Study On Fault Diagnosis And Prediction Based On Multi-source Parameters

Mine hoisting system is the throat part connected with the mine,which is of great significance to the safe production of coal mine.The dynamic improvement and efficiency of mine hoist system is an important index to determine whether a coal mine has efficient modern coal production capacity.The throat of the mine hoist is often known as mine,is the most important key equipment,mine is the only way to underground mine with the outside world,is shouldering the ore,materials,personnel and other important transportation liability.The security and reliability of the operation is critical to the elevator.Traditional mine hoist speed control performance is poorer,the rev.Stop,braking,logic control has many security problems,with the continuous development of computer and PLC technology,the use of advanced control technology of the traditional mining industry transformation of the traditional control system,so that the mine hoist control performance is greatly improved,and its level of automation,security and reliability have reached new heights,and adopt modern management and monitoring measures to ensure the safe running of the elevator.This article first expounds the data fusion estimation theory,summarized various data fusion method of multisensor system,is proposed based on the BP neural network and DS evidence theory quadratic fusion of multi-source data fusion algorithm.Secondly,through the analysis of various fault formation mechanism of mine hoist system,the fault tree of mine hoist system is established.And on the basis of mine system fault tree,analysis the coal mine hoisting system of fault diagnostic method,which is the precondition of multisensor data fusion technology was applied to fault diagnosis of mine hoisting system.At the same time,based on the BP neural network is constructed of steel wire rope tension,improve the mechanical and electrical machine grease temperature and oil pressure,vibration,brake drum brake torque and vibration system of multi-sensor data fusion diagnosis framework,using the BP neural network to eliminate environmental factors influence on sensor fault diagnosis system,and applies conclusions in the fault diagnosis system of ascension.Finally,this paper discusses the data fusion method based on DS evidence theory in fault diagnosis of the hoisting system of the actual application problem,by analyzing the basic principle of DS evidence theory and the basic rules to find outhow data fusion method based on DS evidence theory is applied to fault diagnosis of mine hoisting system.The DS evidence theory is proved to be able to solve the uncertainty and fuzziness of the actual monitoring system.