This chapter aims to develop a machine learning algorithm for residential power energy consumption prediction. The suggested methodology accurately forecasts future energy usage using a dataset of hourly energy consumption and weather data from residential residences. Temperature, humidity, and time of day are just a few of the characteristics that the machine learning system takes into account to forecast how much energy residential dwellings would use. The proposed model uses various machine learning algorithms and deep learning approaches such as recurrent neural network (RNN), decision tree, and random forest, to predict energy consumption. Additionally, the paper incorporates a feature selection approach to extract the most relevant features that affect residential power consumption. The model’s performance is evaluated using metrics such as root mean squared error (RMSE) and mean squared error (MSE). To validate the proposed approach, experiments were conducted on a publicly available dataset of hourly energy consumption and weather data from residential homes. The findings demonstrate that the suggested machine learning model performs better than current approaches in terms of prediction accuracy, offering a useful tool for forecasting residential electricity usage. The proposed machine learning model has practical applications for energy management systems and utilities in predicting future residential power energy consumption. By accurately predicting energy consumption, utilities can efficiently manage their appliances and provide efficient services to customers.