智能螺帽接触式测温传感器在隔离开关触头监测中的应用

为了提升隔离开关安全性，减少漏监测现象，研究智能螺帽接触式测温传感器在隔离开关触头监测中的应用。依据智能螺帽接触式测温传感器的测温与温度补偿原理，在隔离开关触头位置合理布置测温传感器，获取隔离开关触头实时温度后，利用测温传感器自身无线发射模块，将温度数据传输至基于卷积神经网络的温度特征学习框架内，利用该特征学习框架获取隔离开关触头温度特征数据，以该温度特征数据作为温度特征子样本集建立孤立森林，并计算森林内树的深度数值与平均深度数值，依据特征阈值判断树的平均深度是否符合条件，实现隔离开关触头温度监测。实例测试结果表明：该方法数据损失数值小，具备较强的温度数据特征提取能力，且不受电磁波干扰，监测异常隔离开关触头温度仅需0.4s，可有效提升隔离开关安全性。

Application of Intelligent Nut Contact Temperature Sensor in the Contact Monitoring of Isolating Switch

In order to improve the safety of disconnector and reduce leakage monitoring, the application of intelligent nut contact temperature sensor in disconnector contact monitoring is studied. According to the principle of temperature measurement and temperature compensation of the intelligent nut contact temperature sensor. the temperature sensor is reasonably arranged at the contact position of the disconnector. After obtaining the real-time temperature of the disconnector contact, the temperature data is transmitted to the temperature feature learning framework based on the convolution neural network by using the wireless transmission module of the temperature sensor, The temperature characteristic data of disconnector contact is obtained by using the feature learning framework, and the temperature characteristic data is used as the temperature characteristic sub sample set to establish the isolated forest, and calculate the depth value and average depth value of the tree in the forest, and judge whether the average depth of the tree meets the conditions according to the feature threshold, so as to realize the temperature monitoring of disconnector contact. The example test results show that this method has small data loss value, strong temperature data feature extraction ability, and is not disturbed by electromagnetic wave. It only takes 0.4s to monitor the contact temperature of abnormal disconnector, which can effectively improve the safety of disconnector.