EV-Solar-Sim

Digital Twin Framework for EV Charging Optimization

A dynamic, computation-driven platform for optimizing Electric Vehicle (EV) charging infrastructure and policy planning in localized urban systems.

Overview and Contribution

This repository provides the source code for a novel Digital Twin framework designed to manage the complexities of accelerating EV adoption in urban environments. We advance beyond traditional static models by integrating agent-based decision support with embedded metaheuristic optimization.

Our core contribution is a flexible, computation-driven platform for EV infrastructure planning, with a transferable, modular design.

---

Technical Stack and Setup

This project uses GAMA for agent-based simulation and Jupyter Notebook for post-simulation data analysis.

Prerequisites

• GAMA Platform 1.9.3
• Python 3.8+ (for data analysis in Jupyter Notebook)

Installation

1. Clone the Repository:

   git clone https://github.com/dolinh11/EV-Solar-Sim
   cd EV-Solar-Sim

2. GAMA Import:

   • Import the cloned EvSolarSim repository folder directly into the GAMA Platform workspace.

3. Install Python Dependencies:

   • Install all required Python packages for data analysis.

4. Data Placement:

   • Place all new data files (map, weather data, etc.) into the designated folder: ./includes

Configuration Notes

• To tune the model for a new location, modify the driver behavior in ./models/traffic.gaml and charging station configurations in ./models/charging_station.gaml.

Attention (Ensure variable names are consistent): The structure remains the same, but you can tune the logic and map to match your localized urban systems.

---

Run and Analyze

1. Run Simulation:
   • Run the main simulation file in the GAMA Platform: ./models/main.gaml
2. Download Data:
   • Download the simulation output data into the dedicated folder: ./res_data
3. Analysis:
   • Run the analysis notebook in Jupyter Notebook: Data Analysis.ipynb

---

⚖️ License and Citation

Citation

If you use this framework or data in your research, please cite our corresponding work. We recommend citing the ArXiv preprint first: