Automatic and Intelligent Power Quality Disturbances Monitoring Based on Multi-Agent Systems

Power quality monitoring is the first step in the identification of power quality disturbances and reducing them in order to improve the performance of the power system. The aim of this paper is to propose the architecture of a new intelligent strategy for online and offline power quality monitoring system based on multi-agent systems. In this study، a multi-agent system for solving some problems in power quality monitoring، including computational complexity، low accuracy، change in the data pattern، non adaptive structure of detection system to changing conditions is proposed. In the proposed strategy، the agent characteristics، such as automatic and dynamic performance، intelligent، learning، reasoning ability، objectively and interoperability of agents are used. This paper is presented in two stages. In the one stage، to indicate problems in power quality monitoring، different methods of feature extraction، feature selection and classification for automatic recognition of power quality disturbances have been analyzed. Optimal selection of input feature vector of distinguish system is applied using different methods of data mining. Also، three well-known classifiers are considered. In another stage of the paper، to solve some challenges، the design of investigated structures in the form of a multi-agent system is expressed. The results of the experiments in this paper demonstrate the superiority of agents and multi-agent systems for online and offline power quality monitoring.