Forecasting of Hydropower Generation of India using Autoregressive Integrated Moving Average Model

Abstract

Hydropower is a prominent source of energy, contributing for more than 60% of global renewable electricity. It plays a key role in green power generation and has a fundamental influence on power market prices. As a result, precisely predicting the yearly hydro-power generation is need of the hour for the present situations. The present study focussed on predicting the hydropower generation of India through Autoregressive Integrated Moving Average (ARIMA) Model on the basis of the historical data from the year 1971-72 to 2019-20. Significant spikes in the plots of autocorrelation function (ACF) and partial autocorrelation function (PACF) of the hydropower generation data were used to identify the autoregressive (p) and moving average (q) parameters. ARIMA (1, 1, 1) with drift model was found suitable to hydropower generation for forecasting of energy demand for the country needs. Prediction of hydropower generation is done for the next decade using best fitted ARIMA model with lowest AIC and BIC values. The model helps to monitor and understand the nonlinear behaviour of India’s hydropower generation as well as energy markets in India.