Led comprehensive geospatial, temporal, seasonal, and correlational analyses to inform advanced predictive modelling for solar energy forecasting.

Integrated wind speed and direction as external factors into ARIMA model, resulting in a notable increase in R squared value to 93.6 and a significant decrease in Mean Absolute Error (MAE) to 8.840, demonstrating enhanced precision and predictive power.

Orchestrated development of an advanced LSTM model incorporating feature engineering techniques for solar energy forecasting, achieving an outstanding R-squared value of 98.94 and an impressively low MAE of 3.841, showcasing exceptional accuracy in navigating temporal data complexities.