基于RFID的无源无线螺栓松动故障定位算法  

Passive Wireless Bolt Loosening Fault Location Algorithm Based on RFID

作　　者：冉运蒸 鲁罗平 谢蓉 宋坤 田相鹏 梁会军 钟建伟 廖红华 RAN Yunzheng;LU Luoping;XIE Rong;SONG Kun;TIAN Xiangpeng;LIANG Huijun;ZHONG Jianwei;LIAO Honghua(College of Intelligent Systems Science and Engineering,Hubei Minzu University,Enshi 445000,China)

机构地区：[1]湖北民族大学智能科学与工程学院,湖北恩施445000

出　　处：《湖北民族大学学报（自然科学版）》2024年第3期401-408,共8页Journal of Hubei Minzu University:Natural Science Edition

基　　金：国家自然科学基金项目(62341305);恩施州“启航专项”科技计划项目(D20230023);湖北省重点研发计划项目(2023BAB120)。

摘　　要：针对水电机组锥管进人门螺栓松动故障定位问题,提出了基于射频识别(radio frequency identification,RFID)的无源无线水电机组锥管进人门螺栓松动故障定位改进虚拟参考消除(virtual reference elimination,VIRE)算法。首先,确定超高频RFID读写器下所有螺栓松动故障辨识节点内标签的接收信号强度指示(received signal strength indicator,RSSI)值;然后,采用样条插值法插入虚拟标签,计算出虚拟螺栓松动故障辨识节点内标签的RSSI值;最后,采用动态阈值(threshold,Th)使最邻近螺栓松动故障辨识节点内标签数接近最优值,求出待测螺栓松动故障辨识节点内标签的定位坐标并记录,选择几组数据取平均值作为最终结果。结果表明,改进VIRE算法的平均定位误差相对于基于动态有源RFID校准的位置识别(location identification based on dynamic active RFID calibration,LANDMARC)算法、传统VIRE算法分别降低了98.72%、25.27%。该研究成果可以满足复杂环境下的水电机组锥管进人门螺栓松动故障定位需要。Addressing the issue of bolt loosening fault localization on the cone access door of hydroelectric generating unit,the improved virtual reference elimination(VIRE)algorithm was proposed,which was based on passive wireless radio frequency identification(RFID)technology.Initially,the received signal strength indicator(RSSI)values of all tags within the bolt loosening fault identification nodes under an ultra-high-frequency RFID reader were determined.Subsequently,spline interpolation method was employed to insert virtual tags,enabling the calculation of RSSI values for tags within virtual bolt loosening fault identification nodes.Finally,a dynamic threshold(Th)was adopted to optimize the number of tags within the nearest bolt loosening fault identification nodes,and the positioning coordinates of the tags within the tested bolt loosening fault identification nodes were determined and recorded.The average of a set of results was taken as the final outcome.The results indicated that the average positioning error of the improved VIRE algorithm was 98.72%and 25.27%lower than that of the location identification based on dynamic active RFID calibration(LANDMARC)algorithm and the traditional VIRE algorithm,respectively.This research can satisfy the positioning requirements for bolt loosening faults on the cone access door of hydroelectric generating units in complex environments.

关 键 词：RFID 定位算法 RSSI值 螺栓松动 故障定位 样条插值法 动态阈值 

分 类 号：TP391.44[自动化与计算机技术—计算机应用技术]