基于传感器数据采集的接地刀闸机械位置检测  

Mechanical Position Detection of Grounding Knife Switch Based on Sensor Data Acquisition

作　　者：苏淑敏[1] 杜珂 于树海 梁元清 王斌[1] SU Shumin;DU Ke;YU Shuhai;LIANG Yuanqing;WANG Bin(Nanning Power Supply Bureau of Guangxi Power Grid Corporation,Nanning 530000,China)

机构地区：[1]广西电网公司南宁供电局,广西南宁530000

出　　处：《机械与电子》2023年第12期38-42,共5页Machinery & Electronics

摘　　要：位置检测是接地刀闸机械正常运转中不可缺少的技术,但位置检测过程易受机械配置高低、传感器性能和噪声等问题的干扰。为解决上述问题,提出基于传感器数据采集的接地刀闸机械位置检测方法。该方法首先利用混合型形态学滤波器对传感器采集的接地刀闸机械数据实行降噪处理,避免噪声对位置检测结果产生影响。然后通过改进k均值算法对降噪后的数据做聚类处理。最后将预处理后的接地刀闸机械数据输入到结构支持向量机中,通过平面切算法求取结构支持向量机的最优解,完成接地刀闸机械的位置检测。实验结果表明,所提方法的位置检测效率高、准确率高、抗噪能力强,位置检测效果好。Position detection is an indispensable technology in the normal operation of grounding switch machinery,but the position detection process is easily disturbed by problems such as mechanical configuration,sensor performance and noise.In order to solve the above problems,a method for detecting the mechanical position of the grounding knife switch based on sensor data acquisition is proposed.The method firstly uses a hybrid morphological filter to perform noise reduction processing on the mechanical data of the grounding switch collected by the sensor,so as to avoid the influence of noise on the position detection result.Secondly,the denoised data is clustered by the improved k means algorithm.Finally,the preprocessed grounding switch mechanical data is input into the structural support vector machine,and the optimal solution of the structural support vector machine is obtained through the plane cutting algorithm to complete the position detection of the grounding switch machinery.The experimental results show that the proposed method has high position detection efficiency,high accuracy,strong anti noise ability and good position detection effect.

关 键 词：形态学滤波器 聚类中心 结构支持向量机 重叠函数 最大分类间隔 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置] TN911.7[自动化与计算机技术—控制科学与工程]