基于振动和电流信号多域特征联合的高压断路器储能状态辨识方法

为实现高压断路器的储能状态辨识，提出了一种基于伴随振动和电流信号多域特征提取的储能状态辨识方法。对振动信号进行倒频谱分析提取频域周期分量特征来体现电机周期性转动轨迹的基频和高频特性；采用相空间重构理论经关联积分法获取相空间域关联维数与Kolmogorov熵，定量评估储能过程随机运动的复杂程度；计算电机电流持续时间、起始电流值及偏度时域特征，与振动信号特征构建多域联合特征向量；通过支持向量机算法实现断路器储能状态辨识。实验结果表明，基于振动和电流信号多域联合特征的辨识算法准确率达到97%，有效地提高了断路器储能典型工况的识别准确率。

Energy storage state identification method of high-voltage circuit breaker based on multi-domain feature combination of vibration and current signals

In order to realize the energy storage state identification of high-voltage circuit breakers, an energy storage state identification method based on multi-domain feature extraction of accompanying vibration and current signals is proposed. The cepstrum analysis is performed on the vibration signal to extract the periodic component features in the frequency domain, which reflects the fundamental frequency and high-frequency characteristics of the motor periodic rotation trajectory. Then, the phase space reconstruction theory is used to obtain the correlation dimension and Kolmogorov entropy of the phase space domain by the G-P method to quantitatively evaluate the complexity of the random motion of the energy storage process. The motor current duration, initial current value and skewness time domain feature are calculated, and the multi-domain feature vector is constructed with the vibration signal feature. The energy storage state identification of circuit breakers is realized by the support vector machine algorithm. The experimental results show that the accuracy of the proposed identification algorithm based on the multi-domain joint features of vibration and current signals reaches 97%,which effectively improves the identification accuracy of the typical working conditions of the circuit breaker energy storage.