Make test/ a standalone project

LMP Methods & Profiled Units

.github/workflows/test.yml

@@ -1,4 +1,4 @@
-name: Tests
+name: Build & Test
 on:
   push:
   pull_request:
@@ -6,19 +6,30 @@ on:
     - cron: '45 10 * * *'
 jobs:
   test:
+    name: Julia ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }}
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        julia-version: ['1.3', '1.4', '1.5', '1.6']
-        julia-arch: [x64, x86]
-        os: [ubuntu-latest, windows-latest, macOS-latest]
-        exclude:
-          - os: macOS-latest
-            julia-arch: x86
+        version: ['1.6', '1.7', '1.8', '1.9']
+        os:
+          - ubuntu-latest
+        arch:
+          - x64
     steps:
       - uses: actions/checkout@v2
-      - uses: julia-actions/setup-julia@latest
+      - uses: julia-actions/setup-julia@v1
         with:
-          version: ${{ matrix.julia-version }}
-      - uses: julia-actions/julia-buildpkg@latest
-      - uses: julia-actions/julia-runtest@latest
+          version: ${{ matrix.version }}
+          arch: ${{ matrix.arch }}
+      - name: Run tests
+        shell: julia --color=yes --project=test {0}
+        run: |
+          using Pkg
+          Pkg.develop(path=".")
+          Pkg.update()
+          using UnitCommitmentT
+          try
+            runtests()
+          catch
+            exit(1)
+          end

.gitignore

@@ -1,20 +1,38 @@
 *.bak
 *.gz
-*.lastrun
-*.so
-*.mps
 *.ipynb
+*.lastrun
+*.mps
+*.so
+*/Manifest.toml
+.AppleDB
+.AppleDesktop
+.AppleDouble
+.DS_Store
+.DocumentRevisions-V100
+.LSOverride
+.Spotlight-V100
+.TemporaryItems
+.Trashes
+.VolumeIcon.icns
+._*
+.apdisk
+.com.apple.timemachine.donotpresent
+.fseventsd
 .ipy*
+.vscode
+Icon
+Manifest.toml
+Network Trash Folder
+TODO.md
+Temporary Items
 benchmark/results
 benchmark/runs
 benchmark/tables
 benchmark/tmp.json
 build
+docs/_build
 instances/**/*.json
 instances/_source
 local
 notebooks
-TODO.md
-docs/_build
-.vscode
-Manifest.toml

CHANGELOG.md

@@ -11,6 +11,21 @@ All notable changes to this project will be documented in this file.
 [semver]: https://semver.org/spec/v2.0.0.html
 [pkjjl]: https://pkgdocs.julialang.org/v1/compatibility/#compat-pre-1.0
 
+## [0.3.0] - 2022-07-18
+### Added
+- Add support for multiple reserve products and zonal reserves.
+- Add flexiramp reserve products, following WanHob2016's formulation (@oyurdakul, #21).
+- Add 365 variations for each MATPOWER instance, corresponding to each day of the year.
+
+### Changed
+- To support multiple/zonal reserves, the input data format has been modified as follows:
+    - In `Generators`, replace `Provides spinning reserves?` by `Reserve eligibility`
+    - In `Parameters`, remove `Reserve shortfall penalty`
+    - Revise `Reserves` section
+- To allow new versions of UnitCommitment.jl to read old instance files, a new required field `Version` has been added to the `Parameters` section. To load v0.2 files in v0.3, please add `{"Parameters":{"Version":"0.2"}}` to the file.
+- Benchmark test cases are now downloaded on-the-fly as needed, instead of being stored in our GitHub repository. Test cases can also be directly downloaded from: https://axavier.org/UnitCommitment.jl/
+
+
 ## [0.2.2] - 2021-07-21
 ### Fixed
 - Fix small bug in validation scripts related to startup costs

LICENSE.md

@@ -1,4 +1,4 @@
-Copyright © 2020, UChicago Argonne, LLC
+Copyright © 2020-2022, UChicago Argonne, LLC
 
 All Rights Reserved
 

Makefile

@@ -2,31 +2,10 @@
 # Copyright (C) 2020, UChicago Argonne, LLC. All rights reserved.
 # Released under the modified BSD license. See COPYING.md for more details.
 
-JULIA := julia --color=yes --project=@.
-VERSION := 0.2
-
-build/sysimage.so: src/utils/sysimage.jl Project.toml Manifest.toml
-	mkdir -p build
-	mkdir -p benchmark/results/test
-	cd benchmark; $(JULIA) --trace-compile=../build/precompile.jl benchmark.jl test/case14
-	$(JULIA) src/utils/sysimage.jl
-
-clean:
-	rm -rf build/*
+VERSION := 0.3
 
 docs:
-	cd docs; make clean; make dirhtml
-	rsync -avP --delete-after docs/_build/dirhtml/ ../docs/$(VERSION)/
+	cd docs; julia --project=. make.jl; cd ..
+	rsync -avP --delete-after docs/build/ ../docs/$(VERSION)/
 	
-test: build/sysimage.so
-	@echo Running tests...
-	$(JULIA) --sysimage build/sysimage.so -e 'using Pkg; Pkg.test("UnitCommitment")' | tee build/test.log
-
-
-format:
-	julia -e 'using JuliaFormatter; format(["src", "test", "benchmark"], verbose=true);'
-
-install-deps:
-	julia -e 'using Pkg; Pkg.add(PackageSpec(name="JuliaFormatter", version="0.14.4"))'
-
-.PHONY: docs test format install-deps
+.PHONY: docs

Project.toml

@@ -2,10 +2,11 @@ name = "UnitCommitment"
 uuid = "64606440-39ea-11e9-0f29-3303a1d3d877"
 authors = ["Santos Xavier, Alinson <axavier@anl.gov>"]
 repo = "https://github.com/ANL-CEEESA/UnitCommitment.jl"
-version = "0.2.2"
+version = "0.3.0"
 
 [deps]
 DataStructures = "864edb3b-99cc-5e75-8d2d-829cb0a9cfe8"
+Distributed = "8ba89e20-285c-5b6f-9357-94700520ee1b"
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 GZip = "92fee26a-97fe-5a0c-ad85-20a5f3185b63"
 JSON = "682c06a0-de6a-54ab-a142-c8b1cf79cde6"
@@ -15,23 +16,15 @@ Logging = "56ddb016-857b-54e1-b83d-db4d58db5568"
 MathOptInterface = "b8f27783-ece8-5eb3-8dc8-9495eed66fee"
 PackageCompiler = "9b87118b-4619-50d2-8e1e-99f35a4d4d9d"
 Printf = "de0858da-6303-5e67-8744-51eddeeeb8d7"
+Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 SparseArrays = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
 
 [compat]
-Cbc = "0.7"
 DataStructures = "0.18"
 Distributions = "0.25"
 GZip = "0.5"
 JSON = "0.21"
-JuMP = "0.21"
-MathOptInterface = "0.9"
+JuMP = "1"
+MathOptInterface = "1"
 PackageCompiler = "1"
 julia = "1"
-
-[extras]
-Cbc = "9961bab8-2fa3-5c5a-9d89-47fab24efd76"
-Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
-Gurobi = "2e9cd046-0924-5485-92f1-d5272153d98b"
-
-[targets]
-test = ["Cbc", "Test", "Gurobi"]

README.md

@@ -87,19 +87,22 @@ UnitCommitment.write("/tmp/output.json", solution)
 
 ## Documentation
 
-1. [Usage](https://anl-ceeesa.github.io/UnitCommitment.jl/0.2/usage/)
-2. [Data Format](https://anl-ceeesa.github.io/UnitCommitment.jl/0.2/format/)
-3. [Instances](https://anl-ceeesa.github.io/UnitCommitment.jl/0.2/instances/)
-4. [JuMP Model](https://anl-ceeesa.github.io/UnitCommitment.jl/0.2/model/)
+1. [Usage](https://anl-ceeesa.github.io/UnitCommitment.jl/0.3/usage/)
+2. [Data Format](https://anl-ceeesa.github.io/UnitCommitment.jl/0.3/format/)
+3. [Instances](https://anl-ceeesa.github.io/UnitCommitment.jl/0.3/instances/)
+4. [JuMP Model](https://anl-ceeesa.github.io/UnitCommitment.jl/0.3/model/)
+5. [API Reference](https://anl-ceeesa.github.io/UnitCommitment.jl/0.3/api/)
 
 ## Authors
 * **Alinson S. Xavier** (Argonne National Laboratory)
 * **Aleksandr M. Kazachkov** (University of Florida)
+* **Ogün Yurdakul** (Technische Universität Berlin)
+* **Jun He** (Purdue University)
 * **Feng Qiu** (Argonne National Laboratory)
 
 ## Acknowledgments
 
-* We would like to **Yonghong Chen** (Midcontinent Independent System Operator), **Feng Pan** (Pacific Northwest National Laboratory) for valuable feedback on early versions of this package.
+* We would like to thank **Yonghong Chen** (Midcontinent Independent System Operator), **Feng Pan** (Pacific Northwest National Laboratory) for valuable feedback on early versions of this package.
 
 * Based upon work supported by **Laboratory Directed Research and Development** (LDRD) funding from Argonne National Laboratory, provided by the Director, Office of Science, of the U.S. Department of Energy under Contract No. DE-AC02-06CH11357
 
@@ -109,15 +112,15 @@ UnitCommitment.write("/tmp/output.json", solution)
 
 If you use UnitCommitment.jl in your research (instances, models or algorithms), we kindly request that you cite the package as follows:
 
-* **Alinson S. Xavier, Aleksandr M. Kazachkov, Feng Qiu**. "UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment". Zenodo (2020). [DOI: 10.5281/zenodo.4269874](https://doi.org/10.5281/zenodo.4269874).
+* **Alinson S. Xavier, Aleksandr M. Kazachkov, Ogün Yurdakul, Feng Qiu**. "UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment (Version 0.3)". Zenodo (2022). [DOI: 10.5281/zenodo.4269874](https://doi.org/10.5281/zenodo.4269874).
 
-If you use the instances, we additionally request that you cite the original sources, as described in the [instances page](docs/instances.md).
+If you use the instances, we additionally request that you cite the original sources, as described in the documentation.
 
