Solar photovoltaic (PV) systems are gaining popularity as a sustainable and renewable energy source. Accurate prediction of PV output power is essential for efficient system design, grid integration, and energy management. Existing work in this domain has shown limitations in accurately capturing the complex relationships between weather conditions and PV output power. To address these flaws, we propose a machine learning model based on linear regression, support vector regression (SVR), random forest, gradient based, and feedforward neural network (FNN). The proposed model’s accuracy and reliability can aid in optimizing the design and operation of solar PV systems in Visakhapatnam. Among all models, random forest (mean squared error (MSE): 0.0365, mean absolute error (MAE): 0.0912, R2: 0.9997) and gradient boosting (MSE: 0.0541, MAE: 0.1263, R2: 0.9996) provide accurate predictions with low errors and high R-squared values. Decision tree also performs well (MSE: 0.0715, MAE: 0.1226, R2: 0.9995). However, support vector regression (SVR) exhibits poor performance. Overall, random forest and gradient boosting are recommended for accurate solar PV power prediction. The dataset used in this study consists of historical weather data collected from local meteorological station. The data include solar irradiance, temperature, wind speed, humidity, and corresponding PV system output power measurements. The dataset was preprocessed and divided into training and testing sets to evaluate the model’s performance. The findings have implications for energy management, load forecasting, and grid integration, promoting the efficient utilization of solar energy resources.