Vision-Based Measurement for the Transverse–Longitudinal–Rotational Displacement of Hoisting Rope by Modified Lucas–Kanade Algorithm
Vision-Based Measurement for the Transverse–Longitudinal–Rotational Displacement of Hoisting Rope by Modified Lucas–Kanade Algorithm
复制标题
DOI：
10.1109/tim.2023.3284927
复制
发表时间：
2023
影响因子：
5.6
通讯作者：
Yunchang Zhang;Guohua Cao;Gonghua Wang
中科院分区：
文献类型：
--
作者：
Yunchang Zhang;Guohua Cao;Gonghua Wang
在矿井提升系统中，由于外界激励和提升钢丝绳自身结构的影响，提升钢丝绳会产生横向-纵向-旋转（TLR）运动。提升钢丝绳的运动参数是提升系统故障诊断和振动控制的基础。本文提出了一种改进的Lucas-Kanade（LK）方法来识别提升钢丝绳的TLR位移，称为TLR-LK算法。首先，构造了钢丝绳的径向-旋转翘曲函数，推导了其迭代计算公式。提出了一种基于投影最小化的钢丝绳轴线定位和参数识别方法，得到钢丝绳的横向位移和径向尺寸。最后，在运动平台上和煤矿现场实验中，与已有的典型方法进行了对比，验证了TLR-LK算法的性能和可行性。实验结果表明，该算法能够同时获得提升钢丝绳的横向、纵向和旋转参数，具有较高的精度和效率。
In the mine hoisting system, the hoisting rope generates transverse–longitudinal–rotational (TLR) motion due to external excitation and the structure of the hoisting rope. The motion parameters of the hoisting rope are the basis for the fault diagnosis and vibration control of the hoisting system. In this study, a modified Lucas–Kanade (LK) method is proposed to identify the TLR displacement of the hoisting rope, called TLR-LK algorithm. First, the longitudinal–rotational warp function for wire rope is constructed and the iterative calculation formula is derived. Then, a method of wire rope axis positioning and parameter identification is proposed based on projection minimization to obtain the transverse displacement and radial dimension of the rope. Finally, the performance and feasibility of the proposed TLR-LK algorithm are demonstrated by comparing the results with existing representative methods in the motion platform and a field experiment in colliery. Experimental results reveal that the proposed algorithm can obtain the transverse, longitudinal, and rotational parameters of the hoisting rope simultaneously with higher accuracy and efficiency.