🔋 MCP-Based EV Digital Twin Agent

An AI-powered EV Digital Twin platform for battery health monitoring, predictive maintenance, fleet analytics, and intelligent decision support.

The project combines machine learning, digital twin concepts, monitoring systems, and MCP (Model Context Protocol) tools to simulate and analyze electric vehicle battery behavior.

Dashboard Preview

Battery Digital Twin

Fleet Analytics

Monitoring & Drift Detection

Related MCP server: mdfmcp

🚗 Key Features

• Battery State of Health (SOH) Prediction

• Remaining Useful Life (RUL) Estimation

• Battery Range Estimation

• What-if Scenario Simulation

• Charging Intelligence & Risk Analysis

• Driving Intelligence

• Predictive Maintenance Recommendations

• Fleet Monitoring Dashboard

• Data Drift Detection

• LLM-Based EV Analyst

• MCP Server Tools

Architecture

User
 │
 ▼
Streamlit Dashboard
 │
 ▼
EV Digital Twin Agent
 │
 ├── Battery Twin Engine
 ├── Charging Intelligence
 ├── Driving Intelligence
 ├── Predictive Maintenance
 ├── Fleet Monitoring
 ├── Drift Detection
 └── MCP Server Tools

Datasets

NASA Battery Dataset

Used for battery degradation modeling and SOH prediction.

EV Charging Patterns Dataset

Used for charging risk analysis and charging intelligence.

EV Telemetry Dataset

Used for driving behavior and energy consumption analytics.

MCP Tools

• predict_future_soh()

• estimate_rul()

Dashboard Modules

• Battery Twin

• Fleet Analytics

• Charging Intelligence

• Driving Intelligence

• Predictive Maintenance

• Monitoring

• Drift Detection

• AI Assistant

Installation

git clone https://github.com/mahmutcanborann/mcp-ev-digital-twin-agent.git

cd mcp-ev-digital-twin-agent

pip install -r requirements.txt

Run Dashboard

streamlit run dashboard/app.py

Future Work

• Full MCP Tool-Calling Agent

• Multi-Battery Comparison Tools

• Hugging Face Deployment

• Advanced Fleet AI Analyst