Solar power has become an increasingly popular alternative to conventional electricity sources due to its cleanliness, efficiency, and sustainability. Nevertheless, determining whether a residence is powered by solar energy or the grid can pose challenges. This paper aims to forecast the amount of solar energy produced by employing sensors that measure both AC and DC voltages generated by the solar panels while monitoring fluctuations. The collected data will be analyzed using machine learning algorithms to predict solar energy production and ascertain if the residence relies on solar energy or the grid. In the event of any discrepancies or fluctuations, the system will notify the user via alert messages sent to their mobile device using GSM technology. Additionally, LED lights will signal energy transitions. The proposed system leverages machine learning techniques to estimate the solar energy input by implementing various regression algorithms, including linear regression, logistic regression, lasso regression, and random forest regression. Among these, the random forest regression algorithm has been employed to achieve a remarkable accuracy of 97.27% in predicting solar energy production. The forecasted data is then transmitted to an IoT microprocessor, which publishes the information on the Adafruit IoT cloud. This data can subsequently be utilized to determine whether the residence is powered by solar energy or the grid.