Model Solve Infeasible - Multiple Energy Tiers?

[astronobri]

Hi, I used the Custom Rate Builder to create a multi-tier energy rate structure JSON file (below). When I give this file to REopt using ElectricTariff.urdb_response, I receive a warning (also below) about the different energy tiers and a model solve status as infeasible. I’m wondering why the model would be infeasible? I tried removing the 0.02 and 0.6 rate tiers, and the model solve status was optimal. Let me know if there is a better way that I can use multiple energy tiers in REopt. Thanks!

{"energyweekdayschedule":[[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]],"energyweekendschedule":[[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]],"energyratestructure":[[{"rate":0,"unit":"kWh"}],[{"rate":0.01,"max":100.0,"unit":"kWh"},{"rate":0.02,"max":500.0,"unit":"kWh"},{"rate":0.6,"unit":"kWh"}]],"demandratestructure":[[{"rate":0,"unit":"kW"}],[{"rate":0.0,"unit":"kW"}],[{"rate":10.01,"unit":"kW"}]],"demandweekdayschedule":[[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1],[1,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,1,2,2,2,1,1,1,1]],"demandweekendschedule":[[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1],[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]],"fixedchargefirstmeter":40.20,"fixedchargeunits":"$/day"}

The error message from REopt:

REopt | Warn: Energy periods contain different numbers of tiers, using limits of period with most tiers.
└ @ REopt ~/.julia/packages/REopt/Ml1hn/src/core/urdb.jl:218
┌ REopt | Warn: Energy periods contain different numbers of tiers, using limits of period with most tiers.
└ @ REopt ~/.julia/packages/REopt/Ml1hn/src/core/urdb.jl:218
┌ REopt | Warn: Adding binary variable to model gas generator. Some solvers are very slow with integer variables.
└ @ REopt ~/.julia/packages/REopt/Ml1hn/src/core/reopt.jl:588
┌ REopt | Warn: Adding binary variables to model energy cost tiers.
└ @ REopt ~/.julia/packages/REopt/Ml1hn/src/constraints/electric_utility_constraints.jl:237
┌ REopt | Info: Model built. Optimizing...
└ @ REopt ~/.julia/packages/REopt/Ml1hn/src/core/reopt.jl:522
Running HiGHS 1.6.0: Copyright (c) 2023 HiGHS under MIT licence terms
Presolving model
10405 rows, 35076 cols, 93960 nonzeros
Presolve: Infeasible

Solving report
Status Infeasible
Primal bound inf
Dual bound -inf
Gap inf
Solution status -
Timing 0.06 (total)
0.06 (presolve)
0.00 (postsolve)
Nodes 0
LP iterations 0 (total)
0 (strong br.)
0 (separation)
0 (heuristics)

[kurt-verdant]

maybe you need more square brackets?

[astronobri]

More square brackets I think would indicate multiple energy periods with one tier rate per period, rather than the scenario I’m trying to figure out which is just one energy period, but there are multiple energy tiers within that period. More like this example from REopt’s documentation for what the “max” input in energyratestructure means:

"For example, if you enter the following for Period 1:

Max. Energy Purchases (kWh/month)	Energy Charge ($/kWh)
100	$0.01
200	$0.10
900	$0.50

For a month in which Period 1 applies, if the site uses 1000 kWh, the site will be charged the following:

• Tier 1: $0.01*100 kWh
• Tier 2: $0.1*200 kWh (this tier applies to consumption between 100 kWh to 300 kWh
• Tier 3: $0.5*700 kWh (this tier applies to consumption between 300 kWh – 1200 kWh)"

[kurt-verdant]

Ahhh...now I understand. Sorry about that. Any chance you can use your saved rate in a Web Tool optimization, and see if the input JSON produced by the Web Tool contains any clues? I used your rate in a Web Tool run and it seemed to work.

I grabbed the energy rates out of the input JSON produced by the Web Tool and copied them into something else I was working on. It ran fine in REopt installed on my laptop. There are a whole lot of brackets at the end. maybe something is going on there.

[astronobri]

Thanks for sharing that it worked for you, that got me thinking about the other inputs I was using and I think I figured it out! My test demand curve had values that were too large compared to the ElectricStorage and Generator sizes I specified — REopt still gave the multiple energy tier warning, but it ran with an “Optimal” model solve after I scaled the demand curve down. The brackets you’re talking about look correct to me, it encloses all of the information in ElectricTariff. Thanks again!

[kurt-verdant]

Glad to hear you're back in business!