Electrical Charge of Niamey City Modelisation by Neural Network

Abstract

In order to forecast consumption, electric power generation, transmission and distribution companies need model to predict short-term demand for electric power load so that they can use their electricity infrastructure efficiently, safely and economically. The short-term forecast of electrical energy demand is the forecast of consumption over time interval ranging from one hour to few days. For optimal use of electricity grid, energy production must keep pace with demand. To this end, prediction errors can lead to risks and shortcomings in the generation and distribution of electrical load to users. This paper is part of electrical charge prediction of Niamey city. Several are being carried out in this field, but prediction techniques based on artificial neural networks have recently been developed. This work focused on two (2) neural approaches such as the multilayer Perceptron (MLP) and the non-linear autoregressive network with exogenous inputs (NARX). Several configurations of these two models have been developed and tested on actual electrical load data. We carried out the short-term forecast (hourly basis) of electrical load of Niamey city. All configurations have been implemented in MATLAB software. The statistical indicators MAPE (Mean Absolute Average Error in Percent), R2 (the correlation coefficient) and RMSE (Square Root of Mean Square Error) were used to evaluate the performance of the models. Thus, with MAPE of 5.1765%, R2 of 95.3013% and RMSE of 5.6014%, the [ABCD] configuration of NARX model converges better compared to the MLP model with MAPE of 7.1874%, R2 of 92.0622% and RMSE of 7.2199%. Where A is the data charge of the same time of the previous day, B is the charge data of the same time of the previous week, C is the charge data of same time of previous year and D is the average of last 24 charge values. So the NARX model is the most efficient and can be used for future predictions on Niamey city network.