基于容性能量转移变化的高压真空断路器故障智能诊断

Intelligent Fault Diagnosis of High Voltage Vacuum Circuit Breaker Based on Capacitive Energy Transfer Change

高压断路器作为电力系统中最重要的承载、保护与控制单元之一,其绝缘与开断依赖SF6气体。由于常规高压真空断路器存在自然电流零点和电能谷底干扰,其故障检测的核心过程为:电流零区弧隙之间SF6气体介质强度变化引起的信号瞬时启动。为了减低故障漏检情况,结合神经网络+混合布谷鸟算法,设计一种高精度的高压真空断路器故障智能诊断方法。首先,分析故障发生后容性能量转移过程,归纳故障发生后容性能量转移的空间特征和时间特征。其次,构建基于PNN神经网络的高压真空断路器机械故障诊断模型,将容性能量转移特征向量作为训练样本输入到模型中展开训练,完成故障智能诊断。最后,利用混合布谷鸟算法对模型中的参数展开寻优,以此得到高精度诊断结果。实验结果表明:所提方法在高压真空断路器诊断过程中的精度和稳定性具有较高优势。

As one of the most important load-bearing,protection,and control units in the power system,highvoltage circuit breakers rely on SF6 gas for insulation and disconnection.Due to the interference of natural current zero and electrical energy valley in conventional high-voltage vacuum circuit breakers,the core process of fault detection is the instantaneous start of signals caused by changes in the strength of SF6 gas medium between the arc gaps in the current zero zone.In order to reduce the occurrence of missed faults,a high-precision intelligent fault diagnosis method for high-voltage vacuum circuit breakers is designed by combining neural networks and the hybrid cuckoo algorithm.Firstly,the process of capacitive energy transfer after a fault occurs is analyzed,and the spatial and temporal characteristics of capacitive energy transfer after a fault occurs is summarized.Secondly,a high voltage vacuum circuit breaker mechanical fault diagnosis model based on PNN neural network is constructed,and the capacitive energy transfer feature vector is input into the model as training samples for training,completing intelligent fault diagnosis.Finally,the hybrid cuckoo algorithm is used to optimize the parameters in the model,in order to obtain high-precision diagnostic results.The experimental results show that the proposed method has high advantages in accuracy and stability in the diagnosis process of high-voltage vacuum circuit breakers.