Research On Dynamic Coal Quantity Measurement Of Belt Conveyor Based On Machine Vision

At present,the power of the belt conveyor is determined according to the peak capacity of the coal mine production and a certain redundancy factor,and it will keep constant high-speed operation after starting.Because of the imbalance of the production of the working face,the belt conveyor is often in the situation of low load high-speed,or even no-load high-speed,resulting in a great waste of electric energy.Adjusting the belt speed according to the change of coal quantity can effectively overcome the above phenomena,and the real-time accurate measurement of dynamic coal quantity is the key to realize the belt speed adjustment,so it is of great significance to study the technology in depth.The traditional real-time measurement method of dynamic coal quantity of belt conveyor has many problems,such as low accuracy and many restrictions,so this paper studies the real-time measurement technology of dynamic coal quantity based on machine vision.According to the shape of the material transported by the belt conveyor and the camera imaging model,combined with the Scheimpflug condition that the vision system needs to meet,this paper finally determines to use the monocular structured light slanting distance measurement scheme to measure the coal quantity.Based on this scheme,the mathematical model of the volume measurement is established,and the calibration principle of the parameters to be determined in the model is described.Based on this,a dynamic coal quantity measurement system based on machine vision is designed.The key equipment of the system is selected and designed.According to different field environment,two kinds of mounting brackets,quick dismantling structure and hoisting structure,are designed.According to the characteristics of continuity and directionality of light fringes formed by single line structured light on the surface of coal flow,an adaptive threshold binarization algorithm for directional sliding and an algorithm for extracting the center of structured light with directional growth are proposed,and a broken line fitting and connection algorithm based on cardinal spline curve is proposed for the fracture of light fringes.In the aspect of algorithm development,Matlab is used to verify the feasibility,correctness and processing effect of the algorithm in the early stage of development,and then the camera demo is used as the main body,and C++ language is used to realize the algorithm of coal quantity measurement system.The experimental results show that the binarization and center extraction algorithm in this paper has a greater advantage in operating efficiency,and the broken line fitting algorithm has a certain effect in the case of small-scale fracture.The whole algorithm of coal quantity measurement system can process one frame image in about 0.099 S.with the increase of coal quantity,the measurement error of the system is stable at about 1%,which can meet the requirements of actual working conditions.