Research On The Wear Detection Of The Friction Pair Of The Rail Guide Of The Lifting System Based On Machine Vision

The rail guide and sliding cage ears are important parts of the wellbore lifting system.The rail guide guides the lifting container and the sliding cage ears limit the lifting container.Under the combined action of the two,the swing amount of the lifting container is reduced,and it can run safely and stably.Since the rail guide and the sliding cage ears mainly realize the guiding and limiting functions through frictional contact,the two are prone to wear.The "Safety Regulations for Coal Mines" clearly stipulates that if the wear of the two exceeds the prescribed threshold,it will affect the safe operation of the shaft lifting system,causing immeasurable losses and must be replaced.However,limited by the actual working conditions of the well,manual inspection is still the main method of detecting the wear of the rail guide and the sliding cage ears.However,the shaft lifting system must be stopped during the manual inspection process,which reduces the production efficiency.At the same time,it is affected by the harsh and dim wellbore environment and artificial subjective factors,which makes it difficult to ensure the detection accuracy.In order to realize the real-time wear detection of the rail guide and the sliding cage ears,this paper introduces a non-destructive inspection method based on machine vision.This method uses the industrial camera to collect the image of the rail guide and the sliding cage ears,and extracts the edges of the rail guide and the sliding cage ears in the image.Fit the characteristic contour,and finally measure the amount of wear based on the characteristic contour.The main research work of this paper is as follows:(1)A wear detection system for the rail guide and the sliding cage ears is designed.According to the actual working conditions of the wellbore,the wear detection requirements of the rail guide and the sliding cage ears,a wear detection scheme based on machine vision is proposed.This solution uses a combination of a single-digit laser and a camera.Both the camera and the laser are installed on the top of the tank.The laser is irradiated on the rail guide along the upper surface of the sliding cage ear.The camera and the sliding cage ear are at a certain inclination to collect images,and the collected images include all to be measured characteristics.The solution only needs to collect an image to meet all measurement requirements,and the installation is simple and easy to implement.On the basis of the designed scheme,complete the hardware selection and the design of the light source system.(2)Research on the image preprocessing technology of the rail guide and the sliding cage ears.Before acquiring images,Zhang Zhengyou calibration method is used to calibrate the internal parameters and distortion parameters of the optical system to reduce the influence of distortion on the imaging quality.Next,the possible noise pollution in the wellbore was analyzed,and a combination of preprocessing algorithms for weighted average grayscale,adaptive binarization,and Gaussian filtering was determined.This combination of algorithms can reduce the impact of noise while retaining the edge contour information of the rail guide and the sliding cage ears.Finally,the sub-pixel edge detection is used to subdivide the edge pixels after the edge detection,which improves the detection accuracy of the detection system.(3)Research on the image-based measurement method of the rail guide and the sliding cage ear size parameters.The Harris corner detection method is used to find the corner information of the reference object in the image,obtain the inverse perspective transformation matrix,correct the image to the situation where the image is shot vertically,and define the pixel equivalent.For the sliding cage ears,the mask is used to divide them into three parts: left,right and bottom;traverse the contour points to find the lowest point of wear.Using Hough straight line detection to find the boundary line equation,and the distance from the lowest point of wear to the boundary line is the current wear value of the sliding cage ear.For the rail guide,the green laser line is extracted by converting the image to the HSV space,and the left and right side endpoints of the laser line are obtained.The distance between the two ends is the current width of the rail guide.(4)Experimental research on the wear detection of the rail guide and the sliding cage ear.On the basis of the wear detection system,the system software and hardware experiment platform was designed and built,and the feasibility of the detection method and the detection scheme was verified through experiments.Based on the Lab VIEW platform,the host computer system software is designed to display the current image of the rail guide and the sliding cage ear,the amount of wear and the location of the wear in real time.Tested the stability of the detection system under slight shaking.Finally,compare the wear of the tank track and the tank ear detected by this system with the actual wear of the two,the results show that the upper computer software system can accurately display all the detection information,the system has good antiinterference performance,and the detection error is within 5%,which meets the actual detection requirements.There are 69 figures,9 tables,and 91 references in this thesis.