A Hybrid Arima and Discrete Wavelet Transform Model for Predicting the Electricity Consumption of Punjab

Abstract

The significant increase in the world population increases the demand for energy which seems to be alarming for the electricity production boards in the existing time. In the last decade, there are various engineering, simulation tools, and artificial intelligence-based methods such as Support Vector Machine and Artificial Neural Network proposed in the literature to forecast the optimal electricity demand. But these models seldom to work with the linear data. In this paper, a reliable prediction model using the linear time series data of the previous years from January 2013 to December 2017 has been presented to forecast the electricity consumption in Punjab, India. Initially, Discrete Wavelet Transform (DWT) analysis presented to extract the upper and lower limit of the previous year dataand then Auto-Regressive Integrated Moving Average (ARIMA) model has been applied to extract the forecast values. The experimental results compared the original and predicted value using the proposed model to evaluate the effectiveness of the proposed approach. The results show that the difference between the original and proposed modelis only 9% while that of ARIMA only it is 11%. Thus, the proposed model using ARIMA and DWT provides effective results in predicting the forecast value.