With the recent growth in urban and rural sectors by leveraging smart enabling technologies, the consumption of electricity with daily needs also increasing. The demand and supply are expected to be managed proactively to ensure that future power requirements are met well in advance, avoiding any sudden or critical shortages. The power stations are generating powers to meet the demands, however, to meet the future demands, the generation also to be increased or new units to be established before the saturation comes. The study reveals the generation prediction over the years by considering the generation trends based on live history data of electricity generation. The study also analyses the growth in generation and demand gap, that is likely to provide a suggestion to the governing authorities to plan for future establishment of more units or power generation plants as well. A combination of regression techniques such as Linear Regression and Lasso Regression alongside Recurrent Neural Networks (RNN), specifically Long Short-Term Memory (LSTM) models, are employed to predict future power generation patterns accurately on live data obtained from authentic government agencies.