Analysis And Study On Vibration Signal And Fault Of Mine Ventilator

Coal mine ventilation machine is an important electromechanical equipment to ensure the safe operation of coal mines.The main function of the coal mine ventilator is to provide fresh air flow to the next line of workers,to meet the oxygen needs of underground personnel,to dilute and extract toxic and harmful gases and coal dust,dust and cooling down the well to ensure safe production.Once the coal mine ventilator fails or stops working It will pose a huge threat to the safe operation of coal mines and cause serious social and economic losses.Based on this,the common fault problem of the main fan,through the online monitoring system to make a predictive diagnosis can not only reduce the equipment failure rate,extend its service life,but also reduce the accident rate,which is very important for the safe production of coal mines.The occurrence rate of coal mine ventilator fault has,the characteristics of ambiguity,suddenness and uncertainty,so its fault symptom and fault type are difficult to express by the mathematical function of general linear relationship.In this paper,the mining fan is taken as the research object.On the basis of studying the fault diagnosis system of coal mine main ventilator at home and abroad,according to the working characteristics of the ventilator,the principle of vibration and vibration of the main ventilator and the analysis method of vibration are studied.The basic theoretical analysis,numerical simulation and experimental combination method are used,and the vibration data of the fan is transmitted by using the mixed programming and interface technology to realize the online monitoring and fault diagnosis analysis and warning of the main fan.The modal vibration signal data is decomposed by empirical mode decomposition(EMD).The modal aliasing occurs when EMD decomposes the pulse and noise interference signals.When the empirical mode decomposition(EEMD)is used,the vibration signal is used.Decomposition,the fault eigenvalue is extracted according to the distribution of eigenmode energy.The fault type of the coal mine ventilator is diagnosed by the extreme learning machine algorithm.Since ELM randomly sets the input weight and threshold of the hidden layer,it also optimizes its hidden layer parameters through particle swarm optimization(PSO).The PSO algorithm is improved by the shifting operator and the excitation factor(IPSO),which enables the particle swarm to reduce its computation time in the process of optimizing the hidden layer parameters.The fault diagnosis analysis of coal mine ventilator is carried out by IPSO-ELM algorithm,and compared with other algorithms such as BP.The results show that IPSO-ELM algorithm has significant advantages in accuracy,stability and rapidity.Using the interaction between the host computer and MATLAB,the fault type diagnosis of the vibration monitoring data shows that the fault diagnosis algorithm model of this subject is feasible,can effectively monitor the fault of the coal mine fan,and realize the stable and fast signal data transmission.The fault identification efficiency and high accuracy rate,the main fan online monitoring and diagnosis system has a simple structure,reasonable process and good practicability.