In recent years, the idea of a smart grid is being projected in real life. In various countries which constitutes the main idea of deregulation which comes with the conversion of the consumer as a prosumer which affects electricity prices, demand, and power required. In this article Neural Basis Expansion Analysis for Interpretable Time Series Forecasting (N-BEATS) algorithm is used for forecasting of uni-variate data with data-preprocessing stages and multivariate with feature engineering stage, also various other benchmark methods are implemented using python for more flexibility to know robustness of proposed method on the particular case study for power system which is Ontario demand, hourly electricity price, wind speed in Ontario to have precise forecasting which helps in various tasks like demand response to conventional source management especially by detecting sharp spikes in data.