Contribution type: lightning talk & poster
Human-readable mathematics for energy system modelling
Presenter: @brynpickering
Description:
Energy system modelling (ESM) tools are usually linear optimisation wrappers; they process user data into optimisation problems using internally defined mathematical expressions.
Most ESM tools share a common set of these expressions but there is an increasing need for bespoke mathematical constraints, to reflect the complexity of real systems.
Defining this “custom” math is undertaken in the language-specific API of an ESM tool in most cases, creating a barrier to porting math to different contexts and to defining the math in the first place.
Meanwhile, the “core” math is often embedded in the tool source code and therefore completely disconnected from the data that is used to create the model.
In this talk, I will present a human-readable, YAML-based mathematics syntax which we have implemented in Calliope and linopy.
It can be used to store the core and custom math of a model alongside its data, ultimately increasing model transparency and reproducibility.
Optional links:
- Calliope YAML math docs: Defining your own math - Calliope
- Linopy YAML math development: Text-based (YAML) math definition interface · Issue #561 · PyPSA/linopy · GitHub
NOTE: this session could also be a 1.5hr breakout session dedicated to training in use of the math syntax and getting participants to define their own custom constraints.