Design And Research Of Head Sheave Deflection Monitoring System Based On Machine Vision

In the mine hoisting system,head sheave is one of the key operating devices,it undertakes the bearing load force and guidance task.However,due to the busy operation of long time and heavy load,the deflection occurs,which leads to major accidents.Because of the rotation of head sheave,it is not easy to install the sensor itself,and the detection error is large when the sensor is installed in the adjacent structure,this paper proposes a fault monitoring system of head sheave deflection based on machine vision,which monitors the head sheave deflection in a non-contact way.This paper uses machine vision processing technology,then completes the design and development of the measurement system of head sheave’s deflection distance based on image processing technology,with then with the help of the Labview Software,which is developed by United States National Instruments and its related visual module.Next,transmits its measurement data to the fault diagnosis platform of Labview,so as to realize the online monitoring of head sheave deflection fault.Based on the operation characteristics of the mine hoisting system,this paper proposes that the lateral deflection of the head sheave will aggravate the lateral vibration displacement of the wire rope.The article takes the horizontal deflection of the head sheave as the lateral excitation in the lifting system,and establishes the dynamic model of the lateral vibration of the lifting wire rope.Then,the finite difference method was used to solve the vibration equation.Next,using dynamic response simulation,it is verified that the lateral vibration displacement of hoisting wire rope will be significantly aggravated by deflection of the head sheave.According to the working environment of the head sheave,the functional requirements of the monitoring system is determined.Then,the paper designs the hardware system for on-line monitoring of deflection faults of the head sheave based on machine vision,completes the overall system architecture design and hardware equipment selection.So that it achieves effective communication and transmission between hardwares.This paper develops a procedure of measuring the dimension of head sheave deflection,using Vision core module of NI and Vision Assistant.The article uses median filtering method,does image preprocessing such as image enhancement,by obtaining spectral components using fourier transform on the frequency.Then Canny operator is used for edge feature extraction.After smooth filtering by two-dimensional gaussian function,the edge gradient amplitude and gradient direction of the region of interest is obtained by using the first order differential operator.Finally,edge detection,size measurement and other process is achieved.The camera is calibrated by zhang’s calibration method,the conversion between pixel coordinates and world coordinates is realized,and the reliability of calibration information is verified.Adding above calibration information to calibrate the distance measurement system,article realizes the conversion between pixel unit and physical unit of dimension measurement,completes designing of software system for distance measurement of head sheave deflection fault based on image processing technology.The paper built an experimental platform for monitoring the deflection of the head sheave based on machine vision.Then,experiments are carried out on a lifting device.Paper conducts 20 experimental tests on the simulated head sheave with known distance,which verified the accuracy and stability of the distance measurement system.This paper also verifies the stability of the system by installing it in the Coal mine to site test.