⚡ BattSense – Battery Health Prediction Web App

“Predict battery health with real data, real models — and real-time AI assistance.”

---

🌐 Overview

BattSense is a web-based tool that predicts the State of Health (SOH) of lithium-ion batteries using machine learning.

It bridges the gap between raw sensor data and practical diagnostics through an interactive, browser-based dashboard.

Built with React + Vite + Tailwind CSS, this frontend is paired with a trained ML model and enhanced with DeepSeek AI for intelligent analysis.

---

🔍 Key Features

• 🔋 Predict SOH based on voltage, cycles, capacity, and temperature
• 🧠 Powered by a Random Forest Regressor trained on real data
• 💬 Built-in chatbot assistant using DeepSeek API
• 📊 Sample output visualization + Downloadable results
• 🧪 Configured for both web and ML experimentation

---

📦 Tech Stack

Also includes:

• 📦 PostCSS – custom styling and plugin support
• 🧪 Jest – unit testing
• 🧭 ESLint – consistent code formatting
• 🧱 Recharts – data visualization
• 🧠 DeepSeek API – conversational AI assistant
• 📁 Modular file aliasing via Vite config

---

🖼️ Sample Output

After prediction, the result is displayed and can be downloaded as a CSV for further analysis or reporting.

---

🧠 ML Model Details

• Model: Random Forest Regressor
• Dataset includes:
  • Voltage
  • Current
  • Temperature
  • Charge cycles
  • Capacity
• Target: State of Health (SOH)

Handled:

• Missing values
• Outliers
• Feature selection

Metrics Used:

• Mean Squared Error (MSE)
• Root Mean Squared Error (RMSE)
• Mean Absolute Error (MAE)
• R² Score (coming soon)

---

🚀 Getting Started

# Clone the repo
git clone https://github.com/sharvesh1401/BattSense.git
cd BattSense

# Install frontend dependencies
npm install

# Run the local dev server
npm run dev

For backend ML model usage, refer to battery_soh_predictor.py (not included in web build).

---

📁 Project Structure

├── src/                  # Frontend components & views
├── image_*.png           # Sample output graph
├── public/               # Static assets
├── index.html
├── package.json
├── vite.config.ts
├── tailwind.config.js
├── postcss.config.js
├── jest.config.cjs
└── tsconfig.*.json       # TypeScript config files

---

🛠 Improvements Planned

• Connect directly to Python backend for live predictions
• Add downloadable dataset sample
• Expand model support (XGBoost, MLP)
• Add user authentication (optional)

---

🙋‍♂️ About Me

I'm Sharvesh Selvakumar, an engineering student passionate about AI, clean energy, and responsible tech.

🔗 sharveshfolio.netlify.app
🐦 @Sharvesh_14

---

⚡ Built for smarter batteries and better energy tech.
MIT License | © 2025 Sharvesh Selvakumar