Research On Belt Conveyor Fault Diagnosis System Based On Image Processing

With the wide application of belt conveyors in the field of industrial transportation,especially in the fields of port coal,the failure of key parts of belt conveyors,such as motors,rollers,and drums,has gradually increased.Therefore,for the normal operation of the belt conveyor,it is very important to ensure the normal and stable operation of key parts.The traditional fault inspection method is mainly through manual inspection.The long-term manual inspection method gradually reflects its shortcomings.It has low work efficiency,strong subjectivity,and waste of time,which can no longer meet the requirements of modern intelligent ports.Aiming at the shortcomings of traditional inspection methods,this paper takes the belt conveyor in Rizhao Port as the research object,and designs a belt conveyor fault diagnosis system based on infrared image processing.The experiment proves that this system can replace the traditional manual inspection method.Under the condition of no shutdown and no contact,the working condition of the belt conveyor is transmitted to the designed system through the instrument,so as to complete the fault diagnosis and analysis of the belt conveyor and ensure the reliability of the conveyor operation.The main research work of this paper includes:(1)The working environment of the belt conveyor is complex.The collected infrared images generally contain Gaussian noise and salt and pepper noise.The presence of noise will affect the quality of the image and affect the subsequent image recognition.Aiming at the noise problem,this paper uses a median filter and Improve the image denoising method combined with wavelet threshold denoising,which has a better filtering effect on salt and pepper noise and Gaussian noise,and makes up for the shortcomings of traditional denoising algorithms that can only filter out one type of noise.The method is verified by MATLAB.(2)Aiming at the complex background in the infrared image of the belt conveyor,in order to eliminate the influence of background interference on the fault location,this paper adopts a method combining morphology and improved area growth method,and uses the improved area growth method to roughly segment the fault location,And then refine the fault edge by morphological algorithm,and finally use Canny operator to extract the edge information.Through MATLAB experiment simulation,it is proved that this method can effectively segment the key parts of the belt conveyor and lay the foundation for the subsequent operation.(3)For the image recognition of belt conveyors,SVM classifier is used for image recognition.Traditional kernel functions cannot have generalization and learning capabilities at the same time.In order to achieve this,this paper applies a hybrid kernel function,and optimizes the parameters by improving the chaotic particle swarm algorithm,and calls the lib SVM toolbox experiment through MATLAB The classification model of the verification structure can accurately identify the key parts of the belt conveyor.(4)Aiming at the fault diagnosis of the key parts of the belt conveyor,an infrared fault diagnosis system for the belt conveyor is designed using the method of MATLAB and Lab VIEW,and different key parts of the belt conveyor are identified according to the image,and different diagnosis is adopted.Method for fault diagnosis.The system is equipped with an intelligent inspection robot to perform real-time monitoring in the working state of the belt conveyor,and accurately perform image recognition and fault diagnosis of the belt conveyor,which can reduce the failure while avoiding unnecessary economic losses.Time for diagnosis,improve the efficiency of inspection,increase the safety of belt conveyor,and extend the working life of belt conveyor.