ICTACT Journal on Soft Computing

S. Saravanan ¹, S. Kannan ², C. Thangaraj ³

Affiliations
1 Department of Electrical and Electronics Engineering, Kalasalingam University, IN
2 Department of Electrical and Electronics Engineering, Ramco Institute of Technology, IN
3 Vignan University, IN

Abstract

This study aims to provide an accurate and realistic prediction model for electricity demand using population, imports, exports, per capita Gross Domestic Product (GDP) and per capita Gross National Income (GNI) data for India. Four different models were used for different combinations of the above five input variables and the effect of input variables on the estimation of electricity demand has been demonstrated. In order to train the network 29 years data and to test the network 9 years data have been used. The future electricity demand for a period of 8 years from 2013 to 2020 has been predicted. The performance of the ANFIS technique is proved to be better than Multiple Linear Regression (MLR) and Artificial Neural Networks (ANN).

Keywords

ANFIS, ANN, Exports, GDP, GNI, Imports, Load Forecasting, MLR.

Full Text

     

S. Saravanan ¹, S. Kannan ¹, C. Thangaraj ²

Affiliations
1 Department of Electrical and Electronics Engineering, Kalasalingam University, IN
2 Anna University of Technology, Chennai, IN

Abstract

Power System planning starts with Electric load (demand) forecasting. Accurate electricity load forecasting is one of the most important challenges in managing supply and demand of the electricity, since the electricity demand is volatile in nature; it cannot be stored and has to be consumed instantly. The aim of this study deals with electricity consumption in India, to forecast future projection of demand for a period of 19 years from 2012 to 2030. The eleven input variables used are Amount of CO₂ emission, Population, Per capita GDP, Per capita gross national income, Gross Domestic savings, Industry, Consumer price index, Wholesale price index, Imports, Exports and Per capita power consumption. A new methodology based on Artificial Neural Networks (ANNs) using principal components is also used. Data of 29 years used for training and data of 10 years used for testing the ANNs. Comparison made with multiple linear regression (based on original data and the principal components) and ANNs with original data as input variables. The results show that the use of ANNs with principal components (PC) is more effective.

Keywords

Artificial Neural Networks, Electricity Load Forecasting, Regression Analysis, Principal Components.

Full Text