Remote Fault Diagnosis System Of Mine Belt Conveyor Idler

Due to the advantages of large transportation volume long transportation distance high transportation efficiency and continuous transportation belt conveyors are widely applied and have become one of the key transportation equipment in the mining production process.As the mine belt conveyor is in the hub of mine transportation,it is important to ensure its safe operation and prevent vicious accidents.As a typical largescale rotating machine,the mine belt conveyor contains a large number of idler groups,which is the main hidden danger of conveyor fire accident.There are some problems in manual patrol inspection of these faults,such as low efficiency,high labor intensity and poor real-time performance.Taking the belt conveyor idler as the research object,combining with relevant theories and technical means such as signal acquisition,signal transmission,feature extraction and fault diagnosis.The main work f this paper could be summarized as follows:(1)On the basis of introducing the structure of mining belt conveyor and idler,the common faults of mining idler are analyzed,the causes of faults and diagnosis mechanism are studied,and the fault detection method based on acoustic signal is determined according to the fault frequency theory of rolling bearing.At the same time,the corresponding fault simulation test bench is designed for the research content of this subject,in order to obtain the experimental data of the research target.(2)According to the functional requirements of the remote fault diagnosis system for the idler of belt conveyor,the core controller of the system-STM32F767 is selected and the corresponding system hardware circuit is designed around it.The hardware circuits of the system are divided into core controller module,data acquisition module,storage module and wireless transmission module according to their functions.The hardware circuit of each module is designed to complete the design of the hardware circuit of the diagnosis system.(3)By combining information Entropy with signal analysis methods in time domain,frequency domain and time-frequency domain respectively,three different characteristic indexes based on information Entropy are constructed,namely singular spectral Entropy,power spectral Entropy and Wavelet Packet Energy Entropy.The multi-level feature extraction method based on information Entropy is studied.(4)Based on the consideration of correct rate of fault diagnosis,a fault diagnosis algorithm using particle swarm optimization least squares support vector machine is proposed.LSSVM algorithm has good learning ability and generalization ability,and can solve small sample and non-linear problems well.Four different characteristic vectors of rolling bearing motion signals are taken as input vectors of LSSVM model and PSO algorithm is used to optimize the model parameters,thus avoiding blindness in parameter selection.The experimental results show that the optimized LSSVM model can diagnose rolling bearing faults well.(5)Develop fault diagnosis system for mine idler bearing.The remote acoustic signal acquisition system is built on the basis of the built belt conveyor idler fault simulation test bench.Based on the MATLAB GUI platform,the fault diagnosis program which integrates signal acquisition,fault feature extraction and fault diagnosis is programmed,and the overall test of the system is carried out on the test bench,realizing the remote online detection of mine idler fault.At the end of this thesis,the summarization of the research and expectation of the relation technology development were resented.This thesis has 52 figures,14 tables,90 references.