 ## License
 
 ```text
 UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment
-Copyright © 2020-2021, UChicago Argonne, LLC. All Rights Reserved.
+Copyright © 2020-2022, UChicago Argonne, LLC. All Rights Reserved.
 
 Redistribution and use in source and binary forms, with or without modification, are permitted
 provided that the following conditions are met:

benchmark/Project.toml

@@ -1,4 +1,5 @@
 [deps]
+DocOpt = "968ba79b-81e4-546f-ab3a-2eecfa62a9db"
 Gurobi = "2e9cd046-0924-5485-92f1-d5272153d98b"
 JSON = "682c06a0-de6a-54ab-a142-c8b1cf79cde6"
 JuMP = "4076af6c-e467-56ae-b986-b466b2749572"

benchmark/benchmark.jl

@@ -1,158 +0,0 @@
-# UnitCommitment.jl: Optimization Package for Security-Constrained Unit Commitment
-# Copyright (C) 2020, UChicago Argonne, LLC. All rights reserved.
-# Released under the modified BSD license. See COPYING.md for more details.
-
-using Distributed
-using Pkg
-Pkg.activate(".")
-
-@everywhere using Pkg
-@everywhere Pkg.activate(".")
-
-@everywhere using UnitCommitment
-@everywhere using JuMP
-@everywhere using Gurobi
-@everywhere using JSON
-@everywhere using Logging
-@everywhere using Printf
-@everywhere using LinearAlgebra
-@everywhere using Random
-
-@everywhere import UnitCommitment:
-    ArrCon2000,
-    CarArr2006,
-    DamKucRajAta2016,
-    Formulation,
-    Gar1962,
-    KnuOstWat2018,
-    MorLatRam2013,
-    PanGua2016,
-    XavQiuWanThi2019
-
-@everywhere UnitCommitment._setup_logger()
-
-function main()
-    cases = [
-        "pglib-uc/ca/2014-09-01_reserves_0",
-        "pglib-uc/ca/2014-09-01_reserves_1",
-        "pglib-uc/ca/2015-03-01_reserves_0",
-        "pglib-uc/ca/2015-06-01_reserves_0",
-        "pglib-uc/ca/Scenario400_reserves_1",
-        "pglib-uc/ferc/2015-01-01_lw",
-        "pglib-uc/ferc/2015-05-01_lw",
-        "pglib-uc/ferc/2015-07-01_hw",
-        "pglib-uc/ferc/2015-10-01_lw",
-        "pglib-uc/ferc/2015-12-01_lw",
-        "pglib-uc/rts_gmlc/2020-04-03",
-        "pglib-uc/rts_gmlc/2020-09-20",
-        "pglib-uc/rts_gmlc/2020-10-27",
-        "pglib-uc/rts_gmlc/2020-11-25",
-        "pglib-uc/rts_gmlc/2020-12-23",
-        "or-lib/20_0_1_w",
-        "or-lib/20_0_5_w",
-        "or-lib/50_0_2_w",
-        "or-lib/75_0_2_w",
-        "or-lib/100_0_1_w",
-        "or-lib/100_0_4_w",
-        "or-lib/100_0_5_w",
-        "or-lib/200_0_3_w",
-        "or-lib/200_0_7_w",
-        "or-lib/200_0_9_w",
-        "tejada19/UC_24h_290g",
-        "tejada19/UC_24h_623g",
-        "tejada19/UC_24h_959g",
-        "tejada19/UC_24h_1577g",
-        "tejada19/UC_24h_1888g",
-        "tejada19/UC_168h_72g",
-        "tejada19/UC_168h_86g",
-        "tejada19/UC_168h_130g",
-        "tejada19/UC_168h_131g",
-        "tejada19/UC_168h_199g",
-    ]
-    formulations = Dict(
-        "Default" => Formulation(),
-        "ArrCon2000" => Formulation(ramping = ArrCon2000.Ramping()),
-        "CarArr2006" => Formulation(pwl_costs = CarArr2006.PwlCosts()),
-        "DamKucRajAta2016" =>
-            Formulation(ramping = DamKucRajAta2016.Ramping()),
-        "Gar1962" => Formulation(pwl_costs = Gar1962.PwlCosts()),
-        "KnuOstWat2018" =>
-            Formulation(pwl_costs = KnuOstWat2018.PwlCosts()),
-        "MorLatRam2013" => Formulation(ramping = MorLatRam2013.Ramping()),
-        "PanGua2016" => Formulation(ramping = PanGua2016.Ramping()),
-    )
-    trials = [i for i in 1:5]
-    combinations = [
-        (c, f.first, f.second, t) for c in cases for f in formulations for
-        t in trials
-    ]
-    shuffle!(combinations)
-    @sync @distributed for c in combinations
-        _run_combination(c...)
-    end
-end
-
-@everywhere function _run_combination(
-    case,
-    formulation_name,
-    formulation,
-    trial,
-)
-    name = "$formulation_name/$case"
-    dirname = "results/$name"
-    mkpath(dirname)
-    if isfile("$dirname/$trial.json")
-        @info @sprintf("%-4s %-16s %s", "skip", formulation_name, case)
-        return
-    end
-    @info @sprintf("%-4s %-16s %s", "run", formulation_name, case)
-    open("$dirname/$trial.log", "w") do file
-        redirect_stdout(file) do
-            redirect_stderr(file) do
-                return _run_sample(case, formulation, "$dirname/$trial")
-            end
-        end
-    end
-    @info @sprintf("%-4s %-16s %s", "done", formulation_name, case)
-end
-
-@everywhere function _run_sample(case, formulation, prefix)
-    total_time = @elapsed begin
-        @info "Reading: $case"
-        time_read = @elapsed begin
-            instance = UnitCommitment.read_benchmark(case)
-        end
-        @info @sprintf("Read problem in %.2f seconds", time_read)
-        BLAS.set_num_threads(4)
-        model = UnitCommitment.build_model(
-            instance = instance,
-            formulation = formulation,
-            optimizer = optimizer_with_attributes(
-                Gurobi.Optimizer,
-                "Threads" => 4,
-                "Seed" => rand(1:1000),
-            ),
-            variable_names = true,
-        )
-        @info "Optimizing..."
-        BLAS.set_num_threads(1)
-        UnitCommitment.optimize!(
-            model,
-            XavQiuWanThi2019.Method(time_limit = 3600.0, gap_limit = 1e-4),
-        )
-    end
-    @info @sprintf("Total time was %.2f seconds", total_time)
-    @info "Writing solution: $prefix.json"
-    solution = UnitCommitment.solution(model)
-    UnitCommitment.write("$prefix.json", solution)
-    @info "Verifying solution..."
-    return UnitCommitment.validate(instance, solution)
-    # @info "Exporting model..."
-    # return JuMP.write_to_file(model, model_filename)
-end
-
-if length(ARGS) > 0
-    _run_sample(ARGS[1], UnitCommitment.Formulation(), "tmp")
-else
-    main()
-end

benchmark/run.jl

@@ -0,0 +1,209 @@
+# UnitCommitment.jl: Optimization Package for Security-Constrained Unit Commitment
+# Copyright (C) 2020, UChicago Argonne, LLC. All rights reserved.
+# Released under the modified BSD license. See COPYING.md for more details.
+
+doc = """UnitCommitment.jl Benchmark Runner
+
+Usage:
+  run.jl [-s ARG]... [-m ARG]... [-c ARG]... [-f ARG]... [options]
+
+Examples:
+
+    1. Benchmark all solvers, methods and formulations:
+
+        julia run.jl
+
+    2. Benchmark formulations "default" and "ArrCon200" using Gurobi:
+
+        julia run.jl -s gurobi -f default -f ArrCon2000
+
+    3. Benchmark a few test cases, using all solvers, methods and formulations:
+
+        julia run.jl -c or-lib/20_0_1_w -c matpower/case1888rte/2017-02-01
+
+    4. Solve 4 test cases in parallel, with 2 threads available per worker:
+
+        JULIA_NUM_THREADS=2 julia --procs 4 run.jl
+
+Options:
+  -h --help             Show this screen.
+  -s --solver=ARG       Mixed-integer linear solver (e.g. gurobi)
+  -c --case=ARG         Unit commitment test case (e.g. or-lib/20_0_1_w)
+  -m --method=ARG       Solution method (e.g. default)
+  -f --formulation=ARG  Formulation (e.g. ArrCon2000)
+  --time-limit=ARG      Time limit in seconds [default: 3600]
+  --gap=ARG             Relative MIP gap tolerance [default: 0.001]
+  --trials=ARG          Number of trials [default: 5]
+"""
+
+using Distributed
+using Pkg
+Pkg.activate(".")
+@everywhere using Pkg
+@everywhere Pkg.activate(".")
+
+using DocOpt
+args = docopt(doc)
+
+@everywhere using UnitCommitment
+@everywhere UnitCommitment._setup_logger()
+
+using UnitCommitment
+using Gurobi
+using Logging
+using JuMP
+
+import UnitCommitment:
+    ArrCon2000,
+    CarArr2006,
+    DamKucRajAta2016,
+    Formulation,
+    Gar1962,
+    KnuOstWat2018,
+    MorLatRam2013,
+    PanGua2016,
+    XavQiuWanThi2019
+
+# Benchmark test cases
+# -----------------------------------------------------------------------------
+cases = [
+    "pglib-uc/ca/2014-09-01_reserves_0",
+    "pglib-uc/ca/2014-09-01_reserves_1",
+    "pglib-uc/ca/2015-03-01_reserves_0",
+    "pglib-uc/ca/2015-06-01_reserves_0",
+    "pglib-uc/ca/Scenario400_reserves_1",
+    "pglib-uc/ferc/2015-01-01_lw",
+    "pglib-uc/ferc/2015-05-01_lw",
+    "pglib-uc/ferc/2015-07-01_hw",
+    "pglib-uc/ferc/2015-10-01_lw",
+    "pglib-uc/ferc/2015-12-01_lw",
+    "pglib-uc/rts_gmlc/2020-04-03",
+    "pglib-uc/rts_gmlc/2020-09-20",
+    "pglib-uc/rts_gmlc/2020-10-27",
+    "pglib-uc/rts_gmlc/2020-11-25",
+    "pglib-uc/rts_gmlc/2020-12-23",
+    "or-lib/20_0_1_w",
+    "or-lib/20_0_5_w",
+    "or-lib/50_0_2_w",
+    "or-lib/75_0_2_w",
+    "or-lib/100_0_1_w",
+    "or-lib/100_0_4_w",
+    "or-lib/100_0_5_w",
+    "or-lib/200_0_3_w",
+    "or-lib/200_0_7_w",
+    "or-lib/200_0_9_w",
+    "tejada19/UC_24h_290g",
+    "tejada19/UC_24h_623g",
+    "tejada19/UC_24h_959g",
+    "tejada19/UC_24h_1577g",
+    "tejada19/UC_24h_1888g",
+    "tejada19/UC_168h_72g",
+    "tejada19/UC_168h_86g",
+    "tejada19/UC_168h_130g",
+    "tejada19/UC_168h_131g",
+    "tejada19/UC_168h_199g",
+    "matpower/case1888rte/2017-02-01",
+    "matpower/case1951rte/2017-02-01",
+    "matpower/case2848rte/2017-02-01",
+    "matpower/case3012wp/2017-02-01",
+    "matpower/case3375wp/2017-02-01",
+    "matpower/case6468rte/2017-02-01",
+    "matpower/case6515rte/2017-02-01",
+]
+
+# Formulations
+# -----------------------------------------------------------------------------
+formulations = Dict(
+    "default" => Formulation(),
+    "ArrCon2000" => Formulation(ramping = ArrCon2000.Ramping()),
+    "CarArr2006" => Formulation(pwl_costs = CarArr2006.PwlCosts()),
+    "DamKucRajAta2016" => Formulation(ramping = DamKucRajAta2016.Ramping()),
