ASSESSMENT OF THE ADEQUACY OF ELECTRICAL ENERGY DEMAND FORECAST MODEL FOR THE NIGERIA POWER DISTRIBUTION SYSTEM VIA STATIONARITY TEST

Abstract

Electric energy demand forecasting model is an essential tool in the course of planning in electricity industry. Though, there has been increasing concern to fix models for various domains. The adequacy and accuracy of these models for forecasting reasonable energy generation capacity, scheduling and system management planning are paramount. Inaccurate model will give forecasting that are either underrated that will incapacitate socio-economic growth by not supply enough electrical energy for development, or overestimated leading to excess electrical energy generation without commensurate returns on investment, another form of economic jeopardy. In this paper, Assessment of the adequacy of Electrical Energy Demand Forecasting Model for the Nigeria Power Distribution System via Stationarity test was performed as crucial stage in development of time series technique of energy demand forecast model. In the stationarity investigation of data set under the null hypothesis as a test tool for the confirmation of stationarity and non-stationarity of energy demand data for processing and further analyses of energy demand in power distribution system in Nigeria. Data were collected from five 33kV feeders each with sixty-point of monthly peak Load demand for five years (2015-2019) from Ibadan Electricity distribution Company (IBEDC). R- Software was used as optimization tool for the analyses. The end result was interpreted by Critical values for Augmented Dickey-Fuller method.  Findings shows that data from three of the feeders were non-stationary they will go through data differencing to make the data suitable for further investigations as a mixture of an autoregressive integrated moving average ARIMA while two are stationary and can be authenticated for further analyses. The application of this test to further difference the datapoints that are non-stationary will lead to stationary dataset, hence, give viable model for accurate energy demand forecasting model development.