ahmed-maher77 / wind-turbine-power-prediction-app-using-machine-learning

"Wind Power Predictor" is a machine learning project that forecasts turbine output using real-time data from Turkish wind farms. Its web app interface offers convenient access to predictions, enabling informed decisions for maximizing energy production and advancing renewable energy usage.

Wind Turbine Power Predictor

This project involves the development of a machine learning model deployed on a web application to predict the power output of a wind turbine. The model utilizes real-time environmental and operational data sourced from a wind turbine's Supervisory Control and Data Acquisition (SCADA) system in Turkey. Features such as wind speed, wind direction, and theoretical power curve are analyzed to optimize wind turbine performance and enhance energy production efficiency. By integrating the model into a web application, users can access predictions and insights remotely, facilitating informed decision-making and promoting the sustainable utilization of renewable energy resources.

Used Technologies: Python - Flask Framework - JSON - JavaScript - HTML/HTML5 - CSS/CSS3 - PWA - Python Libraries (pandas - numpy - matplotlib - seaborn - sklearn - catboost - xgboost) - ML Algorithms (GradientBoostingRegressor - SVR - RandomForestRegressor - LinearRegression - ExtraTreesRegressor - AdaBoostRegressor - DecisionTreeRegressor - XGBRegressor - XGBRFRegressor - CatBoostRegressor)

Demo (Live Preview): https://mlwindturbine.pythonanywhere.com/

Jupyter Notebook (ML Code): https://www.kaggle.com/code/ahmedmaheralgohary/wind-turbine-eda-and-modeling

Key Features:

• Machine Learning Model: Utilizes advanced algorithms to predict wind turbine power output.
• Real-Time Data Analysis: Incorporates live data from SCADA systems for accurate predictions.
• Web Application: serves as the primary platform for accessing predictions and insights generated by our machine learning model. Users can conveniently access these resources remotely, empowering informed decision-making regarding wind turbine operations.
• Enhanced Decision-Making: Empowers users with actionable insights to optimize turbine operations.
• Promotes Sustainability: Encourages the eco-friendly use of renewable energy resources.

Web Application Features:

• Responsive Design: Seamlessly accessible across various devices.
• High Performance: Utilizing optimal code structure and lazy loading for images to ensure lightning-fast speed and responsiveness.
• Accessibility: Our platform caters to users with special needs, ensuring compatibility with screen readers and enabling access for individuals with disabilities.
• High SEO: Implementing meta tags, titles for images, alternative texts, and semantic elements to enhance search engine visibility.
• Clean Code and Best Practices: Prioritizing code clarity, organization, and utilization of modern technologies to ensure browser compatibility and incorporate the latest features and techniques.
• Simple Animations: Enhancing user experience with subtle yet effective animations.
• High-Quality UX: Prioritizing user experience with high contrast colors, clear fonts, easy navigation, and smooth interactions.
• PWA: Our web application is installable on various devices, offering the convenience of a mobile app across multiple platforms.
• Dark/Light Mode: Enhance user experience with the option to switch between dark and light modes, providing flexibility and reducing eye strain, while accessing predictions and insights for informed decision-making regarding wind turbine operations.

Contributors:

• Ahmed Maher Algohary (Me)   LinkedIn
• Ahmed Waheed   LinkedIn
• Youssif Qzamel   LinkedIn
• Mohamed Ehab   LinkedIn
• Mina Farid   LinkedIn
• Mohamed Khedr   LinkedIn
• Peter Nabil   LinkedIn
• Mahmoud Elspaiy