SEASONAL AND CYCLIC FORECASTING FOR HYDRO- ELECTRIC POWER GENERATING STATION

Abstract

Effective forecasting is an inevitable tool for managers and administrators who are always occupied with strategic decision making under uncertainty. Electric power load forecasting is a vital process in the planning of electricity industry and the operation of electric power systems. It facilitates electric power-generation strategies, by pre-informing power providers to take appropriate mitigating actions to minimize risks and manage demand. In this paper, monthly electric power load data collected from hydro-electric power generating station, Kainji, for seven years (2008 – 2014), were analyzed and used to forecast future power values. The patterns of distribution of previous power generation data over a period of time (i.e., in form of Time series), which include trend, seasonality, cyclicality, and randomness, were identified using Decomposition approach. The choice of an appropriate forecasting method was then determined, future values were projected and power forecasts were made for the next few periods. For reasons of clarity and uniformity, points of significance and mathematical procedures were discussed and shown in the context of a business case typical of most industries