基于深度置信网络的电能质量扰动检测与分类

Power Quality Disturbance Detection and Classification Based on Deep Belief Network

为准确识别各类电能质量扰动,提出一种新型的、基于深度置信网络(deep belief network, DBN)的电能质量(power quality, PQ)扰动检测和分类方法。该方法为纯数据驱动方法,通过使用DBN对数据样本进行深度学习,从而形成稳定模型用于检测与分类。为了获得足够的样本进行训练,搭建1个周期内的PQ扰动数学模型,进行数据采集;然后对DBN的结构及参数进行设计和选取。为验证该方法的有效性,使用训练好的DBN对常见的PQ扰动信号进行检测和分类,并与现有的常规检测分类方法进行比较。对比仿真结果表明,与现有的检测分类方法相比,该方法具有更高的精度和较强的鲁棒性。

To accurately identify various power quality(PQ)disturbance,this paper proposes a new type of PQ disturbance detection and classification method based on deep belief network(DBN).The method is a pure data-driven method,and the data samples are deeply learned by using DBN to form a stable model for detection and classification.In order to obtain enough samples for training,this paper builds a mathematical model of PQ disturbance in a cycle,and performs data acquisition,and then optimizes the number and size of hidden layers of DBN.In order to verify the effectiveness of the proposed method,the paper uses the trained DBN network to detect and classify common PQ disturbance signals and compares it with existing conventional detection classification methods.The comparative simulation results show that the proposed method has higher precision and stronger robustness than the existing detection classification methods.