Study On Coal Quantity Monitoring System Of Belt Conveyor Based On Machine Vision

With the concept of building smart mine,the production of China's coal industry is developing from an extensive way to an intensive way.As an important means of transport in the coal production,belt conveyor causes much unnecessary cost during the process of production as it always runs at full speed,without adjusting speed according to the load.Domestic equipment manufacturers are actively studying intelligent speed control technology for belt conveyors,the core of which lies in the accurate measurement of coal volume and the establishment of intelligent control strategies on the basis of dynamic coal volume.To solve this problem,this paper puts forward a kind of coal monitoring system for belt conveyor based on machine vision,in which the dynamic volume of the coal on the primary conveyor is measured using a combination of laser triangulation detection and binocular vision detection,after which the results are used as a parameter to change the operating frequency of the inverter by the PLC programmable controller,thereby controlling the secondary belt conveyor to achieve energy-saving speed control operation.The experimental results show that the system can realize the dynamic identification of the coal volume on the conveyor belt,with high accuracy and can meet the real-time speed regulation requirements,which has a high practical value.The main research of this paper are as follows:1)The overall design and construction of the research platform for the coal quantity monitoring system for belt conveyor based on machine vision is carried out.This system mainly includes two functional subsystems: volume detection and speed control.The main hardware equipment includes two-stage belt conveyor,line laser binocular stereo camera,computer,Siemens PLC,frequency converter and so on.Under the development platform of Visual Studio 2017,the software design of coal volume detection system is realized based on C++ programming language combined with Qt framework,and the programming of PLC speed control system is realized based on TIA Portal software.2)For the volume detection system,the calibration of laser binocular stereo camera is studied.The linear imaging model,nonlinear imaging model and binocular vision imaging model of the camera are established.The calibration experiments of the binocular camera are carried out by using the Zhang's calibration method which is based on two-dimensional/2D checkerboard calibration.Finally,the internal and external parameters of the camera and the transformation structure parameters of the system are obtained,which is a good preparation for the following research.3)The point cloud data acquisition of the center line of the line laser fringe is studied.Firstly,the collected laser fringe image is smoothed and de-noised by means of average filtering method.Then,an iterative threshold barycenter method is proposed to extract the center line of laser fringe.For the two images of laser stripe center line extracted from the left and right cameras,the SGBM semi-global stereo matching algorithm based on Open CV is used to obtain the 3D coordinates of all points on the laser stripe centerline based on the principle of parallax and the principle of polar lines,combined with the system parameters calibrated by the cameras,to obtain the 3D point cloud data.Finally,the 3D point cloud data is reconstructed by the OpenGL module encapsulated by Qt,and the 3D image scanned by laser is drawn and displayed in real time in the upper computer interface.4)The research and experimental verification of coal volume algorithm based on point cloud data are carried out.The key of volume detection is to calculate the transverse sectional area of coal.This paper presents an improved trapezoidal area accumulation method to calculate the transverse sectional area.Then,in view of realtime dynamic volume detection of conveyor belt,a method of using RFID reader to read the electronic tag is proposed to carry out fixed segment of conveyor belt,which reduces the cumulative calculation error caused by speed deviation.After the algorithm is designed,experimental verification is carried out.The overall absolute error of the experimental results is less than 5%,which proves that the volume detection algorithm is of high accuracy,and the upper computer software runs stably,which can display 3D images of coal transportation in real time.Finally,the experimental results and causes of errors are analyzed.5)A study of the speed control system was carried out for the volume detection results.In the study,a set of speed control scheme is designed and a two-stage speed control experiment for two belt conveyors is conducted in a laboratory environment,using a USB serial relay module to connect computer and PLC,and finally realized the control of two belt conveyors by dividing the detected coal volume into levels,and finally realized the sequential start-stop and speed control of the two-stage/secondary belt conveyor.There are 50 figures,11 tables and 72 references in this paper.