Research and Analysis Applying Neural Network on Low Voltage Contract-Consumers

• Main Authors: CHENG JEN MING, 鄭哲民
• Other Authors: Yung-Nan Hu
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/59861300144086181072

碩士 === 大葉大學 === 電機工程學系碩士在職專班 === 93 === Due to the increasing public awareness，consumers reject power-equipments to be set near their houses. Building constructor and electrician also try to avoid setting power equipment near consumers’ houses at all cost. Contracted-consumers have a specific power requirement，so Engineers usually design and lay equipments in shopping and schooling districts. In order to reject power-equipments，some consumers usually underestimate measure of power-depletion. This will result in some problems, i.e., full load or overload operation, difficulty in equipment maintenance, power quality degradation and life shortage of equipment. The purpose of this thesis is solving aiming at such problem. Two hundred fifty three contracted-consumers, were studied.To obtain influence factor by using regression analysis for consume’s data ，then applying Neural Network to calculate suitable weight . Fifty consumer data were used for evaluating the performance of the proposed method by comparing the estimated and real values. A better way of grouping and network calculation can be done based on the calculated error. This can also act as reference for contracted- consumers .