基于传感器融合技术的高压开关柜缺陷识别

Defect recognition of high-voltage switchgear based on sensor fusion technology

高压开关柜是变电站的重要设备之一,对其进行绝缘缺陷检测能够保证系统供电的可靠性。基于此,首先制作了四种开关柜典型模型,并搭建了试验平台;其次采用传感器融合技术同时获取暂态地电压信号和脉冲电流信号;再次采用粒子群优化的多维支持向量机算法进行缺陷识别;最后分别针对TEV信号、脉冲电流传感器信号以及组合信号的缺陷识别,与神经网络识别算法进行对比,结果表明,基于传感融合技术的开关柜缺陷识别的准确率达到96.7%,证明缺陷识别算法的准确性。

High-voltage switchgear is one of the important equipments in substations,and its insulation defect detection can ensure the reliability of system power supply.Based on this,firstly,four typical models of switchgear are made,and a test platform is built.Secondly,the sensor fusion technology is used to obtain the transient earth voltage(TEV)signal and the pulse current signal at the same time,and the particle swarm optimized multi-dimensional support vector machine algorithm is again used for defect recognition.Finally,the defect recognition of TEV signal,pulse current sensor signal and combined signal is compared separately,and the neural network recognition algorithm is compared at the same time.The result shows that the accurate rate of switchgear defect recognition based on sensor fusion technology reaches 96.7%,which proves the accuracy of the defect recognition algorithm.