+    "Gar1962" => Formulation(pwl_costs = Gar1962.PwlCosts()),
+    "KnuOstWat2018" => Formulation(pwl_costs = KnuOstWat2018.PwlCosts()),
+    "MorLatRam2013" => Formulation(ramping = MorLatRam2013.Ramping()),
+    "PanGua2016" => Formulation(ramping = PanGua2016.Ramping()),
+)
+
+# Solution methods
+# -----------------------------------------------------------------------------
+const gap_limit = parse(Float64, args["--gap"])
+const time_limit = parse(Float64, args["--time-limit"])
+methods = Dict(
+    "default" => XavQiuWanThi2019.Method(
+        time_limit = time_limit,
+        gap_limit = gap_limit,
+    ),
+)
+
+# MIP solvers
+# -----------------------------------------------------------------------------
+optimizers = Dict(
+    "gurobi" => optimizer_with_attributes(
+        Gurobi.Optimizer,
+        "Threads" => Threads.nthreads(),
+    ),
+)
+
+# Parse command line arguments
+# -----------------------------------------------------------------------------
+if !isempty(args["--case"])
+    cases = args["--case"]
+end
+if !isempty(args["--formulation"])
+    formulations = filter(p -> p.first in args["--formulation"], formulations)
+end
+if !isempty(args["--method"])
+    methods = filter(p -> p.first in args["--method"], methods)
+end
+if !isempty(args["--solver"])
+    optimizers = filter(p -> p.first in args["--solver"], optimizers)
+end
+const ntrials = parse(Int, args["--trials"])
+
+# Print benchmark settings
+# -----------------------------------------------------------------------------
+function printlist(d::Dict)
+    for key in keys(d)
+        @info "  - $key"
+    end
+end
+
+function printlist(d::Vector)
+    for key in d
+        @info "  - $key"
+    end
+end
+
+@info "Computational environment:"
+@info "  - CPU: $(Sys.cpu_info()[1].model)"
+@info "  - Logical CPU cores: $(length(Sys.cpu_info()))"
+@info "  - System memory: $(round(Sys.total_memory() / 2^30, digits=2)) GiB"
+@info "  - Available workers: $(nworkers())"
+@info "  - Available threads per worker: $(Threads.nthreads())"
+
+@info "Parameters:"
+@info "  - Number of trials: $ntrials"
+@info "  - Time limit (s): $time_limit"
+@info "  - Relative MIP gap tolerance: $gap_limit"
+
+@info "Solvers:"
+printlist(optimizers)
+
+@info "Methods:"
+printlist(methods)
+
+@info "Formulations:"
+printlist(formulations)
+
+@info "Cases:"
+printlist(cases)
+
+# Run benchmarks
+# -----------------------------------------------------------------------------
+UnitCommitment._run_benchmarks(
+    cases = cases,
+    formulations = formulations,
+    methods = methods,
+    optimizers = optimizers,
+    trials = 1:ntrials,
+)

docs/Makefile

@@ -1,14 +0,0 @@
-SPHINXOPTS    ?=
-SPHINXBUILD   ?= sphinx-build
-SOURCEDIR     = .
-BUILDDIR      = _build
-
-help:
-	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
-
-.PHONY: help Makefile
-
-# Catch-all target: route all unknown targets to Sphinx using the new
-# "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
-%: Makefile
-	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

docs/Project.toml

@@ -0,0 +1,5 @@
+[deps]
+Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
+JuMP = "4076af6c-e467-56ae-b986-b466b2749572"
+Revise = "295af30f-e4ad-537b-8983-00126c2a3abe"
+UnitCommitment = "64606440-39ea-11e9-0f29-3303a1d3d877"

docs/_static/custom.css

@@ -1,49 +0,0 @@
-h1.site-logo {
-    font-size: 30px !important;
-}
-
-h1.site-logo small {
-    font-size: 20px !important;
-}
-
-h1.site-logo {
-    font-size: 30px !important;
-}
-
-h1.site-logo small {
-    font-size: 20px !important;
-}
-
-tbody, thead, pre {
-    border: 1px solid rgba(0, 0, 0, 0.25);
-}
-
-table td, th {
-    padding: 8px;
-}
-
-table p {
-    margin-bottom: 0;
-}
-
-table td code {
-    white-space: nowrap;
-}
-
-table tr,
-table th {
-    border-bottom: 1px solid rgba(0, 0, 0, 0.1);
-}
-
-table tr:last-child {
-    border-bottom: 0;
-}
-
-pre {
-    box-shadow: inherit !important;
-    background-color: #fff;
-}
-
-.text-align\:center {
-    text-align: center;
-}

docs/conf.py

@@ -1,16 +0,0 @@
-project = "UnitCommitment.jl"
-copyright = "2020-2021, UChicago Argonne, LLC"
-author = ""
-release = "0.2"
-extensions = ["myst_parser"]
-templates_path = ["_templates"]
-exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"]
-html_theme = "sphinx_book_theme"
-html_static_path = ["_static"]
-html_css_files = ["custom.css"]
-html_theme_options = {
-    "repository_url": "https://github.com/ANL-CEEESA/UnitCommitment.jl/",
-    "use_repository_button": True,
-    "extra_navbar": "",
-}
-html_title = f"UnitCommitment.jl<br/><small>{release}</small>"

docs/make.jl

@@ -0,0 +1,16 @@
+using Documenter, UnitCommitment, JuMP
+
+makedocs(
+    sitename="UnitCommitment.jl",
+    pages=[
+        "Home" => "index.md",
+        "usage.md",
+        "format.md",
+        "instances.md",
+        "model.md",
+        "api.md",
+    ],
+    format = Documenter.HTML(
+        assets=["assets/custom.css"],
+    )
+)

docs/src/api.md

@@ -0,0 +1,62 @@
+# API Reference
+
+## Read data, build model & optimize
+
+```@docs
+UnitCommitment.read
+UnitCommitment.read_benchmark
+UnitCommitment.build_model
+UnitCommitment.optimize!
+UnitCommitment.solution
+UnitCommitment.validate
+UnitCommitment.write
+```
+
+## Locational Marginal Prices
+
+### Conventional LMPs
+```@docs
+UnitCommitment.compute_lmp(::JuMP.Model,::UnitCommitment.ConventionalLMP)
+```
+
+### Approximated Extended LMPs
+```@docs
+UnitCommitment.AELMP
+UnitCommitment.compute_lmp(::JuMP.Model,::UnitCommitment.AELMP)
+```
+
+
+## Modify instance
+
+```@docs
+UnitCommitment.slice
+UnitCommitment.randomize!(::UnitCommitment.UnitCommitmentInstance)
+UnitCommitment.generate_initial_conditions!
+```
+
+## Formulations
+
+```@docs
+UnitCommitment.Formulation
+UnitCommitment.ShiftFactorsFormulation
+UnitCommitment.ArrCon2000
+UnitCommitment.CarArr2006
+UnitCommitment.DamKucRajAta2016
+UnitCommitment.Gar1962
+UnitCommitment.KnuOstWat2018
+UnitCommitment.MorLatRam2013
+UnitCommitment.PanGua2016
+UnitCommitment.WanHob2016
+```
+
+## Solution Methods
+
+```@docs
+UnitCommitment.XavQiuWanThi2019.Method
+```
+
+## Randomization Methods
+
+```@docs
+UnitCommitment.XavQiuAhm2021.Randomization
+```

docs/src/assets/custom.css

@@ -0,0 +1,36 @@
+@media screen and (min-width: 1056px) {
+    #documenter .docs-main {
+      max-width: 65rem !important;
+    }
+}
+  
+
+tbody, thead, pre {
+    border: 1px solid rgba(0, 0, 0, 0.25);
+}
+
+table td, th {
+    padding: 8px;
+}
+
+table p {
+    margin-bottom: 0;
+}
+
+table td code {
+    white-space: nowrap;
+}
+
+table tr,
+table th {
+    border-bottom: 1px solid rgba(0, 0, 0, 0.1);
+}
+
+table tr:last-child {
+    border-bottom: 0;
+}
+
+code {
+    background-color: transparent;
+    color: rgb(232, 62, 140);
+}

docs/src/format.md

@@ -1,38 +1,29 @@
-```{sectnum}
----
-start: 2
-depth: 2
-suffix: .
----
-```
-
-
 Data Format
 ===========
 
-
 Input Data Format
 -----------------
 
 Instances are specified by JSON files containing the following main sections:
 
-* Parameters
-* Buses
-* Generators
-* Price-sensitive loads
-* Transmission lines
-* Reserves
-* Contingencies
+* [Parameters](#Parameters)
+* [Buses](#Buses)
+* [Generators](#Generators)
+* [Price-sensitive loads](#Price-sensitive-loads)
+* [Transmission lines](#Transmission-lines)
+* [Reserves](#Reserves)
+* [Contingencies](#Contingencies)
 
-Each section is described in detail below. For a complete example, see [case14](https://github.com/ANL-CEEESA/UnitCommitment.jl/tree/dev/instances/matpower/case14).
+Each section is described in detail below. See [case118/2017-01-01.json.gz](https://axavier.org/UnitCommitment.jl/0.3/instances/matpower/case118/2017-01-01.json.gz) for a complete example.
 
 ### Parameters
 
-This section describes system-wide parameters, such as power balance penalties,  optimization parameters, such as the length of the planning horizon and the time.
+This section describes system-wide parameters, such as power balance penalty, and optimization parameters, such as the length of the planning horizon and the time.
 
 | Key                            | Description                                       | Default  | Time series?
 | :----------------------------- | :------------------------------------------------ | :------: | :------------:
-| `Time horizon (h)`                     | Length of the planning horizon (in hours). | Required | N
+| `Version` | Version of UnitCommitment.jl this file was written for. Required to ensure that the file remains readable in future versions of the package. If you are following this page to construct the file, this field should equal `0.3`. | Required | N
+| `Time horizon (h)` | Length of the planning horizon (in hours). | Required | N
 | `Time step (min)` | Length of each time step (in minutes). Must be a divisor of 60 (e.g. 60, 30, 20, 15, etc). | `60` | N
 | `Power balance penalty ($/MW)` | Penalty for system-wide shortage or surplus in production (in $/MW). This is charged per time step. For example, if there is a shortage of 1 MW for three time steps, three times this amount will be charged. | `1000.0` | Y
 
@@ -41,6 +32,7 @@ This section describes system-wide parameters, such as power balance penalties,
 ```json
 {
     "Parameters": {
+        "Version": "0.3",
         "Time horizon (h)": 4,
         "Power balance penalty ($/MW)": 1000.0
     }
@@ -78,11 +70,17 @@ This section describes the characteristics of each bus in the system.
 
