Flash and HDA Flowsheet Tutorial Updates (#160)

* Flash Unite and HDA Flowsheet Tutorial Revisions to reflect changes to initialization and scaling

* Flash Unite and HDA Flowsheet Tutorial Revisions to reflect changes to initialization and scaling

* Updated version numbers for GitHub Actions to version 4

* Revised both Tutorials based on feedback and ensured they both work as intended

* Fixed spelling and ran Black

* Fixing pytesting for earlier versions

* Created config files to implement modular properties framework and initialized the HDA flowsheet files to work with the modular properties framework. Currently not working due to unit model initialization error.

* Revised HDA Flowsheet to use the new State Definition FpTPxpc and add the necessary supporting files to make it work. Included a Python file of the flowsheet.

* Flash Unite and HDA Flowsheet Tutorial Revisions to reflect changes to initialization and scaling

* Flash Unite and HDA Flowsheet Tutorial Revisions to reflect changes to initialization and scaling

* Revised both Tutorials based on feedback and ensured they both work as intended

* Fixed spelling and ran Black

* Fixing pytesting for earlier versions

* Created config files to implement modular properties framework and initialized the HDA flowsheet files to work with the modular properties framework. Currently not working due to unit model initialization error.

* Revised HDA Flowsheet to use the new State Definition FpTPxpc and add the necessary supporting files to make it work. Included a Python file of the flowsheet.

* Revised and fixed the HDA tutorial to work with updated FpTPxpc state definition and initialization methods

* Ran black

* Ran black

* Ran ideasx pre

* Spelling fixes

* Spelling

* Fixed some testing and reverted `hda_ideal_VLE.py` back to its original state

* Relaxes assertion check for reactor temperature due to result variance between python/os versions in testing and populated doc notebook files

* Relaxes assertion check for reactor temperature due to result variance between python/os versions in testing

* Relaxes assertion check for reactor temperature due to result variance between python/os versions in testing

* Relaxes assertion check for reactor temperature due to result variance between python/os versions in testing

* Fixed reactor outlet temperature to remove indeterminancy and adjusted exercise cells to accommodate

* Changed reactor heat duty units from J to W and expanded explanation for fixing reactor outlet temperature

Author: tannerpolley

.github/workflows/core.yml

@@ -69,7 +69,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout source
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
       - name: Run Spell Checker
         uses: crate-ci/typos@master
 
@@ -95,7 +95,7 @@ jobs:
           - os: win64
             runner-image: windows-2022
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Set up Conda environment
         uses: conda-incubator/setup-miniconda@v2.2.0
         with:

