MicroGridsPy 2.1

Description

The MicroGridsPy model main objective is to provide an open-source alternative to the problem of sizing and dispatch of energy in micro-grids in isolated places. It’s written in python(pyomo) and use excel and text files as input and output data handling and visualisation.

Here the link to the online documentation: MicroGridsPy Documentation.

⚠️ Warning: For a MACOS compatible version, please refer to version 2.0. The project is under active development.

Main Features

• Optimal sizing of PV panels, wind turbines, other renewable technologies, back-up genset and electrochemical storage system for least cost electricity supply in rural remote areas
• Optimal dispatch from the identified supply systems
• Possibility to optimize on NPC or operation costs
• LCOE evaluation for the identified system
• Two-stage stochastic optimization
• User Interface

Advanced Features

MicroGridsPy 2.1 incorporates several advanced features, enhancing its versatility and effectiveness in various scenarios:

• Variable Fuel Costs: Adjust fuel costs dynamically based on market prices.
• MILP Formulation for Unit Commitment: Ensures optimal operation of generation sources.
• Partial Load Operation of Diesel Genset: Optimizes diesel genset operations, even under partial load, using MILP.
• High-Resolution Optimization: Conducts optimization at a 1-minute time-step, suitable for coupling with RAMP load profiles.
• Evolving Load Demand & Multi-Step Capacity Expansion: Adapts to changing load demands over multiple years.

• Optimization of Grid-Connected Microgrids: Enhances the efficiency of grid-connected systems.

• Two-Objective Optimization: Focuses on minimizing either Net Present Cost or CO2 emissions.

• Brownfield Optimization: Tailors solutions for existing infrastructure upgrades.
• Built-In Load Archetypes: Specifically designed for rural users in Sub-Sahara Africa.
• NASA POWER Integration: Utilizes NASA POWER for calculating renewable energy production.

Environments and Solvers

In the current repository under the Environments branch, environments for both macOS and Windows are available. Currently, MicroGridsPy has been confirmed to work with Gurobi and GLPK.

⚠️ Warning: While GLPK is a capable solver for many optimization problems, it tends to operate slower compared to commercial solvers like Gurobi, especially for large or complex problems. The performance difference can be significant, potentially ranging from several times to orders of magnitude faster, depending on the specifics of the problem. These observations are general, and actual performance will vary with each unique problem. It is advisable to consider this factor when choosing a solver for time-sensitive or large-scale applications.

Getting Started with MicroGridsPy

To install MicroGridsPy, the conda package manager is recommended. Follow this step-by-step guide:

1. Install Anaconda:

   • Download and install the Anaconda distribution for Python 3 from the Anaconda Archive. Anaconda simplifies package management and deployment, especially for data science and machine learning applications.

2. Download the MicroGridsPy Environment File:

   • The environment file for MicroGridsPy is available on the MicroGridsPy/Environments SESAM GitHub repository. Download the file to prepare your modeling environment.

3. Create and Activate the MicroGridsPy Environment:

   • Save the downloaded mgpy_win.yml file in an accessible directory (e.g., C:\\Users\\youruser).
   • Open the Anaconda Prompt, navigate to the directory, and execute the following commands:

     conda env create -f mgpy_win.yml
     conda activate mgpy
     

4. Development Environment Setup:

   • Make sure your development environment includes necessary Python packages such as Pyomo (minimum version 6.4.3), Pandas, NumPy, and Matplotlib.

5. Launch Spyder IDE for Development:

   • Use the Spyder IDE for development and debugging, available in the Anaconda distribution. Start Spyder with the following commands:

     conda activate mgpy
     spyder
     

6. Run the GUI Application:

   • MicroGridsPy includes a GUI for easier data input. Launch the GUI by running app_main.py from the Code/User Interface folder using Spyder or your preferred IDE.

These steps will guide you through a successful installation of MicroGridsPy.

Maintenance and Development

MicroGridsPy is currently mantained and developed by the SESAM group in the Department of Energy Engineering at Politecnico di Milano. The research activity of SESAM focuses on using mathematical models for the study of systems, components, and processes in the energy field and industrial ecology.

Contributors of the current version:

• Nicolò Stevanato, Politecnico di Milano nicolo.stevanato@polimi.it
• Alessandro Onori, Politecnico di Milano alessandro.onori@polimi.it
• Riccardo Mereu, Politecnico di Milano riccardo.mereu@polimi.it
• Emanuela Colombo, Politecnico di Milano emanuela.colombo@polimi.it

Past contributors:

• Gianluca Pellecchia, Politecnico di Milano gianluca.pellecchia@polimi.it
• Ivan Sangiorgio, Politecnico di Milano ivan.sangiorgio@polimi.it
• Francesco Lombardi, TU Delft f.lombardi@tudelft.nl
• Giulia Guidicini, Politecnico di Milano giulia.guidicini@polimi.it
• Lorenzo Rinaldi, Politecnico di Milano lorenzo.rinaldi@polimi.it

Based on the original model by:

• Sergio Balderrama, Universidad Mayor de San Simón s.balderrama@umss.edu
• Sylvain Quoilin, Université de Liège squoilin@uliege.be

License

This is a free software licensed under the “European Union Public Licence" EUPL v1.1. It can be redistributed and/or modified under the terms of this license.