Should natural resources be defined on a per-technology basis or at the site level?

[johnjasa]

How should natural resources be defined? As coupled to technologies or at the site level?

Site level makes sense based on physical reality, but coupled to the technologies makes more sense computationally as certain performance and cost models may assume that you're inputting resource data in a certain format.

To clarify through an example, which part of the model should "own" the wind resource data? Is it the site itself or the wind plant technology?

[johnjasa]

Gen suggests that resources should be coupled to the technologies, not at the site level

[jaredthomas68]

I think this may be the wrong question. I don't think resources should be owned by the technology. I think resources should be owned by the system, kind of in between the site and the technology levels.

However, I think it could be helpful to allow any resource to be attached to a given site as an input resource that just exists at that site. I'm not sure how that would work in practice though. This may be how we differentiate between starting resources and inbetween/output resources. Initial resource profiles could possibly come from the site and other resources would come from technologies.

[jaredthomas68]

One other thought here is that resources should not be owned by technologies because multiple technologies may need to share resources

[bayc]

This is a tricky one, as we want to require for certain techs that certain resources be available, thinking wind, but then maybe a simple constant wind speed could be specified.

One point of reference would be the WindIO schema, where the resource is tied to the site portion of the definition, and not the wind energy system (technology) definition.

[kbrunik]

I think this question is very applicable to offshore/onshore configurations and virtual power plants where the site might span several kilometers.

I'd like the resource to be defined at the plant level but can be attached to a specific technology if in needs to be constrained. I'd imagine that you have offshore wind and onshore wind, each gets a different wind resource profile (maybe different hub height resource). It gets constrained in the input YAML that offshore resource is for offshore turbines and same for onshore. Then you get electricity from both but they are spatially constrained and can't be combined unless a cable connects them to a substation. Substation could be a whole new location.

Handling resources at the site level would allows for simpler use cases where everything is co-located. It would allow for easier summary statistics of resource.