Short-term load prediction model for electrical power distribution network

Abstract

Electrical Load prediction model for power distribution network becomes necessary as a result of fast increase in the demand for electricity for residential and industrial usage. This work developed a short-term model for an 11-KV feeder that supplied a particular area. The model was based on the multiple linear regression principle that used the past data of the feeder line. The model was used to predict the consumers’ load demand for a period of 1 week. The result of the regression analysis gave a Mean Absolute Percentage Error (MAPE) of forecast of 0.9% for the 11-KV feeder. Conclusively, the accuracy of the developed load forecast model was 99.1%.