 ### Generators
 
-This section describes all generators in the system, including thermal units, renewable units and virtual units.
+This section describes all generators in the system. Two types of units can be specified:
+
+- **Thermal units:** Units that produce power by converting heat into electrical energy, such as coal and oil power plants. These units use a more complex model, with binary decision variables, and various constraints to enforce ramp rates and minimum up/down time.
+- **Profiled units:** Simplified model for units that do not require the constraints mentioned above, only a maximum and minimum power output for each time period. Typically used for renewables and hydro.
+
+#### Thermal Units
 
 | Key                       | Description                                      | Default | Time series?
 | :------------------------ | :------------------------------------------------| ------- | :-----------:
 | `Bus`                     | Identifier of the bus where this generator is located (string). | Required | N
+| `Type`                     | Type of the generator (string). For thermal generators, this must be `Thermal`.  | Required | N
 | `Production cost curve (MW)` and `Production cost curve ($)` | Parameters describing the piecewise-linear production costs. See below for more details. | Required | Y
 | `Startup costs ($)` and `Startup delays (h)` | Parameters describing how much it costs to start the generator after it has been shut down for a certain amount of time. If `Startup costs ($)` and `Startup delays (h)` are set to `[300.0, 400.0]` and `[1, 4]`, for example, and the generator is shut down at time `00:00` (h:min), then it costs \$300 to start up the generator at any time between `01:00` and `03:59`, and \$400 to start the generator at time `04:00` or any time after that.  The number of startup cost points is unlimited, and may be different for each generator. Startup delays must be strictly increasing and the first entry must equal `Minimum downtime (h)`. | `[0.0]` and `[1]` | N
 | `Minimum uptime (h)`      | Minimum amount of time the generator must stay operational after starting up (in hours). For example, if the generator starts up at time `00:00` (h:min) and `Minimum uptime (h)` is set to 4, then the generator can only shut down at time `04:00`. | `1` | N
@@ -94,18 +92,31 @@ This section describes all generators in the system, including thermal units, re
 | `Initial status (h)`  | If set to a positive number, indicates the amount of time (in hours) the generator has been on at the beginning of the simulation, and if set to a negative number, the amount of time the generator has been off. For example, if `Initial status (h)` is `-2`, this means that the generator was off since `-02:00` (h:min). The simulation starts at time `00:00`. If `Initial status (h)` is `3`, this means that the generator was on since `-03:00`. A value of zero is not acceptable. | Required | N
 | `Initial power (MW)`  | Amount of power the generator at time step `-1`, immediately before the planning horizon starts. | Required | N
 | `Must run?`               | If `true`, the generator should be committed, even if that is not economical (Boolean). | `false` | Y
-| `Provides spinning reserves?`    | If `true`, this generator may provide spinning reserves (Boolean). | `true` | Y
+| `Reserve eligibility` | List of reserve products this generator is eligibe to provide. By default, the generator is not eligible to provide any reserves. | `[]` | N
+| `Commitment status` | List of commitment status over the time horizon. At time `t`, if `true`, the generator must be commited at that time period; if `false`, the generator must not be commited at that time period. If `null` at time `t`, the generator's commitment status is then decided by the model. By default, the status is a list of `null` values. | `null` | Y
+
+#### Profiled Units
+
+| Key               | Description                                       | Default  | Time series?
+| :---------------- | :------------------------------------------------ | :------: | :------------:
+| `Bus`             | Identifier of the bus where this generator is located (string). | Required | N
+| `Type`            | Type of the generator (string). For profiled generators, this must be `Profiled`.  | Required | N
+| `Cost ($/MW)`     | Cost incurred for serving each MW of power by this generator. | Required | Y
+| `Minimum power (MW)` | Minimum amount of power this generator may supply. | `0.0` | Y
+| `Maximum power (MW)` | Maximum amount of power this generator may supply. | Required | Y
 
 #### Production costs and limits
 
 Production costs are represented as piecewise-linear curves. Figure 1 shows an example cost curve with three segments, where it costs \$1400, \$1600, \$2200 and \$2400 to generate, respectively, 100, 110, 130 and 135 MW of power. To model this generator, `Production cost curve (MW)` should be set to `[100, 110, 130, 135]`, and `Production cost curve ($)`  should be set to `[1400, 1600, 2200, 2400]`.
 Note that this curve also specifies the production limits. Specifically, the first point identifies the minimum power output when the unit is operational, while the last point identifies the maximum power output.
 
+```@raw html
 <center>
-    <img src="../_static/cost_curve.png" style="max-width: 500px"/>
+    <img src="../assets/cost_curve.png" style="max-width: 500px"/>
     <div><b>Figure 1.</b> Piecewise-linear production cost curve.</div>
     <br/>
 </center>
+```
 
 #### Additional remarks:
 
@@ -121,6 +132,7 @@ Note that this curve also specifies the production limits. Specifically, the fir
     "Generators": {
         "gen1": {
             "Bus": "b1",
+            "Type": "Thermal",
             "Production cost curve (MW)": [100.0, 110.0, 130.0, 135.0],
             "Production cost curve ($)": [1400.0, 1600.0, 2200.0, 2400.0],
             "Startup costs ($)": [300.0, 400.0],
@@ -132,14 +144,26 @@ Note that this curve also specifies the production limits. Specifically, the fir
             "Minimum downtime (h)": 4,
             "Minimum uptime (h)": 4,
             "Initial status (h)": 12,
+            "Initial power (MW)": 115,
             "Must run?": false,
-            "Provides spinning reserves?": true,
+            "Reserve eligibility": ["r1"]
         },
         "gen2": {
             "Bus": "b5",
+            "Type": "Thermal",
             "Production cost curve (MW)": [0.0, [10.0, 8.0, 0.0, 3.0]],
             "Production cost curve ($)": [0.0, 0.0],
-            "Provides spinning reserves?": true,
+            "Initial status (h)": -100,
+            "Initial power (MW)": 0,
+            "Reserve eligibility": ["r1", "r2"],
+            "Commitment status": [true, false, null, true]
+        },
+        "gen3": {
+            "Bus": "b6",
+            "Type": "Profiled",
+            "Minimum power (MW)": 10.0,
+            "Maximum power (MW)": 120.0,
+            "Cost ($/MW)": 100.0
         }
     }
 }
@@ -169,7 +193,7 @@ This section describes components in the system which may increase or reduce the
 }
 ```
 
-### Transmission Lines
+### Transmission lines
 
 This section describes the characteristics of transmission system, such as its topology and the susceptance of each transmission line.
 
@@ -204,24 +228,39 @@ This section describes the characteristics of transmission system, such as its t
 
 ### Reserves
 
-This section describes the hourly amount of operating reserves required.
+This section describes the hourly amount of reserves required.
 
 
 | Key                   | Description                                        | Default   |  Time series?
 | :-------------------- | :------------------------------------------------- | --------- |  :----:
-| `Spinning (MW)`       | Minimum amount of system-wide spinning reserves (in MW). Only generators which are online may provide this reserve. | `0.0` | Y
+| `Type` | Type of reserve product. Must be either "spinning" or "flexiramp". | Required | N
+| `Amount (MW)` | Amount of reserves required. | Required | Y
+| `Shortfall penalty ($/MW)` | Penalty for shortage in meeting the reserve requirements (in $/MW). This is charged per time step. Negative value implies reserve constraints must always be satisfied. | `-1` | Y
 
-#### Example
+#### Example 1
 
