Store process parameters in `Asset`

[tsmbland]

Given that Assets need to look up various process parameters for their commission year, I think it would be better to just do that lookup once and store the process parameters as a field in the asset. You could store Rc<ProcessParameter>s in the parameter map to make them cheap to copy (also useful for when there are ranges of years specified).

Originally posted by @alexdewar in #475 (review)

[tsmbland]

While we're at it, we should do this for flows and energy limits as well

[tsmbland]

While we're at it, we should do this for flows and energy limits as well

I made a start tackling #495, but realised it would be easier if we do this first

[alexdewar]

I made a start on this yesterday, so I can try to tack on these other bits today and wrap it up.

[alexdewar]

I've had a quick look and the problem with ProcessEnergyLimitsMap is that it's keyed by time slice in addition to region and year, so to do this we'd either have to turn it into a map of maps (potentially a bit gross) or create a new map with just time slices as keys, but that's also a bit more of a faff and less of an obvious win (we only use it in one place currently). Both of these are options though.

I'm not sure what you mean with flows actually... it's not a map currently anyway. Were you planning to add region/year fields to process_flows.csv? There is an issue open for turning it into a map (#449), but that would only be with commodity ID as a key, so that's kind of separate.

[tsmbland]

I've had a quick look and the problem with ProcessEnergyLimitsMap is that it's keyed by time slice in addition to region and year, so to do this we'd either have to turn it into a map of maps (potentially a bit gross) or create a new map with just time slices as keys, but that's also a bit more of a faff and less of an obvious win (we only use it in one place currently). Both of these are options though.

Hmm yes this is a bit of a conundrum. Maybe we leave this for now and if we end up using this data a lot (e.g. in agent investment) we can come back to it

I'm not sure what you mean with flows actually... it's not a map currently anyway. Were you planning to add region/year fields to process_flows.csv? There is an issue open for turning it into a map (#449), but that would only be with commodity ID as a key, so that's kind of separate.

Yes #495

I started this (#538), but it got a bit fiddly as we already use this data a lot so figured it would be worth doing this first

An alternative is that we store the full Process object in Asset (as an rc), then add methods to get the appropriate data based on the asset's region and commission year