基于ITD分解和孪生支持向量机的电能质量扰动识别方法研究

Power Quality Disturbance Identification Based on ITD Decomposition and Twin Support Vector Machine

电能质量扰动信号的识别与分类是电能质量分析、评估和治理的基础和关键。针对电能质量扰动信号种类复杂、识别速度慢且准确率低等问题,提出一种基于ITD分解和孪生支持向量机的电能质量扰动识别方法。首先,对电能质量扰动信号做ITD分解,得到一系列固有旋转分量(PRC),并通过云模型的熵和超熵筛选出有效的PRC分量,减少特征冗余;其次,计算有效的PRC分量的模糊熵和能量熵,并根据模糊熵和能量熵求得混合特征矩阵;最后,基于混合特征采用麻雀优化的孪生支持向量机对扰动信号进行分类。仿真分析结果表明,该方法能识别多种电能质量扰动信号,且提高了对单一电能质量的识别准确率,进而为电能质量的分析、评估和治理提供辅助决策,以进一步提高供电质量。

The recognition and classification of power quality disturbance signals is the basis and key of power quality analysis,evaluation and control.Aiming at the problems of complex types of power quality disturbance signals,slow recognition speed and low accuracy rate,a power quality disturbance identification method based on ITD decomposition and twin support vector machine is proposed.Firstly,the ITD decomposition of power quality disturbance signals is performed to obtain a series of intrinsic rotating components(PRC).The entropy and hyper-entropy of cloud model are used to select the effective PRC components and reduce the characteristic redundancy.Secondly,the fuzzy entropy and the energy entropy of the effective PRC component are calculated,and the mixed characteristic matrix is obtained according to the fuzzy entropy and the energy entropy.Finally,the mixed characteristic is used to classify the disturbance signals by using the sparrow optimized twin support vector machine.The simulation results show that this method can recognize many kinds of power quality disturbance signals,and improve the recognition accuracy of single power quality,thus providing assistant decision-making for power quality analysis,evaluation and control,and further facilitating the improvement of power supply quality.