 ```json
 {
     "Reserves": {
-        "Spinning (MW)": [
-            57.30552,
-            53.88429,
-            51.31838,
-            50.46307
-        ]
+        "r1": {
+            "Type": "spinning",
+            "Amount (MW)": [
+                57.30552,
+                53.88429,
+                51.31838,
+                50.46307
+            ],
+            "Shortfall penalty ($/MW)": 5.0
+        },
+        "r2": {
+            "Type": "flexiramp",
+            "Amount (MW)": [
+                20.31042,
+                23.65273,
+                27.41784,
+                25.34057
+            ],
+        }
     }
 }
 ```
@@ -284,9 +323,8 @@ The output data format is also JSON-based, but it is not currently documented si
 Current limitations
 -------------------
 
-* All reserves are system-wide. Zonal reserves are not currently supported.
 * Network topology remains the same for all time periods
 * Only N-1 transmission contingencies are supported. Generator contingencies are not currently supported.
 * Time-varying minimum production amounts are not currently compatible with ramp/startup/shutdown limits.
-
+* Flexible ramping products can only be acquired under the `WanHob2016` formulation, which does not support spinning reserves. 
 

docs/src/index.md

@@ -6,24 +6,24 @@
 
 * **Data Format:** The package proposes an extensible and fully-documented JSON-based data specification format for SCUC, developed in collaboration with Independent System Operators (ISOs), which describes the most important aspects of the problem. The format supports all the most common generator characteristics (including ramping, piecewise-linear production cost curves and time-dependent startup costs), as well as operating reserves, price-sensitive loads, transmission networks and contingencies.
 * **Benchmark Instances:** The package provides a diverse collection of large-scale benchmark instances collected from the literature, converted into a common data format, and extended using data-driven methods to make them more challenging and realistic.
-* **Model Implementation**: The package provides a Julia/JuMP implementations of state-of-the-art formulations and solution methods for SCUC, including multiple ramping formulations ([ArrCon2000][ArrCon2000], [MorLatRam2013][MorLatRam2013], [DamKucRajAta2016][DamKucRajAta2016], [PanGua2016][PanGua2016]), multiple piecewise-linear costs formulations ([Gar1962][Gar1962], [CarArr2006][CarArr2006], [KnuOstWat2018][KnuOstWat2018]) and contingency screening methods ([XavQiuWanThi2019][XavQiuWanThi2019]). Our goal is to keep these implementations up-to-date as new methods are proposed in the literature.
+* **Model Implementation**: The package provides a Julia/JuMP implementations of state-of-the-art formulations and solution methods for SCUC, including multiple ramping formulations ([ArrCon2000](https://doi.org/10.1109/59.871739), [MorLatRam2013](https://doi.org/10.1109/TPWRS.2013.2251373), [DamKucRajAta2016](https://doi.org/10.1007/s10107-015-0919-9), [PanGua2016](https://doi.org/10.1287/opre.2016.1520)), multiple piecewise-linear costs formulations ([Gar1962](https://doi.org/10.1109/AIEEPAS.1962.4501405), [CarArr2006](https://doi.org/10.1109/TPWRS.2006.876672), [KnuOstWat2018](https://doi.org/10.1109/TPWRS.2017.2783850)) and contingency screening methods ([XavQiuWanThi2019](https://doi.org/10.1109/TPWRS.2019.2892620)). Our goal is to keep these implementations up-to-date as new methods are proposed in the literature.
 * **Benchmark Tools:** The package provides automated benchmark scripts to accurately evaluate the performance impact of proposed code changes.
 
-[ArrCon2000]: https://doi.org/10.1109/59.871739
-[CarArr2006]: https://doi.org/10.1109/TPWRS.2006.876672
-[DamKucRajAta2016]: https://doi.org/10.1007/s10107-015-0919-9
-[Gar1962]: https://doi.org/10.1109/AIEEPAS.1962.4501405
-[KnuOstWat2018]: https://doi.org/10.1109/TPWRS.2017.2783850
-[MorLatRam2013]: https://doi.org/10.1109/TPWRS.2013.2251373
-[PanGua2016]: https://doi.org/10.1287/opre.2016.1520
-[XavQiuWanThi2019]: https://doi.org/10.1109/TPWRS.2019.2892620
+## Table of Contents
 
-### Authors
+```@contents
+Pages = ["usage.md", "format.md", "instances.md", "model.md", "api.md"]
+Depth = 3
+```
+
+## Authors
 * **Alinson S. Xavier** (Argonne National Laboratory)
 * **Aleksandr M. Kazachkov** (University of Florida)
+* **Ogün Yurdakul** (Technische Universität Berlin)
+* **Jun He** (Purdue University)
 * **Feng Qiu** (Argonne National Laboratory)
 
-### Acknowledgments
+## Acknowledgments
 
 * We would like to thank **Yonghong Chen** (Midcontinent Independent System Operator), **Feng Pan** (Pacific Northwest National Laboratory) for valuable feedback on early versions of this package.
 
@@ -31,19 +31,19 @@
 
 * Based upon work supported by the **U.S. Department of Energy Advanced Grid Modeling Program** under Grant DE-OE0000875.
 
-### Citing
+## Citing
 
 If you use UnitCommitment.jl in your research (instances, models or algorithms), we kindly request that you cite the package as follows:
 
-* **Alinson S. Xavier, Aleksandr M. Kazachkov, Feng Qiu**, "UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment". Zenodo (2020). [DOI: 10.5281/zenodo.4269874](https://doi.org/10.5281/zenodo.4269874).
+* **Alinson S. Xavier, Aleksandr M. Kazachkov, Ogün Yurdakul, Feng Qiu**, "UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment (Version 0.3)". Zenodo (2022). [DOI: 10.5281/zenodo.4269874](https://doi.org/10.5281/zenodo.4269874).
 
 If you use the instances, we additionally request that you cite the original sources, as described in the [instances page](instances.md).
 
-### License
+## License
 
 ```text
 UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment
-Copyright © 2020, UChicago Argonne, LLC. All Rights Reserved.
+Copyright © 2020-2022, UChicago Argonne, LLC. All Rights Reserved.
 
