow fan in mine.This paper introduces the mine ventilation system and the types,working principle and characteristics of the fan,the structure of each part of the fan and the layout of the sensor of the wind turbine system.It also systematically study the characteristics of fan failure and diagnostic method.It provides a theoretical basis for the on-line monitoring and fault diagnosis of fan.In this paper,several of the time series change trend of fan can be acquired through extraction and analysis of the fan signal by means of time series method.Combining with the time series prediction toolbox of MATLAB dynamic neural network to predict the future trend of the fan,a method of mine fan fault prediction is presented based on time series.By means of blind source separation pretreatment method and research on blind source separation algorithm,a method of mine fan fault diagnosis is acquired based on blind source separation.In this paper,by means of the simulation experiments and time series prediction based on and Fast ICA blind source separation algorithm to extract the source signal study of wind turbine bearings,the method based on time series prediction and fast ICA blind source separation effectively used to the fan bearing fault feature extraction and diagnosis.According to the technical requirement,some improvements are made on the basis of the existing monitoring system.The monitoring interface can be created through using Kingview software.The data storage can be realized through using SQL Server database.Manage and analyze the data through using Matlab's powerfulfunction.Based on the effective combination of software,the transmission,storage,analysis and feedback of data can be realized,the complete system of wind turbine can be established.Finally,in the field practice process,the AIC9900 instrument is applied to the actual fan fault detection,and realize the combination of theory and practice.