idaes_examples/mod/hda/hda_flowsheet_extras.py

@@ -0,0 +1,237 @@
+import numpy as np
+from idaes.core.util.model_statistics import degrees_of_freedom
+from idaes.core.solvers import get_solver
+import idaes.logger as idaeslog
+
+
+def FpcTP_to_FpTPxpc(flow_mol_phase_comp):
+
+    flow_mol_phase = {
+        "Vap": 0,
+        "Liq": 0,
+    }
+    mole_frac_phase_comp = {}
+
+    for p, c in flow_mol_phase_comp.keys():
+        flow_mol_phase[p] += flow_mol_phase_comp[(p, c)]
+    for p, c in flow_mol_phase_comp.keys():
+        if flow_mol_phase[p] == 0.0:
+            mole_frac_phase_comp[p, c] = 0.0
+        else:
+            mole_frac_phase_comp[p, c] = flow_mol_phase_comp[(p, c)] / flow_mol_phase[p]
+
+    return flow_mol_phase, mole_frac_phase_comp
+
+
+def fix_inlet_states(m):
+
+    eps = 1e-5
+    flow_mol_phase_comp_101 = {
+        ("Vap", "benzene"): eps,
+        ("Vap", "toluene"): eps,
+        ("Vap", "hydrogen"): eps,
+        ("Vap", "methane"): eps,
+        ("Liq", "benzene"): eps,
+        ("Liq", "toluene"): 0.30,
+    }
+
+    flow_mol_phase_101, mole_frac_phase_comp_101 = FpcTP_to_FpTPxpc(
+        flow_mol_phase_comp_101
+    )
+
+    m.fs.I101.flow_mol_phase[0, "Vap"].fix(flow_mol_phase_101["Vap"])
+    m.fs.I101.flow_mol_phase[0, "Liq"].fix(flow_mol_phase_101["Liq"])
+    m.fs.I101.mole_frac_phase_comp[0, "Vap", "benzene"].fix(
+        mole_frac_phase_comp_101["Vap", "benzene"]
+    )
+    m.fs.I101.mole_frac_phase_comp[0, "Vap", "toluene"].fix(
+        mole_frac_phase_comp_101["Vap", "toluene"]
+    )
+    m.fs.I101.mole_frac_phase_comp[0, "Vap", "hydrogen"].fix(
+        mole_frac_phase_comp_101["Vap", "hydrogen"]
+    )
+    m.fs.I101.mole_frac_phase_comp[0, "Vap", "methane"].fix(
+        mole_frac_phase_comp_101["Vap", "methane"]
+    )
+    m.fs.I101.mole_frac_phase_comp[0, "Liq", "benzene"].fix(
+        mole_frac_phase_comp_101["Liq", "benzene"]
+    )
+    m.fs.I101.mole_frac_phase_comp[0, "Liq", "toluene"].fix(
+        mole_frac_phase_comp_101["Liq", "toluene"]
+    )
+    m.fs.I101.temperature.fix(303.2)
+    m.fs.I101.pressure.fix(350000)
+
+    flow_mol_phase_comp_102 = {
+        ("Vap", "benzene"): eps,
+        ("Vap", "toluene"): eps,
+        ("Vap", "hydrogen"): 0.30,
+        ("Vap", "methane"): 0.02,
+        ("Liq", "benzene"): eps,
+        ("Liq", "toluene"): eps,
+    }
+
+    flow_mol_phase_102, mole_frac_phase_comp_102 = FpcTP_to_FpTPxpc(
+        flow_mol_phase_comp_102
+    )
+
+    m.fs.I102.flow_mol_phase[0, "Vap"].fix(flow_mol_phase_102["Vap"])
+    m.fs.I102.flow_mol_phase[0, "Liq"].fix(flow_mol_phase_102["Liq"])
+    m.fs.I102.mole_frac_phase_comp[0, "Vap", "benzene"].fix(
+        mole_frac_phase_comp_102["Vap", "benzene"]
+    )
+    m.fs.I102.mole_frac_phase_comp[0, "Vap", "toluene"].fix(
+        mole_frac_phase_comp_102["Vap", "toluene"]
+    )
+    m.fs.I102.mole_frac_phase_comp[0, "Vap", "hydrogen"].fix(
+        mole_frac_phase_comp_102["Vap", "hydrogen"]
+    )
+    m.fs.I102.mole_frac_phase_comp[0, "Vap", "methane"].fix(
+        mole_frac_phase_comp_102["Vap", "methane"]
+    )
+    m.fs.I102.mole_frac_phase_comp[0, "Liq", "benzene"].fix(
+        mole_frac_phase_comp_102["Liq", "benzene"]
+    )
+    m.fs.I102.mole_frac_phase_comp[0, "Liq", "toluene"].fix(
+        mole_frac_phase_comp_102["Liq", "toluene"]
+    )
+    m.fs.I102.temperature.fix(303.2)
+    m.fs.I102.pressure.fix(350000)
+
+    tear_guesses = {
+        "flow_mol_phase": {
+            (0, "Liq"): flow_mol_phase_101["Liq"],
+            (0, "Vap"): flow_mol_phase_102["Vap"],
+        },
+        "mole_frac_phase_comp": {
+            (0, "Liq", "benzene"): mole_frac_phase_comp_101["Liq", "benzene"],
+            (0, "Liq", "toluene"): mole_frac_phase_comp_101["Liq", "toluene"],
+            (0, "Vap", "benzene"): mole_frac_phase_comp_102["Vap", "benzene"],
+            (0, "Vap", "toluene"): mole_frac_phase_comp_102["Vap", "toluene"],
+            (0, "Vap", "methane"): mole_frac_phase_comp_102["Vap", "hydrogen"],
+            (0, "Vap", "hydrogen"): mole_frac_phase_comp_102["Vap", "methane"],
+        },
+        "temperature": {0: 303},
+        "pressure": {0: 350000},
+    }
+
+    return tear_guesses
+
+