 Redistribution and use in source and binary forms, with or without modification, are permitted
 provided that the following conditions are met:
@@ -67,16 +67,3 @@ THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING N
 OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 POSSIBILITY OF SUCH DAMAGE.
 ```
-
-## Site contents
-
-```{toctree}
----
-maxdepth: 2
----
-usage.md
-format.md
-instances.md
-model.md
-```
-

docs/src/instances.md

@@ -1,20 +1,11 @@
-```{sectnum}
----
-start: 3
-depth: 2
-suffix: .
----
-```
-
 Instances
 =========
 
-UnitCommitment.jl provides a large collection of benchmark instances collected
-from the literature and converted to a [common data format](format.md). In some cases, as indicated below, the original instances have been extended, with realistic parameters, using data-driven methods. 
-If you use these instances in your research, we request that you cite UnitCommitment.jl, as well as the original sources.
+UnitCommitment.jl provides a large collection of benchmark instances collected from the literature and converted to a [common data format](format.md). In some cases, as indicated below, the original instances have been extended, with realistic parameters, using data-driven methods. If you use these instances in your research, we request that you cite UnitCommitment.jl, as well as the original sources, as listed below. Benchmark instances can be loaded with `UnitCommitment.read_benchmark(name)`, as explained in the [usage section](usage.md). Instance files can also be [directly downloaded from our website](https://axavier.org/UnitCommitment.jl/0.3/instances/).
 
-Raw instances files are [available at our GitHub repository](https://github.com/ANL-CEEESA/UnitCommitment.jl/tree/dev/instances). Benchmark instances can also be loaded with
-`UnitCommitment.read_benchmark(name)`, as explained in the [usage section](usage.md).
+!!! warning
+
+    The instances included in UC.jl are still under development and may change in the future. If you use these instances in your research, for reproducibility, you should specify what version of UC.jl they came from.
 
 
 MATPOWER
@@ -34,7 +25,7 @@ Because most MATPOWER test cases were originally designed for power flow studies
 
 * **Contingencies** were set to include all N-1 transmission line contingencies that do not generate islands or isolated buses. More specifically, there is one contingency for each transmission line, as long as that transmission line is not a bridge in the network graph.
 
-For each MATPOWER test case, UC.jl provides two variations (`2017-02-01` and `2017-08-01`) corresponding respectively to a winter and to a summer test case.
+For each MATPOWER test case, UC.jl provides 365 variations (`2017-01-01` to `2017-12-31`) corresponding different days of the year.
 
 ### MATPOWER/UW-PSTCA
 
@@ -42,16 +33,11 @@ A variety of smaller IEEE test cases, [compiled by University of Washington](htt
 
 | Name | Buses | Generators | Lines | Contingencies | References |
 |------|-------|------------|-------|---------------|--------|
-| `matpower/case14/2017-02-01` | 14 | 5 | 20 | 19 | [MTPWR, PSTCA]
-| `matpower/case14/2017-08-01` | 14 | 5 | 20 | 19 | [MTPWR, PSTCA]
-| `matpower/case30/2017-02-01` | 30 | 6 | 41 | 38 | [MTPWR, PSTCA]
-| `matpower/case30/2017-08-01` | 30 | 6 | 41 | 38 | [MTPWR, PSTCA]
-| `matpower/case57/2017-02-01` | 57 | 7 | 80 | 79 | [MTPWR, PSTCA]
-| `matpower/case57/2017-08-01` | 57 | 7 | 80 | 79 | [MTPWR, PSTCA]
-| `matpower/case118/2017-02-01` | 118 | 54 | 186 | 177 | [MTPWR, PSTCA]
-| `matpower/case118/2017-08-01` | 118 | 54 | 186 | 177 | [MTPWR, PSTCA]
-| `matpower/case300/2017-02-01` | 300 | 69 | 411 | 320 | [MTPWR, PSTCA]
-| `matpower/case300/2017-08-01` | 300 | 69 | 411 | 320 | [MTPWR, PSTCA]
+| `matpower/case14/2017-01-01` | 14 | 5 | 20 | 19 | [MTPWR, PSTCA]
+| `matpower/case30/2017-01-01` | 30 | 6 | 41 | 38 | [MTPWR, PSTCA]
+| `matpower/case57/2017-01-01` | 57 | 7 | 80 | 79 | [MTPWR, PSTCA]
+| `matpower/case118/2017-01-01` | 118 | 54 | 186 | 177 | [MTPWR, PSTCA]
+| `matpower/case300/2017-01-01` | 300 | 69 | 411 | 320 | [MTPWR, PSTCA]
 
 
 ### MATPOWER/Polish
@@ -60,22 +46,14 @@ Test cases based on the Polish 400, 220 and 110 kV networks, originally provided
 
 | Name | Buses | Generators | Lines | Contingencies | References |
 |------|-------|------------|-------|---------------|--------|
-| `matpower/case2383wp/2017-02-01` | 2383 | 323 | 2896 | 2240 | [MTPWR]
-| `matpower/case2383wp/2017-08-01` | 2383 | 323 | 2896 | 2240 | [MTPWR]
-| `matpower/case2736sp/2017-02-01` | 2736 | 289 | 3504 | 3159 | [MTPWR]
-| `matpower/case2736sp/2017-08-01` | 2736 | 289 | 3504 | 3159 | [MTPWR]
-| `matpower/case2737sop/2017-02-01` | 2737 | 267 | 3506 | 3161 | [MTPWR]
-| `matpower/case2737sop/2017-08-01` | 2737 | 267 | 3506 | 3161 | [MTPWR]
-| `matpower/case2746wop/2017-02-01` | 2746 | 443 | 3514 | 3155 | [MTPWR]
-| `matpower/case2746wop/2017-08-01` | 2746 | 443 | 3514 | 3155 | [MTPWR]
-| `matpower/case2746wp/2017-02-01` | 2746 | 457 | 3514 | 3156 | [MTPWR]
-| `matpower/case2746wp/2017-08-01` | 2746 | 457 | 3514 | 3156 | [MTPWR]
-| `matpower/case3012wp/2017-02-01` | 3012 | 496 | 3572 | 2854 | [MTPWR]
-| `matpower/case3012wp/2017-08-01` | 3012 | 496 | 3572 | 2854 | [MTPWR]
-| `matpower/case3120sp/2017-02-01` | 3120 | 483 | 3693 | 2950 | [MTPWR]
-| `matpower/case3120sp/2017-08-01` | 3120 | 483 | 3693 | 2950 | [MTPWR]
-| `matpower/case3375wp/2017-02-01` | 3374 | 590 | 4161 | 3245 | [MTPWR]
-| `matpower/case3375wp/2017-08-01` | 3374 | 590 | 4161 | 3245 | [MTPWR]
+| `matpower/case2383wp/2017-01-01` | 2383 | 323 | 2896 | 2240 | [MTPWR]
+| `matpower/case2736sp/2017-01-01` | 2736 | 289 | 3504 | 3159 | [MTPWR]
+| `matpower/case2737sop/2017-01-01` | 2737 | 267 | 3506 | 3161 | [MTPWR]
+| `matpower/case2746wop/2017-01-01` | 2746 | 443 | 3514 | 3155 | [MTPWR]
+| `matpower/case2746wp/2017-01-01` | 2746 | 457 | 3514 | 3156 | [MTPWR]
+| `matpower/case3012wp/2017-01-01` | 3012 | 496 | 3572 | 2854 | [MTPWR]
+| `matpower/case3120sp/2017-01-01` | 3120 | 483 | 3693 | 2950 | [MTPWR]
+| `matpower/case3375wp/2017-01-01` | 3374 | 590 | 4161 | 3245 | [MTPWR]
 
 ### MATPOWER/PEGASE
 
@@ -83,16 +61,11 @@ Test cases from the [Pan European Grid Advanced Simulation and State Estimation
 
 | Name | Buses | Generators | Lines | Contingencies | References |
 |------|-------|------------|-------|---------------|--------|
-| `matpower/case89pegase/2017-02-01` | 89 | 12 | 210 | 192 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case89pegase/2017-08-01` | 89 | 12 | 210 | 192 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case1354pegase/2017-02-01` | 1354 | 260 | 1991 | 1288 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case1354pegase/2017-08-01` | 1354 | 260 | 1991 | 1288 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case2869pegase/2017-02-01` | 2869 | 510 | 4582 | 3579 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case2869pegase/2017-08-01` | 2869 | 510 | 4582 | 3579 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case9241pegase/2017-02-01` | 9241 | 1445 | 16049 | 13932 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case9241pegase/2017-08-01` | 9241 | 1445 | 16049 | 13932 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case13659pegase/2017-02-01` | 13659 | 4092 | 20467 | 13932 | [JoFlMa16, FlPaCa13, MTPWR]
-| `matpower/case13659pegase/2017-08-01` | 13659 | 4092 | 20467 | 13932 | [JoFlMa16, FlPaCa13, MTPWR]
+| `matpower/case89pegase/2017-01-01` | 89 | 12 | 210 | 192 | [JoFlMa16, FlPaCa13, MTPWR]
+| `matpower/case1354pegase/2017-01-01` | 1354 | 260 | 1991 | 1288 | [JoFlMa16, FlPaCa13, MTPWR]
+| `matpower/case2869pegase/2017-01-01` | 2869 | 510 | 4582 | 3579 | [JoFlMa16, FlPaCa13, MTPWR]
+| `matpower/case9241pegase/2017-01-01` | 9241 | 1445 | 16049 | 13932 | [JoFlMa16, FlPaCa13, MTPWR]
+| `matpower/case13659pegase/2017-01-01` | 13659 | 4092 | 20467 | 13932 | [JoFlMa16, FlPaCa13, MTPWR]
 
 ### MATPOWER/RTE
 
@@ -100,22 +73,14 @@ Test cases from the R&D Division at [Reseau de Transport d'Electricite](https://
 
 | Name | Buses | Generators | Lines | Contingencies | References |
 |------|-------|------------|-------|---------------|--------|
-| `matpower/case1888rte/2017-02-01` | 1888 | 296 | 2531 | 1484 | [MTPWR, JoFlMa16]
-| `matpower/case1888rte/2017-08-01` | 1888 | 296 | 2531 | 1484 | [MTPWR, JoFlMa16]
-| `matpower/case1951rte/2017-02-01` | 1951 | 390 | 2596 | 1497 | [MTPWR, JoFlMa16]
-| `matpower/case1951rte/2017-08-01` | 1951 | 390 | 2596 | 1497 | [MTPWR, JoFlMa16]
-| `matpower/case2848rte/2017-02-01` | 2848 | 544 | 3776 | 2242 | [MTPWR, JoFlMa16]
-| `matpower/case2848rte/2017-08-01` | 2848 | 544 | 3776 | 2242 | [MTPWR, JoFlMa16]
-| `matpower/case2868rte/2017-02-01` | 2868 | 596 | 3808 | 2260 | [MTPWR, JoFlMa16]
-| `matpower/case2868rte/2017-08-01` | 2868 | 596 | 3808 | 2260 | [MTPWR, JoFlMa16]
-| `matpower/case6468rte/2017-02-01` | 6468 | 1262 | 9000 | 6094 | [MTPWR, JoFlMa16]
-| `matpower/case6468rte/2017-08-01` | 6468 | 1262 | 9000 | 6094 | [MTPWR, JoFlMa16]
-| `matpower/case6470rte/2017-02-01` | 6470 | 1306 | 9005 | 6085 | [MTPWR, JoFlMa16]
-| `matpower/case6470rte/2017-08-01` | 6470 | 1306 | 9005 | 6085 | [MTPWR, JoFlMa16]
-| `matpower/case6495rte/2017-02-01` | 6495 | 1352 | 9019 | 6060 | [MTPWR, JoFlMa16]
-| `matpower/case6495rte/2017-08-01` | 6495 | 1352 | 9019 | 6060 | [MTPWR, JoFlMa16]
-| `matpower/case6515rte/2017-02-01` | 6515 | 1368 | 9037 | 6063 | [MTPWR, JoFlMa16]
-| `matpower/case6515rte/2017-08-01` | 6515 | 1368 | 9037 | 6063 | [MTPWR, JoFlMa16]
+| `matpower/case1888rte/2017-01-01` | 1888 | 296 | 2531 | 1484 | [MTPWR, JoFlMa16]
+| `matpower/case1951rte/2017-01-01` | 1951 | 390 | 2596 | 1497 | [MTPWR, JoFlMa16]
+| `matpower/case2848rte/2017-01-01` | 2848 | 544 | 3776 | 2242 | [MTPWR, JoFlMa16]
+| `matpower/case2868rte/2017-01-01` | 2868 | 596 | 3808 | 2260 | [MTPWR, JoFlMa16]
+| `matpower/case6468rte/2017-01-01` | 6468 | 1262 | 9000 | 6094 | [MTPWR, JoFlMa16]
+| `matpower/case6470rte/2017-01-01` | 6470 | 1306 | 9005 | 6085 | [MTPWR, JoFlMa16]
+| `matpower/case6495rte/2017-01-01` | 6495 | 1352 | 9019 | 6060 | [MTPWR, JoFlMa16]
+| `matpower/case6515rte/2017-01-01` | 6515 | 1368 | 9037 | 6063 | [MTPWR, JoFlMa16]
 
 
 PGLIB-UC Instances
@@ -315,7 +280,7 @@ Tejada19
 References
 ----------
 
-* [UCJL] **Alinson S. Xavier, Aleksandr M. Kazachkov, Feng Qiu.** "UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment". Zenodo (2020). [DOI: 10.5281/zenodo.4269874](https://doi.org/10.5281/zenodo.4269874)
+* [UCJL] **Alinson S. Xavier, Aleksandr M. Kazachkov, Ogün Yurdakul, Feng Qiu.** "UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment (Version 0.3)". Zenodo (2022). [DOI: 10.5281/zenodo.4269874](https://doi.org/10.5281/zenodo.4269874)
 
 * [KnOsWa20] **Bernard Knueven, James Ostrowski and Jean-Paul Watson.** "On Mixed-Integer Programming Formulations for the Unit Commitment Problem". INFORMS Journal on Computing (2020). [DOI: 10.1287/ijoc.2019.0944](https://doi.org/10.1287/ijoc.2019.0944)
 
@@ -323,14 +288,9 @@ References
 
 * [BaBlEh19] **Clayton Barrows, Aaron Bloom, Ali Ehlen, Jussi Ikaheimo, Jennie Jorgenson, Dheepak Krishnamurthy, Jessica Lau et al.** "The IEEE Reliability Test System: A Proposed 2019 Update." IEEE Transactions on Power Systems (2019). [DOI: 10.1109/TPWRS.2019.2925557](https://doi.org/10.1109/TPWRS.2019.2925557)
 
-* [JoFlMa16] **C. Josz, S. Fliscounakis, J. Maeght, and P. Panciatici.** "AC Power Flow
-Data in MATPOWER and QCQP Format: iTesla, RTE Snapshots, and PEGASE". [ArXiv (2016)](https://arxiv.org/abs/1603.01533).
