Logo Lanfrica
  • Home
  • Atlas
  • Insights
  • Docs
  • Sign in

© 2026 Lanfrica. All rights reserved. All copyrights of the resources shown on the Lanfrica website belong to the original copyright holders, unless explicitly stated otherwise.

RahmaDawy/Solar-Power-Prediction-Egypt

Domain:

environment and energy

Record type:

project
Creator:
Rah
Host:
Machine Learning analysis of solar panel output in Aswan. # Solar Photovoltaic Power Forecasting in Aswan, Egypt ### Advanced Predictive Modeling using Meteorological Data --- ## 🚀 Project Overview This repository contains a comprehensive Machine Learning framework designed to predict Solar Photovoltaic (PV) power output in the Aswan region of Egypt. The project addresses the challenge of predicting renewable energy generation using only low-cost, standard meteorological variables (**Temperature, Humidity, Wind Speed, and Pressure**) rather than expensive irradiance sensors. The study implements both **Classification** (predicting output levels) and **Regression** (predicting continuous power values), utilizing a robust pipeline of dimensionality reduction and non-linear modeling. --- ## 👤 Author & Supervision * **Presenter:** Rahma Asem Dawi * **Program:** Artificial Intelligence and Data Science (AID) * **Institution:** E-JUST (Egypt-Japan University of Science and Technology) * **Supervisors:** Dr. Ahmed Anter and Eng. Sama AlQasaby * **Date:** December 2025 --- ## 🏗️ Technical Pipeline & Methodology ### 1. Data Engineering & Statistical Validation * **Data Cleaning:** The initial dataset of 421 observations was refined to **398 unique chronological records** by removing redundant data blocks (specifically duplicated April 2022 entries). * **Preprocessing:** Automated handling of missing values, outlier removal (resulting in **312 usable samples**), and target binning for categorical classification (Low, Medium, and High generation). * **Statistical Profiling:** Rigorous analysis of **Skewness** and **Kurtosis** to identify and handle data distribution irregularities. * **Hypothesis Testing:** Utilization of **ANOVA** and **T-Tests** ($p < 0.05$) to statistically prove the influence of weather features, particularly Temperature, on solar power variance. ### 2. Feature Reduction (Dimensionality Reduction) To optimize performance and address multicollinearity, three distinct reduction techniques were implemente …

Visit

github.com