AhmedElmasry27/solar-energy-prediction-aswan

Machine Learning project predicting solar output using weather data from Aswan, Egypt. # Solar Energy Production Prediction (Aswan, Egypt) ## Project Overview This project applies Machine Learning to predict solar energy output based on meteorological data collected from **Aswan, Egypt** (Benban Solar Park region). The goal is to classify daily solar production into **Low, Medium, or High** categories to assist in grid stability and energy management. ## Dataset The dataset includes daily weather observations: - **Features:** Average Temperature, Humidity, Wind Speed, Pressure, Dew Point. - **Target:** Solar PV Output (Classified into Low, Medium, High). - **Source:** [Mention source if you have it, e.g., NASA Power or Local Station]. ## Technologies Used - **Python** (Pandas, NumPy, Matplotlib, Seaborn) - **Scikit-Learn** (Decision Tree, KNN, LDA, Logistic Regression) - **Data Viz:** Heatmaps, 3D Scatter Plots, ROC Curves. ## Key Results - **Best Model:** Decision Tree Classifier (**65% Accuracy**). - **Key Insight:** The relationship between Temperature and Solar Output is **non-linear**. Simple linear models (Logistic Regression) failed (~30% acc), while rule-based models (Trees) succeeded. - **Top Features:** Temperature and Humidity were the strongest predictors of solar output. ## How to Run 1. Clone the repo: ```bash git clone github.com