https://github.com/sharvesh1401/battsense

BattSense is a machine learning project focused on predicting the State of Health (SOH) of lithium-ion batteries using operational parameters such as voltage, current, temperature, and capacity. The model enables accurate, data-driven diagnostics for battery performance monitoring in electric vehicles and portable devices.
https://github.com/sharvesh1401/battsense

battery-diagnostics battery-health battery-health-prediction battery-soh data-analysis electric-vehicles energy-storage machine-learning predictive-maintenance python regression scikit-learn

Last synced: about 1 month ago
JSON representation

Host: GitHub
URL: https://github.com/sharvesh1401/battsense
Owner: sharvesh1401
Created: 2025-06-18T17:43:01.000Z (12 months ago)
Default Branch: main
Last Pushed: 2025-07-27T19:41:00.000Z (10 months ago)
Last Synced: 2025-07-27T21:35:27.068Z (10 months ago)
Topics: battery-diagnostics, battery-health, battery-health-prediction, battery-soh, data-analysis, electric-vehicles, energy-storage, machine-learning, predictive-maintenance, python, regression, scikit-learn
Language: TypeScript
Homepage: https://battsense.netlify.app
Size: 250 KB
Stars: 0
Watchers: 0
Forks: 0
Open Issues: 0
Metadata Files:
- Readme: README.md

Awesome Lists containing this project

README

# ⚡ BattSense – Battery Health Prediction Web App

[![Live App](https://img.shields.io/badge/Live%20App-battsense.netlify.app-%237B3FE4?style=flat-square&logo=netlify&logoColor=white)](https://battsense.netlify.app)
[![Follow on Twitter](https://img.shields.io/badge/Twitter-%40Sharvesh_14-1DA1F2?style=flat-square&logo=twitter&logoColor=white)](https://x.com/Sharvesh_14)

Typing animation

> “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

![skills](https://skillicons.dev/icons?i=react,tailwind,typescript,vite,python,netlify,git&theme=light)

**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

![SOH Prediction Chart](./image_2025-06-19_075905942.png)

> 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

```bash
# 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](https://sharveshfolio.netlify.app)
🐦 [@Sharvesh_14](https://x.com/Sharvesh_14)

---

ecosyste.ms

Data

Tools

Indexes

Applications

Experiments

Awesome

https://github.com/sharvesh1401/battsense

Awesome Lists containing this project

README