+def initialize_unit(unit):
+    from idaes.core.util.exceptions import InitializationError
+    import idaes.logger as idaeslog
+
+    optarg = {
+        "nlp_scaling_method": "user-scaling",
+        "OF_ma57_automatic_scaling": "yes",
+        "max_iter": 1000,
+        "tol": 1e-8,
+    }
+
+    try:
+        initializer = unit.default_initializer(solver_options=optarg)
+        initializer.initialize(unit, output_level=idaeslog.INFO_LOW)
+    except InitializationError:
+        solver = get_solver(solver_options=optarg)
+        solver.solve(unit)
+
+
+def manual_propagation(m, tear_guesses):
+    from idaes.core.util.initialization import propagate_state
+
+    print(f"The DOF is {degrees_of_freedom(m)} initially")
+    m.fs.s03_expanded.deactivate()
+    print(f"The DOF is {degrees_of_freedom(m)} after deactivating the tear stream")
+
+    for k, v in tear_guesses.items():
+        for k1, v1 in v.items():
+            getattr(m.fs.s03.destination, k)[k1].fix(v1)
+
+    DOF_initial = degrees_of_freedom(m)
+
+    print(f"The DOF is {degrees_of_freedom(m)} after setting the tear stream")
+
+    optarg = {
+        "nlp_scaling_method": "user-scaling",
+        "OF_ma57_automatic_scaling": "yes",
+        "max_iter": 300,
+        # "tol": 1e-10,
+    }
+
+    solver = get_solver(solver_options=optarg)
+
+    initialize_unit(m.fs.H101)  # Initialize Heater
+    propagate_state(m.fs.s04)  # Establish connection between Heater and Reactor
+    initialize_unit(m.fs.R101)  # Initialize Reactor
+    propagate_state(
+        m.fs.s05
+    )  # Establish connection between Reactor and First Flash Unit
+    initialize_unit(m.fs.F101)  # Initialize First Flash Unit
+    propagate_state(
+        m.fs.s06
+    )  # Establish connection between First Flash Unit and Splitter
+    propagate_state(
+        m.fs.s07
+    )  # Establish connection between First Flash Unit and Second Flash Unit
+    initialize_unit(m.fs.S101)  # Initialize Splitter
+    propagate_state(m.fs.s08)  # Establish connection between Splitter and Compressor
+    initialize_unit(m.fs.C101)  # Initialize Compressor
+    propagate_state(m.fs.s09)  # Establish connection between Compressor and Mixer
+    initialize_unit(m.fs.I101)  # Initialize Toluene Inlet
+    propagate_state(m.fs.s01)  # Establish connection between Toluene Inlet and Mixer
+    initialize_unit(m.fs.I102)  # Initialize Hydrogen Inlet
+    propagate_state(m.fs.s02)  # Establish connection between Hydrogen Inlet and Mixer
+    initialize_unit(m.fs.M101)  # Initialize Mixer
+    propagate_state(m.fs.s03)  # Establish connection between Mixer and Heater
+    solver.solve(m.fs.F102)
+    propagate_state(
+        m.fs.s10
+    )  # Establish connection between Second Flash Unit and Benzene Product
+    propagate_state(
+        m.fs.s11
+    )  # Establish connection between Second Flash Unit and Toluene Product
+    propagate_state(m.fs.s12)  # Establish connection between Splitter and Purge Product
+
+    optarg = {
+        "nlp_scaling_method": "user-scaling",
+        "OF_ma57_automatic_scaling": "yes",
+        "max_iter": 300,
+        "tol": 1e-8,
+    }
+    solver = get_solver("ipopt_v2", options=optarg)
+    solver.solve(m, tee=False)
+
+    for k, v in tear_guesses.items():
+        for k1, v1 in v.items():
+            getattr(m.fs.H101.inlet, k)[k1].unfix()
+
+    m.fs.s03_expanded.activate()
+    print(
+        f"The DOF is {degrees_of_freedom(m)} after unfixing the values and reactivating the tear stream"
+    )
+
+
+def automatic_propagation(m, tear_guesses):
+
+    from pyomo.network import SequentialDecomposition
+
+    seq = SequentialDecomposition()
+    seq.options.select_tear_method = "heuristic"
+    seq.options.tear_method = "Wegstein"
+    seq.options.iterLim = 5
+
+    # Using the SD tool
+    G = seq.create_graph(m)
+    heuristic_tear_set = seq.tear_set_arcs(G, method="heuristic")
+    order = seq.calculation_order(G)
+
+    # Pass the tear_guess to the SD tool
+    seq.set_guesses_for(heuristic_tear_set[0].destination, tear_guesses)
+
+    print(f"Tear Stream starts at: {heuristic_tear_set[0].destination.name}")
+
+    for o in order:
+        print(o[0].name)
+
+    seq.run(m, initialize_unit)