+* [JoFlMa16] **C. Josz, S. Fliscounakis, J. Maeght, and P. Panciatici.** "AC Power Flow Data in MATPOWER and QCQP Format: iTesla, RTE Snapshots, and PEGASE". [ArXiv (2016)](https://arxiv.org/abs/1603.01533).
 
-* [FlPaCa13] **S. Fliscounakis, P. Panciatici, F. Capitanescu, and L. Wehenkel.**
-"Contingency ranking with respect to overloads in very large power
-systems taking into account uncertainty, preventive and corrective
-actions", Power Systems, IEEE Trans. on, (28)4:4909-4917, 2013.
-[DOI: 10.1109/TPWRS.2013.2251015](https://doi.org/10.1109/TPWRS.2013.2251015)
+* [FlPaCa13] **S. Fliscounakis, P. Panciatici, F. Capitanescu, and L. Wehenkel.** "Contingency ranking with respect to overloads in very large power systems taking into account uncertainty, preventive and corrective actions", Power Systems, IEEE Trans. on, (28)4:4909-4917, 2013. [DOI: 10.1109/TPWRS.2013.2251015](https://doi.org/10.1109/TPWRS.2013.2251015)
 
 * [MTPWR] **D. Zimmerman, C. E. Murillo-Sandnchez and R. J. Thomas.** "Matpower:  Steady-state  operations,  planning,  and  analysis  tools  forpower systems research and education", IEEE Transactions on PowerSystems, vol. 26, no. 1, pp. 12 –19, Feb. 2011. [DOI: 10.1109/TPWRS.2010.2051168](https://doi.org/10.1109/TPWRS.2010.2051168)
 

docs/src/model.md

@@ -1,11 +1,3 @@
-```{sectnum}
----
-start: 4
-depth: 2
-suffix: .
----
-```
-
 JuMP Model
 ==========
 
@@ -16,21 +8,30 @@ Decision variables
 
 ### Generators
 
+#### Thermal Units
+
 Name | Symbol | Description | Unit
------|:--------:|-------------|:------:
+:-----|:--------:|:-------------|:------:
 `is_on[g,t]` | $u_{g}(t)$ | True if generator `g` is on at time `t`. | Binary
 `switch_on[g,t]` | $v_{g}(t)$ | True is generator `g` switches on at time `t`. | Binary
 `switch_off[g,t]` | $w_{g}(t)$ | True if generator `g` switches off at time `t`. | Binary
 `prod_above[g,t]` |$p'_{g}(t)$ | Amount of power produced by generator `g` above its minimum power output at time `t`. For example, if the minimum power of generator `g` is 100 MW and `g` is producing 115 MW of power at time `t`, then `prod_above[g,t]` equals `15.0`. | MW
 `segprod[g,t,k]` | $p^k_g(t)$ | Amount of power from piecewise linear segment `k` produced by generator `g` at time `t`. For example, if cost curve for generator `g` is defined by the points `(100, 1400)`, `(110, 1600)`, `(130, 2200)` and `(135, 2400)`, and if the generator is producing 115 MW of power at time `t`, then `segprod[g,t,:]` equals `[10.0, 5.0, 0.0]`.| MW
-`reserve[g,t]` | $r_g(t)$ | Amount of reserves provided by generator `g` at time `t`. | MW
+`reserve[r,g,t]` | $r_g(t)$ | Amount of reserve `r` provided by unit `g` at time `t`. | MW
 `startup[g,t,s]` | $\delta^s_g(t)$ | True if generator `g` switches on at time `t` incurring start-up costs from start-up category `s`. | Binary
 
 
+#### Profiled Units
+
+Name | Symbol | Description | Unit
+:-----|:------:|:-------------|:------:
+`prod_profiled[s,t]` | $p^{\dagger}_{g}(t)$ | Amount of power produced by profiled unit `g` at time `t`. | MW
+
+
 ### Buses
 
 Name | Symbol | Description | Unit
------|:------:|-------------|:------:
+:-----|:------:|:-------------|:------:
 `net_injection[b,t]` | $n_b(t)$ | Net injection at bus `b` at time `t`. | MW
 `curtail[b,t]` | $s^+_b(t)$ | Amount of load curtailed at bus `b` at time `t` | MW
 
@@ -38,69 +39,24 @@ Name | Symbol | Description | Unit
 ### Price-sensitive loads
 
 Name | Symbol | Description | Unit
------|:------:|-------------|:------:
+:-----|:------:|:-------------|:------:
 `loads[s,t]` | $d_{s}(t)$ | Amount of power served to price-sensitive load `s` at time `t`. | MW
 
 ### Transmission lines
 
 Name | Symbol | Description | Unit
------|:------:|-------------|:------:
+:-----|:------:|:-------------|:------:
 `flow[l,t]` | $f_l(t)$ | Power flow on line `l` at time `t`. | MW
 `overflow[l,t]` | $f^+_l(t)$ | Amount of flow above the limit for line `l` at time `t`. | MW
 
-```{warning}
+!!! warning
 
-Since transmission and N-1 security constraints are enforced in a lazy way, most of the `flow[l,t]` variables are never added to the model. Accessing `model[:flow][l,t]` without first checking that the variable exists will likely generate an error.
-```
+    Since transmission and N-1 security constraints are enforced in a lazy way, most of the `flow[l,t]` variables are never added to the model. Accessing `model[:flow][l,t]` without first checking that the variable exists will likely generate an error.
 
 Objective function
 ------------------
 
-$$
-\begin{align}
-    \text{minimize} \;\; &
-        \sum_{t \in \mathcal{T}}
-          \sum_{g \in \mathcal{G}}
-          C^\text{min}_g(t) u_g(t) \\
-    &
-        + \sum_{t \in \mathcal{T}}
-          \sum_{g \in \mathcal{G}}
-          \sum_{g \in \mathcal{K}_g}
-          C^k_g(t) p^k_g(t) \\
-    &
-        + \sum_{t \in \mathcal{T}}
-          \sum_{g \in \mathcal{G}}
-          \sum_{s \in \mathcal{S}_g}
-          C^s_{g}(t) \delta^s_g(t) \\
-    &
-        + \sum_{t \in \mathcal{T}}
-          \sum_{l \in \mathcal{L}}
-          C^\text{overflow}_{l}(t) f^+_l(t) \\
-    &
-        + \sum_{t \in \mathcal{T}}
-           \sum_{b \in \mathcal{B}}
-           C^\text{curtail}(t) s^+_b(t) \\
-    &
-        - \sum_{t \in \mathcal{T}}
-           \sum_{s \in \mathcal{PS}}
-           R_{s}(t) d_{s}(t) \\
-          
-\end{align}
-$$
-where
-- $\mathcal{B}$ is the set of buses
-- $\mathcal{G}$ is the set of generators
-- $\mathcal{L}$ is the set of transmission lines
-- $\mathcal{PS}$ is the set of price-sensitive loads
-- $\mathcal{S}_g$ is the set of start-up categories for generator $g$
-- $\mathcal{T}$ is the set of time steps
-- $C^\text{curtail}(t)$ is the curtailment penalty (in \$/MW)
-- $C^\text{min}_g(t)$ is the cost of keeping generator $g$ on and producing at minimum power during time $t$ (in \$)
-- $C^\text{overflow}_{l}(t)$ is the flow limit penalty for line $l$ at time $t$ (in \$/MW)
-- $C^k_g(t)$ is the cost for generator $g$ to produce 1 MW of power at time $t$ under piecewise linear segment $k$
-- $C^s_{g}(t)$ is the cost of starting up generator $g$ at time $t$ under start-up category $s$ (in \$)
-- $R_{s}(t)$ is the revenue obtained from serving price-sensitive load $s$ at time $t$ (in \$/MW)
-
+TODO
 
 Constraints
 -----------

docs/src/usage.md

@@ -0,0 +1,226 @@
+Usage
+=====
+
+Installation
+------------
+
+UnitCommitment.jl was tested and developed with [Julia 1.7](https://julialang.org/). To install Julia, please follow the [installation guide on the official Julia website](https://julialang.org/downloads/). To install UnitCommitment.jl, run the Julia interpreter, type `]` to open the package manager, then type:
+
+```text
+pkg> add UnitCommitment@0.3
+```
+
+To test that the package has been correctly installed, run:
+
+```text
+pkg> test UnitCommitment
+```
+
+If all tests pass, the package should now be ready to be used by any Julia script on the machine.
+
+To solve the optimization models, a mixed-integer linear programming (MILP) solver is also required. Please see the [JuMP installation guide](https://jump.dev/JuMP.jl/stable/installation/) for more instructions on installing a solver. Typical open-source choices are [Cbc](https://github.com/JuliaOpt/Cbc.jl) and [GLPK](https://github.com/JuliaOpt/GLPK.jl). In the instructions below, Cbc will be used, but any other MILP solver listed in JuMP installation guide should also be compatible.
+
+Typical Usage
+-------------
+
+### Solving user-provided instances
+
+The first step to use UC.jl is to construct a JSON file describing your unit commitment instance. See [Data Format](format.md) for a complete description of the data format UC.jl expects. The next steps, as shown below, are to: (1) read the instance from file; (2) construct the optimization model; (3) run the optimization; and (4) extract the optimal solution.
+
+```julia
+using Cbc
+using JSON
+using UnitCommitment
+
+# 1. Read instance
+instance = UnitCommitment.read("/path/to/input.json")
+
+# 2. Construct optimization model
+model = UnitCommitment.build_model(
+    instance=instance,
+    optimizer=Cbc.Optimizer,
+)
+
+# 3. Solve model
+UnitCommitment.optimize!(model)
+
+# 4. Write solution to a file
+solution = UnitCommitment.solution(model)
+UnitCommitment.write("/path/to/output.json", solution)
+```
+
+### Solving benchmark instances
+
+UnitCommitment.jl contains a large number of benchmark instances collected from the literature and converted into a common data format. To solve one of these instances individually, instead of constructing your own, the function `read_benchmark` can be used, as shown below. See [Instances](instances.md) for the complete list of available instances.
+
+```julia
+using UnitCommitment
+instance = UnitCommitment.read_benchmark("matpower/case3375wp/2017-02-01")
+```
+
+## Customizing the formulation
+
+By default, `build_model` uses a formulation that combines modeling components from different publications, and that has been carefully tested, using our own benchmark scripts, to provide good performance across a wide variety of instances. This default formulation is expected to change over time, as new methods are proposed in the literature. You can, however, construct your own formulation, based on the modeling components that you choose, as shown in the next example.
+
+```julia
+using Cbc
+using UnitCommitment
+
+import UnitCommitment:
+    Formulation,
+    KnuOstWat2018,
+    MorLatRam2013,
+    ShiftFactorsFormulation
+
+instance = UnitCommitment.read_benchmark(
+    "matpower/case118/2017-02-01",
+)
+
+model = UnitCommitment.build_model(
+    instance = instance,
+    optimizer = Cbc.Optimizer,
+    formulation = Formulation(
+        pwl_costs = KnuOstWat2018.PwlCosts(),
+        ramping = MorLatRam2013.Ramping(),
+        startup_costs = MorLatRam2013.StartupCosts(),
+        transmission = ShiftFactorsFormulation(
+            isf_cutoff = 0.005,
+            lodf_cutoff = 0.001,
+        ),
+    ),
+)
+```
+
+## Generating initial conditions
+
+When creating random unit commitment instances for benchmark purposes, it is often hard to compute, in advance, sensible initial conditions for all generators. Setting initial conditions naively (for example, making all generators initially off and producing no power) can easily cause the instance to become infeasible due to excessive ramping. Initial conditions can also make it hard to modify existing instances. For example, increasing the system load without carefully modifying the initial conditions may make the problem infeasible or unrealistically challenging to solve.
+
+To help with this issue, UC.jl provides a utility function which can generate feasible initial conditions by solving a single-period optimization problem, as shown below:
+
+```julia
+using Cbc
+using UnitCommitment
+
+# Read original instance
+instance = UnitCommitment.read("instance.json")
+
+# Generate initial conditions (in-place)
+UnitCommitment.generate_initial_conditions!(instance, Cbc.Optimizer)
+
+# Construct and solve optimization model
+model = UnitCommitment.build_model(
+    instance=instance,
+    optimizer=Cbc.Optimizer,
+)
+UnitCommitment.optimize!(model)
+```
+
+!!! warning
+
+    The function `generate_initial_conditions!` may return different initial conditions after each call, even if the same instance and the same optimizer is provided. The particular algorithm may also change in a future version of UC.jl. For these reasons, it is recommended that you generate initial conditions exactly once for each instance and store them for later use.
