Multi-Agent System Based Distributed Voltage Control in Distribution Systems

摘要

Distribution System is a standout among the most complex entities of the electric power grid. Moreover, voltage quality sustainability till customer premises, with the introduction of Distributed Generation (DG), is one of the most frenzied control areas. Previously, SCADA in cohesion with Wide Area Measurement Systems (WAMS) was a dependable control strategy, yet as the ever growing and complex distribution system is advancing towards the Smart Grids, control strategies are becoming more and more distributed in spite of the centralized one.A detailed literature review of the voltage control methods ranging from the centralized one to the fully distributed agent based control is conducted. In the light of the previous researches, a distributed voltage control based on Multi-Agent System is proposed, as the agents based control strategies, are becoming well known day by day, due to its autonomous control and decision making capacity. To make the proposed algorithm fully distributed, token transversal through the network and agents communication to remove voltage violation over least correspondence and measurements of the system, are utilized. Following instant voltage control at the load nodes, a penalty function is employed to keep the voltage value curve throughout the network as close as possible to the nominal, with minimum network losses and minimum voltage damage.The authentication of the devised control algorithm is acknowledged by utilizing a Greenfield distribution Network, which is based on the realistic loading data. Agents and the controlling logic are codded in Matlab ® programming software. A sensitivity analysis is performed based on DG penetration to have the complete overview of the proposed methodology. The principle objective of the technique is to keep the voltage value within the standard limit of ±10% of the nominal, at all load nodes while instantly utilizing voltage control entities like DGs, Static VAR Compensator (SVCs) and On-Load Tap Changer (OLTC). In addition, the optimization of network losses and voltage level close to nominal is to be accomplished by the penalty function implementation.