Reactive Power and Voltage Control in a Distorted Distribution System Considering Distributed Generation

• Main Authors: Jhen-Rong Huang, 黃振榮
• Other Authors: Ruey-Hsun Liang
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/36778513936900688105

碩士 === 國立雲林科技大學 === 電機工程系碩士班 === 99 === Reactive power and voltage control in a distribution system in a daily system operation is very important. A proper control for the reactive power and voltage will improve the voltage profile, reduce system losses, and increase system efficiency. The reactive power and voltage are efficiently controlled that will improve the voltage quality and decrease power generation cost. First, this thesis presents a non-revisiting memetic algorithm to solve the dispatch problem of under load tap changer (ULTC) and capacitors in a harmonic distorted distribution system. The main purpose is to find an optimum dispatch of ULTC and capacitors status, to reduce real power losses, and to inhibit bus voltage total harmonic distortion (THD). To demonstrate the effectiveness of the proposed method, the reactive power and voltage control problem is performed in a distorted distribution system. And then a non-revisiting enhanced particle swarm optimization is proposed for solving the problem of reactive power and voltage control in a distorted distribution system considering distributed generation. The purpose of the reactive power and voltage control problem is to find an optimum dispatching schedule of tap position for the main transformer ULTC, on/off status for the switched capacitors, and active power and power factor for the distributed generators in a day. The constraints must be satisfied to minimize the purchase cost for buying electricity from the grid, operating cost of distributed generators and real power loss cost on feeders. In order to reach the objective satisfying some constraints, a non-revisiting enhanced particle swarm optimization is proposed to solve the reactive power and voltage control problem. The proposed method is to combine non-revisiting scheme and local search technique into a particle swarm optimization, so that the global optimum solution can be efficiently and quickly achieved. To demonstrate the effectiveness of the proposed method, the reactive power and voltage control problem is performed in a distorted distribution system with distributed generation.