Research On Magnetic Leakage Detection System For Defects Of Mining Wire Rope

Steel wire rope has the advantages of high tensile strength,wear resistance,low elongation,and is not easy to break suddenly,and is widely used in coal mines and other fields.With the increase of service time,the wire rope will have defects such as broken wire and wear,which will affect production safety in severe cases.As a relatively mature nondestructive testing technology for wire rope,magnetic leakage detection technology has the advantages of fast detection speed and low cost.However,at this stage,the detection equipment is less intelligent,the magnetic leakage signal is weak,and small defects are prone to missed detection and misjudgment.In order to optimize the non-destructive testing equipment of wire rope and improve the detection accuracy,this thesis studies the magnetic leakage detection system for defects of mining wire rope,and the main research work and innovation points are as follows:Firstly,according to the theory of magnetic leakage detection,the finite element model of wire rope detection is established by COMSOL.On the basis of verifying the accuracy of the model,the influence of defect size,probe lift height,and detection direction on the simulation signal is explored.Through simulation,the general structure of the detection probe is designed,the excitation intensity and excitation method required for the magnetization of the wire rope to saturation are determined,and the magnetic concentration structure that can improve the detection signal is designed,which provides a theoretical basis for the system design and signal processing in the following article.Secondly,a set of wire rope non-destructive testing device was built,including two parts: probe and hardware system.The probe is composed of excitation device,encoder,Hall sensor and other parts to realize the functions of wire rope excitation,signal collection,and position recording;The hardware system includes signal acquisition device,communication circuit,power supply device and other parts to realize signal acquisition,saving,display and other functions,and write supporting applications to drive the normal operation of the detection device.Then,a remote monitoring system based on the Internet of Things was designed.Use WIFI module to connect non-destructive testing equipment to the Internet to achieve TCP(Transmission Control Protocol)transmission;Build TCP server and web server in the cloud to provide services for remote monitoring system;Use a non-relational database to save inspection data;Visualize data in the web interface.Finally,the remote monitoring function of wire rope condition is realized,and multiple devices can be monitored at the same time.Finally,an intelligent identification method for wire rope defects based on convolutional neural network is proposed.Experimental test and evaluation were carried out through sample preparation and construction of experimental equipment,and the original one-dimensional magnetic leakage signal was obtained;It is preprocessed for detrending,filtering,envelope and other preprocessing to improve the signal quality.The preprocessed raw signal is interpolated and imaged to construct a machine learning dataset;The transfer learning network based on Alex Net structure is trained,and the solution to the defect recognition problem in small sample cases is proposed,and intelligent and high-accuracy defect classification is realized,and the experimental results show that the recognition accuracy is as high as 99.2%.This thesis optimizes the nondestructive testing equipment of wire rope,improves the accuracy of defect judgment,and provides a reference basis for the field leakage detection of wire rope defects,which has certain theoretical and application value.This thesis has 120 figures,11 tables,and 92 references.