Many industies have equipments sensitive to bad power quality that affects their production and product quality. Therefore, it is important to automatically monitor the quality of power with minimum human intervention. It is possible to analyze and interprete raw data from the industrial equipments to useful information with the help of signal processing and artificial intelligence system. This paper presents the automatic classification of power quality events using Extreme Learning Machine (ELM) in combination with optimization techniques. S transform is used for extraction of useful features of the disturbance signal. The features are used to train the ELM for classifying PQ events. Further the parameters of ELM are tuned through Grey Wolf Optimization (GWO) approach to improve the classification accuracy. Seventeen different categories of PQ events are used for the classification purpose. The efficiency of GWO-ELM is compared with other widely used classifiers such as K-Nearest Neighbour (KNN), Support Vector Machine (SVM) and ELM. The simulation results reveal that the proposed approach can accurately detect and classify the PQ events.

许多行业拥有对不良电能质量敏感的设备，这些设备会影响其生产和产品质量。因此，以最少的人工干预自动监视电源质量非常重要。借助信号处理和人工智能系统，可以将工业设备中的原始数据分析和解释为有用的信息。本文介绍了使用极限学习机（ELM）结合优化技术对电能质量事件进行自动分类的方法。S变换用于提取干扰信号的有用特征。这些功能用于训练ELM以对PQ事件进行分类。此外，通过灰狼优化（GWO）方法对ELM的参数进行了调整，以提高分类的准确性。十七种不同类别的PQ事件用于分类目的。将GWO-ELM的效率与其他广泛使用的分类器（例如K最近邻（KNN），支持向量机（SVM）和ELM）进行了比较。仿真结果表明，该方法可以准确地对PQ事件进行检测和分类。