基于优化VMD的高压断路器机械状态检测

为有效检测高压断路器的机械状态，提高其运行可靠性，基于优化变分模式分解(VMD)法对高压断路器分合闸过程中的振动信号进行了分析。首先利用粒子群优化算法基于整体正交系数得到了最优的VMD结果，然后对振动信号Hilbert变换的时频谱进行了合理划分，据此定义了振动信号的特征向量及相似度指标。对某40.5 kV断路器正常与典型故障下振动信号的分析结果表明，所提出的优化VMD算法的分解结果更为准确，所定义的相似度指标能有效识别断路器的典型故障。当相似度大于0.9时，断路器机械状态正常；当相似度在0.7～0.9之间，可能发生缓冲机构故障；当相似度小于0.7时，可能发生传动机构故障。

Mechanical state detection of high voltage circuit breaker based on optimized VMD algorithm

In order to effectively detect the mechanical state of HV(High Voltage) circuit breaker and improve its operation reliability, the vibration signals during the opening/closing operation of HV circuit breaker are analyzed based on the optimized VMD(Variational Mode Decomposition) algorithm. Firstly, the optimal VMD results are obtained based on the global orthogonal coefficients by particle swarm optimization algorithm. Then the time-frequency spectrums obtained from Hilbert transformation of vibration signals are reasonably divided, accordingly the feature vector and similarity degree index of vibration signals are defined. The vibration signals of a 40.5 kV circuit breaker under normal and typical fault conditions are analyzed, and the analytical results show that the decomposition results of the proposed optimized VMD algorithm are more accurate, and the defined similarity degree index can effectively identify typical faults of circuit breaker. When the similarity degree index is greater than 0.9, the circuit breaker is in normal condition; when the similarity degree index is between 0.7 and 0.9, the circuit breaker may exist buffering mechanism defects; when the similarity degree index is smaller than 0.7, the circuit breaker may exist transmission mechanism defects.