MAGISTRA
Vol 19, No 61 (2007): Magistra Edisi Juni

PERAMALAN BEBAN LISTRIK JARAK PENDEK DENGAN MENGGUNAKAN JARINGAN SYARAF TIRUAN DI P3B PT. PLN REGION III JAWA TENGAH DAN DIY

Purnomo, Harry (Unknown)



Article Info

Publish Date
01 Jun 2007

Abstract

Short-term load forcasting need to planning activity of electrical generating and demand of power system, to schedull and controlling power system or running generatingresources. Short-term load forecasting in P3B PT. PLN Region III Central Java and DIY have load coofesien mothode. This method is not practically and need a long time to forecasting. Artificial neural-network hort-term load forecasting (ANNSTLF) with backpropagation algorithm used a new methode for short-term load forecasting. The load were divided into 10 design : 6 load work day, 1 load for Sunday or Holiday, 3 load for special day (new year, Christmas and Idul Fitri). A load not linear. Input for Artificial neural-network is historical load for 2 day ago and yesterday (48 hour), and the output is load forecasting, load of hour by hour for 24 hour tomorrow. Artificial neural-network short-term load forecasting (ANNSTLF) with backpropagantion of algorithm have been worked successful by simulation and forecast a load with a relative less error.   Keyword : Artificial neural-network, backpropagation, electric load forecasting.

Copyrights © 2007