Research On Belt Conveyor Fault Diagnosis System Based On Infrared Image Processing

With the steady development of transportation industry and the continuous growth of coal demand,the belt conveyor used for transporting coal plays a huge role in ports,coal mines and other places.The belt conveyor is prone to various system faults in the operation process,mainly including the tearing and deviation of the conveyor belt and the damage of the supporting roller,these faults often cause immeasurable economic losses and security accident,so the fault diagnosis of the belt conveyor has become an urgent problem to be solved.The temperature in fault area of belt conveyor is higher than that in non fault area,and the infrared image is a kind of image that reflect the target temperature,which can be targeted fault diagnosis,taking the belt conveyor of a Port in Rizhao,Shandong Province as the research object,this article designs a belt conveyor fault diagnosis system based on infrared image processing.The main research contents of this article are as follows:(1)When the thermal imager and the belt conveyor move relative to each other,the motion blur phenomenon will occur in infrared image,aiming at the motion blur phenomenon and noise problem of infrared imagery,a restoration filter based on improved median filter and Wiener filter is used for image restoration,MATLAB simulation results show that the algorithm gives consideration to both running speed and recovery effect,and has excellent comprehensive performance.Aiming at the problems of poor imaging effect and low resolution of infrared image,the adaptive histogram equalization enhancement algorithm is adopted to enhance the image of infrared image,improve the image quality and make the image clearer.(2)In order to reduce the influence of background region on target region in infrared image,an improved two-dimensional Otsu image segmentation algorithm based on morphology is proposed,firstly,the original infrared image is segmented by the improved two-dimensional Otsu algorithm,and then processed by the erosion operation and open operation in mathematical morphology,finally,the Canny operator is used to extract the edge to get the final target region,the segmentation region obtained by this algorithm is complete and the segmentation edge is obvious.SIFT algorithm is used to extract features from infrared image of belt conveyor fault,and the BOW model is used to reduce the dimension of eigenvector,after obtaining the sample database of fault infrared image,SVM classifier is used for image classification,according to the classification results of infrared image,the fault classification of belt conveyor is realized,so as to carry out troubleshooting.(3)Design the diagnosis scheme of the belt conveyor fault diagnosis system,and verify the stability of the fault diagnosis system.First,according to the temperature characteristics and damage conditions of belt conveyor fault,the fault diagnosis scheme suitable for belt conveyor are proposed.Second,fault diagnosis experiment platform of belt conveyor was designed and set up,and analysis the external factors that will affect the fault diagnosis system.Finally,the fault diagnosis experiments under different influencing factors are carried out,the experimental results show that the fault diagnosis system of belt conveyor has high reliability and accuracy.