Publications

Abstract : During COVID-19 impact especially on energy markets, reliable electricity pricing has now become unpredictable and it becomes a challenging task to get prepared for the future price forecasting. The pandemic has mostly affected energy markets and efficient operation of the restructured electricity market effectively all over the world. In this work, the analysis of electricity price and forecasting is carried out on the wholesale market of United States namely MISO electricity market. Due to uncertainty of demand occurring during the pandemic period, the market price data is analyzed. And, using statistical learning and deep learning method day ahead price is forecasted which would prepare the electricity market to operate in an efficient manner to face such pandemics in the future. In this study, three methods are proposed namely Auto Regressive Integrated Moving Average (ARIMA), decision-tree-based ensemble Machine Learning algorithm namely Extreme Gradient Boosting (XGboost) and Recurrent Neural Network (RNN) for forecasting the electricity price. Depending upon the electricity price data attributes, the electricity price of MISO electricity market is predicted and forecasted. The performance of the methods to predict and forecast the electricity price is compared based on the processing speed and error.