ECMAS-v1

This repository contains the official implementation of the paper:

“Gradual Introduction of Carbon Allowance Auctions Facilitate the Sustainable Emission Reduction in the Power Sector”

This study develops a bottom-up multi-agent simulation model (ECMAS) to analyze the interaction between the electricity market and carbon market under different policy designs.

• ⚡ Integrated electricity market + carbon market (primary & secondary)

• 🧠 Multi-agent simulation with 2,000+ heterogeneous power enterprises

• 🔄 Dynamic policy design:

  • Quota reduction pathways
  • Auction ratio evolution

• 📊 Evaluation of:

  • Carbon prices
  • Emission reduction
  • Energy structure
  • Market stability

The ECMAS model simulates interactions among three core agents:

• Government
• Power grid
• Power enterprises (2241 agents)

And three coupled markets:

• Electricity market
• Primary carbon market
• Secondary carbon market

The model consists of six modules:

1. Electricity production
2. Electricity trading
3. Low-carbon technology transformation
4. Carbon allowance allocation
5. Carbon allowance trading
6. Unit investment and retirement

• 🚫 Aggressive auction introduction + rapid quota reduction

  • Causes carbon price collapse
  • Leads to supply risk

• ✅ Gradual auction introduction (Smooth scenario)

  • Stabilizes carbon prices
  • Supports low-carbon investment
  • Achieves sustainable emission reduction

• ⚖️ Optimal policy path:

  • Gradual tightening + phased auction increase

• Python 3.9+
• Mesa (agent-based modeling framework)

This project is licensed under the CC BY-NC 4.0 License.

• 🚫 Non-commercial use only
• 📚 Research purposes only
• 📄 Citation required

If you use this code, please cite:

@article{li2026ecmas,
  title={Gradual Introduction of Carbon Allowance Auctions Facilitate the Sustainable Emission Reduction in the Power Sector},
  author={Li, Zhao-Yuan and Zhao, Lu-Tao and Qu, Zhi and Chen, Zhe-Yi and Qiu, Rui-Xiang and An, Xing-Yu and Wang, Dai-Song},
  journal={iScience},
  year={2026}
}

This code is provided "as is" without warranty. The authors are not responsible for any damages arising from its use.