Current Status of Models 1b and 2b¶
Outlined here is the status of the Model 1b and 2b code framework.
Status of Model Components¶
Dispatch Model: [complete] a Pyomo dispatch MILP has been created and runs correctly within the project. It sets up an MILP and solves it when called.
SSC Model: [in progress] new classes have been added to SSC suite and work but calling through PySAM is currently not working (only through PySSC).
mdot_solver
also needs to be overloaded with correct stream calculations.NE2 Model: [complete] a Python interface class has been created and runs correctly within the project. It creates dispatch model instances, solves them, converts results to dispatch targets, and calls on SSC to execute at every timestep of the time horizon.
Framework within SSC¶
Compute module (cmod): cmod_nuclear_mspt_indirect_tes
This part works
major modifications include new parameters and setting them correctly within member objects
csp_solver: csp_dual_indirect_solver_core
This part mostly works
needs some debugging within the estimation of plant component modes and operating mode selection
csp_solver_mono_eq: csp_dual_indirect_solver_mono_eq_methods
This part is not finished
Need to overload mass flow solver - calculate the correct streams and temperatures and set the proper inputs within the PC and TES object (mix the streams, do HX calculations, etc.)
Plant Component solvers: within the compute module, an object instance of all plant components are initialized with corresponding SSC inputs. Then these objects are used as inputs for the
csp_solver
.- TES: csp_solver_two_tank_tes
No changes
- PC: csp_solver_pc_Rankine_indirect_224
Some changes enacted, debugging
- CSP: csp_solver_mspt_collector_receiver
Receiver: csp_solver_mspt_receiver_222
- Heliostatfield: csp_solver_pt_heliostatfield
No changes
- LFR: csp_solver_nuclear_plant
- Nuclear: csp_solver_nuclear
Some changes enacted, debugging