wilberkamaa/fault-prediction

Predictive Analytics for Hybrid Solar Systems in Kenya # Energy System Fault Prediction A comprehensive solution for fault prediction in hybrid energy systems, combining synthetic data generation, machine learning models, and an interactive dashboard. ## 📋 Overview This project provides tools for: - **Synthetic Data Generation**: Create realistic time-series data for hybrid energy systems - **Fault Detection**: Train and evaluate ML models (Random Forest and XGBoost) for fault prediction - **Interactive Dashboard**: Visualize system data and predictions through a Streamlit interface - Available: fault-prediction.streamlit.… The system is designed to detect and predict faults in a hybrid energy system with components including: - 1500 kW Solar PV System - 1 MVA Diesel Generator - 3 MWh Battery Storage - 25 kV Grid Connection ## 🔧 Installation ### Prerequisites - Python 3.8 or higher - pip (Python package installer) ### Setup 1. Clone the repository: ```bash git clone github.com cd fault-prediction ``` 2. Create and activate a virtual environment: ```bash python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate ``` 3. Install dependencies: ```bash pip install -r requirements.txt ``` ## Usage ### Data Generation Generate synthetic data for a hybrid energy system: ```python from src.data_generator import HybridSystemDataGenerator # Initialize generator generator = HybridSystemDataGenerator(seed=42) # Generate 1 year of data df = generator.generate_dataset( start_date='2023-01-01', periods_years=1, output_file='data/hybrid_system_data.parquet' ) ``` ### Running the Dashboard Launch the interactive Streamlit dashboard: ```bash streamlit run streamlit_app.py # OR streamlit run prediction_app.py ``` The dashboard provides: - System Overview visualization - Fault Analysis and prediction - Feature Importance analysis - Performance Metrics evaluation ## 📊 Dashboard Features ### 1. System Overview - Power distribution visualizatio …