GridPenguin

Introduction

GridPenguin is a powerful and comprehensive tool to simulate a district heating system(DHS) with both good accuracy and high speed. We designed GridPenguin as a benchmark tool to compare different DHS optimization algorithms. It also give possibilities to new optimization algorithms that requires a fast and iterative process with a DHS simulator, e.g. data-driven machine learning algorithms. For a detailed comprehensive documentary about the physical & computational logic and scientific significiency, please go to our publication at NEFI: presentation and paper.

Dependencies

• Python 3.10.6
• Numpy 1.23.3
• Scipy 1.9.1
• BeautifulTable 1.1.0

Get Started

Install local package with pip install -e .

Go to one folder above the root, then run $python -m grid-penguin.examples.example_one_consumer. This script build and run a one-consumer DHS with demand, supply temperature etc. specified in the script.

You can also run a 3-consumer DHS by $python -m grid-penguin.examples.example_3_consumers.

Directory Structure

• Interfaces: the interfaces of GridPenguin
  • grid_interface.py
• Case: construction of different network structure
  • one_consumer.py: network with one producer, one consumer and two pipes
  • parallel_consumer.py: network with parallel multiple consumers, one producer
• Models: building blocks of the network
  • Producers: models of the producer, currently only contains the CHP
    • CHP.py: model of the CHP
  • grid.py: container for all nodes and edges
  • producer.py: production unit (CHP is also a producer, but here the producer is modeled much simplified)
  • consumer.py: end consumer that has heat demand
  • edge.py: water pipes
  • transfer.py: Heat exchange station
• Util: utilities and configurations
• Wanda_compare: to build dummy grid and compare performance with Wanda

Support For questions and issues, please open a github issue.