Get Started

For a more detailed description of each of the steps mentioned below, see the following pages:

• Creating Model Templates
• Defining System Topology
• Creating Input Data Templates
• Define Input Data
• Define Model Configuration
• Model construction and solving
• Check Model Diagnostics
• Manage results

In addition to the workflow documentation below, we provide three documented case studies which can be useful to see how AdOpT-NET0 can be used to implement your optimization problem.

In short

This workflow documentation will guide you through all the steps that are required to prepare an exemplary model. Here, we optimize an energy system with a flat electricity demand over a full year that can be served by imports from the grid, or by installing photovoltaic panels and a battery.

The modelling steps are as follows:

• Create an empty working directory for all input data. The subsequent functions will create the data structure you need for the model

• Create the templates for the system topology and the model configuration.

  import adopt_net0 as adopt
  import json
  from pathlib import Path
  import pandas as pd
  
  input_data_path = Path("path_to_your_input_data_folder")
  adopt.create_optimization_templates(input_data_path)
  

• Define your system topology and the model configuration.

  with open(input_data_path / "Topology.json", "r") as json_file:
      topology = json.load(json_file)
  # Nodes
  topology["nodes"] = ["node1"]
  # Carriers:
  topology["carriers"] = ["electricity"]
  # Investment periods:
  topology["investment_periods"] = ["period1"]
  # Save json template
  with open(input_data_path / "Topology.json", "w") as json_file:
      json.dump(topology, json_file, indent=4)
  

• Create the folder structure and templates for the input data files.

  adopt.create_input_data_folder_template(input_data_path)
  

• Load and define input data (e.g., weather data, technology performance, demand data, etc.).

  # Define node locations (here an exemplary location in the Netherlands)
  node_locations = pd.read_csv(input_data_path / "NodeLocations.csv", sep=";",
                       index_col=0)
  node_locations.loc["node1", "lon"] = 5.5
  node_locations.loc["node1", "lat"] = 52.5
  node_locations.loc["node1", "alt"] = 10
  node_locations.to_csv(input_data_path / "NodeLocations.csv", sep=";")
  
  # Define new technologies
  with open(input_data_path / "period1" / "node_data" / "node1" / "Technologies.json", "r") as json_file:
      technologies = json.load(json_file)
  technologies["new"] = ["Photovoltaic", "Storage_Battery"]
  with open(input_data_path / "period1" / "node_data" / "node1" / "Technologies.json", "w") as json_file:
      json.dump(technologies, json_file, indent=4)
  
  # Copy over technology files
  adopt.copy_technology_data(input_data_path)
  
  # Define climate data
  adopt.load_climate_data_from_api(input_data_path)
  
  # Define demand
  adopt.fill_carrier_data(input_data_path, value_or_data=0.01, columns=['Demand'], carriers=['electricity'], nodes=['node1'])
  adopt.fill_carrier_data(input_data_path, value_or_data=100, columns=['Import price'], carriers=['electricity'], nodes=['node1'])
  adopt.fill_carrier_data(input_data_path, value_or_data=1, columns=['Import limit'], carriers=['electricity'], nodes=['node1'])
  

• Define model configuration if you want to change something from the defaults. Make sure that the result folder path in reporting/save_path refers to an existing folder

• Construct and solve the model and, possibly, incorporate options to lower the complexity of the model.

  m = adopt.ModelHub()
  m.read_data(input_data_path, start_period=None, end_period=None)
  m.quick_solve()
  

• If something unexpected happens: check the model diagnostics.

• Obtain and interpret the optimization results. Therefore, you can upload the H5 file saved on the visualization platform provided: https://resultvisualization.streamlit.app/.

An elaborate example of how to set up the model accordingly can be found below. To understand what happens behind the scenes, please take a look at the Source Code Documentation.