+    
+## Verifying solutions
+
+When developing new formulations, it is very easy to introduce subtle errors in the model that result in incorrect solutions. To help with this, UC.jl includes a utility function that verifies if a given solution is feasible, and, if not, prints all the validation errors it found. The implementation of this function is completely independent from the implementation of the optimization model, and therefore can be used to validate it. The function can also be used to verify solutions produced by other optimization packages, as long as they follow the [UC.jl data format](format.md).
+
+```julia
+using JSON
+using UnitCommitment
+
+# Read instance
+instance = UnitCommitment.read("instance.json")
+
+# Read solution (potentially produced by other packages) 
+solution = JSON.parsefile("solution.json")
+
+# Validate solution and print validation errors
+UnitCommitment.validate(instance, solution)
+```
+
+## Computing Locational Marginal Prices
+
+Locational marginal prices (LMPs) refer to the cost of supplying electricity at a particular location of the network. Multiple methods for computing LMPs have been proposed in the literature. UnitCommitment.jl implements two commonly-used methods: conventional LMPs and Approximated Extended LMPs (AELMPs). To compute LMPs for a given unit commitment instance, the `compute_lmp` function can be used, as shown in the examples below. The function accepts three arguments -- a solved SCUC model, an LMP method, and a linear optimizer -- and it returns a dictionary mapping `(bus_name, time)` to the marginal price.
+
+
+!!! warning
+
+    Most mixed-integer linear optimizers, such as `HiGHS`, `Gurobi` and `CPLEX` can be used with `compute_lmp`, with the notable exception of `Cbc`, which does not support dual value evaluations. If using `Cbc`, please provide `Clp` as the linear optimizer.
+
+### Conventional LMPs
+
+LMPs are conventionally computed by: (1) solving the SCUC model, (2) fixing all binary variables to their optimal values, and (3) re-solving the resulting linear programming model. In this approach, the LMPs are defined as the dual variables' values associated with the net injection constraints. The example below shows how to compute conventional LMPs for a given unit commitment instance. First, we build and optimize the SCUC model. Then, we call the `compute_lmp` function, providing as the second argument `ConventionalLMP()`.
+
+
+```julia
+using UnitCommitment
+using HiGHS
+
+import UnitCommitment: ConventionalLMP
+
+# Read benchmark instance
+instance = UnitCommitment.read_benchmark("matpower/case118/2018-01-01")
+
+# Build the model
+model = UnitCommitment.build_model(
+    instance = instance,
+    optimizer = HiGHS.Optimizer,
+)
+
+# Optimize the model
+UnitCommitment.optimize!(model)
+
+# Compute the LMPs using the conventional method
+lmp = UnitCommitment.compute_lmp(
+    model,
+    ConventionalLMP(),
+    optimizer = HiGHS.Optimizer,
+)
+
+# Access the LMPs
+# Example: "s1" is the scenario name, "b1" is the bus name, 1 is the first time slot
+@show lmp["s1","b1", 1]
+```
+
+### Approximate Extended LMPs
+
+Approximate Extended LMPs (AELMPs) are an alternative method to calculate locational marginal prices which attemps to minimize uplift payments. The method internally works by modifying the instance data in three ways: (1) it sets the minimum power output of each generator to zero, (2) it averages the start-up cost over the offer blocks for each generator, and (3) it relaxes all integrality constraints. To compute AELMPs, as shown in the example below, we call `compute_lmp` and provide `AELMP()` as the second argument.
+
+This method has two configurable parameters: `allow_offline_participation` and `consider_startup_costs`. If `allow_offline_participation = true`, then offline generators are allowed to participate in the pricing. If instead `allow_offline_participation = false`, offline generators are not allowed and therefore are excluded from the system. A solved UC model is optional if offline participation is allowed, but is required if not allowed. The method forces offline participation to be allowed if the UC model supplied by the user is not solved. For the second field, If `consider_startup_costs = true`, then start-up costs are integrated and averaged over each unit production; otherwise the production costs stay the same. By default, both fields are set to `true`.
+
+!!! warning
+
+    This approximation method is still under active research, and has several limitations. The implementation provided in the package is based on MISO Phase I only. It only supports fast start resources. More specifically, the minimum up/down time of all generators must be 1, the initial power of all generators must be 0, and the initial status of all generators must be negative. The method does not support time-varying start-up costs. The method does not support multiple scenarios. If offline participation is not allowed, AELMPs treats an asset to be  offline if it is never on throughout all time periods. 
+
+```julia
+using UnitCommitment
+using HiGHS
+
+import UnitCommitment: AELMP
+
+# Read benchmark instance
+instance = UnitCommitment.read_benchmark("matpower/case118/2017-02-01")
+
+# Build the model
+model = UnitCommitment.build_model(
+    instance = instance,
+    optimizer = HiGHS.Optimizer,
+)
+
+# Optimize the model
+UnitCommitment.optimize!(model)
+
+# Compute the AELMPs
+aelmp = UnitCommitment.compute_lmp(
+    model,
+    AELMP(
+        allow_offline_participation = false,
+        consider_startup_costs = true
+    ),
+    optimizer = HiGHS.Optimizer
+)
+
+# Access the AELMPs
+# Example: "s1" is the scenario name, "b1" is the bus name, 1 is the first time slot
+# Note: although scenario is supported, the query still keeps the scenario keys for consistency.
+@show aelmp["s1", "b1", 1]
+```

docs/usage.md

@@ -1,149 +0,0 @@
-```{sectnum}
----
-start: 1
-depth: 2
-suffix: .
----
-```
-
-Usage
-=====
-
-Installation
-------------
-
-UnitCommitment.jl was tested and developed with [Julia 1.6](https://julialang.org/). To install Julia, please follow the [installation guide on the official Julia website](https://julialang.org/downloads/platform.html). To install UnitCommitment.jl, run the Julia interpreter, type `]` to open the package manager, then type:
-
-```text
-pkg> add UnitCommitment@0.2
-```
-
-To test that the package has been correctly installed, run:
-
-```text
-pkg> test UnitCommitment
-```
-
-If all tests pass, the package should now be ready to be used by any Julia script on the machine.
-
-To solve the optimization models, a mixed-integer linear programming (MILP) solver is also required. Please see the [JuMP installation guide](https://jump.dev/JuMP.jl/stable/installation/) for more instructions on installing a solver. Typical open-source choices are [Cbc](https://github.com/JuliaOpt/Cbc.jl) and [GLPK](https://github.com/JuliaOpt/GLPK.jl). In the instructions below, Cbc will be used, but any other MILP solver listed in JuMP installation guide should also be compatible.
-
-Typical Usage
--------------
-
-### Solving user-provided instances
-
-The first step to use UC.jl is to construct a JSON file describing your unit commitment instance. See [Data Format](format.md) for a complete description of the data format UC.jl expects. The next steps, as shown below, are to: (1) read the instance from file; (2) construct the optimization model; (3) run the optimization; and (4) extract the optimal solution.
-
-```julia
-using Cbc
-using JSON
-using UnitCommitment
-
-# 1. Read instance
-instance = UnitCommitment.read("/path/to/input.json")
-
-# 2. Construct optimization model
-model = UnitCommitment.build_model(
-    instance=instance,
-    optimizer=Cbc.Optimizer,
-)
-
-# 3. Solve model
-UnitCommitment.optimize!(model)
-
-# 4. Write solution to a file
-solution = UnitCommitment.solution(model)
-UnitCommitment.write("/path/to/output.json", solution)
-```
-
-### Solving benchmark instances
-
-UnitCommitment.jl contains a large number of benchmark instances collected from the literature and converted into a common data format. To solve one of these instances individually, instead of constructing your own, the function `read_benchmark` can be used, as shown below. See [Instances](instances.md) for the complete list of available instances.
-
-```julia
-using UnitCommitment
-instance = UnitCommitment.read_benchmark("matpower/case3375wp/2017-02-01")
-```
-
-Advanced usage
---------------
-
-### Customizing the formulation
-
-By default, `build_model` uses a formulation that combines modeling components from different publications, and that has been carefully tested, using our own benchmark scripts, to provide good performance across a wide variety of instances. This default formulation is expected to change over time, as new methods are proposed in the literature. You can, however, construct your own formulation, based on the modeling components that you choose, as shown in the next example.
-
-```julia
-using Cbc
-using UnitCommitment
-
-import UnitCommitment:
-    Formulation,
-    KnuOstWat2018,
-    MorLatRam2013,
-    ShiftFactorsFormulation
-
-instance = UnitCommitment.read_benchmark(
-    "matpower/case118/2017-02-01",
-)
-
-model = UnitCommitment.build_model(
-    instance = instance,
-    optimizer = Cbc.Optimizer,
-    formulation = Formulation(
-        pwl_costs = KnuOstWat2018.PwlCosts(),
-        ramping = MorLatRam2013.Ramping(),
-        startup_costs = MorLatRam2013.StartupCosts(),
-        transmission = ShiftFactorsFormulation(
-            isf_cutoff = 0.005,
-            lodf_cutoff = 0.001,
-        ),
-    ),
-)
-```
-
-### Generating initial conditions
-
-When creating random unit commitment instances for benchmark purposes, it is often hard to compute, in advance, sensible initial conditions for all generators. Setting initial conditions naively (for example, making all generators initially off and producing no power) can easily cause the instance to become infeasible due to excessive ramping. Initial conditions can also make it hard to modify existing instances. For example, increasing the system load without carefully modifying the initial conditions may make the problem infeasible or unrealistically challenging to solve.
-
-To help with this issue, UC.jl provides a utility function which can generate feasible initial conditions by solving a single-period optimization problem, as shown below:
-
-```julia
-using Cbc
-using UnitCommitment
-
-# Read original instance
-instance = UnitCommitment.read("instance.json")
-
-# Generate initial conditions (in-place)
-UnitCommitment.generate_initial_conditions!(instance, Cbc.Optimizer)
-
-# Construct and solve optimization model
-model = UnitCommitment.build_model(
-    instance=instance,
-    optimizer=Cbc.Optimizer,
-)
-UnitCommitment.optimize!(model)
-```
-
-```{warning}
-The function `generate_initial_conditions!` may return different initial conditions after each call, even if the same instance and the same optimizer is provided. The particular algorithm may also change in a future version of UC.jl. For these reasons, it is recommended that you generate initial conditions exactly once for each instance and store them for later use.
-```
-    
-### Verifying solutions
-
-When developing new formulations, it is very easy to introduce subtle errors in the model that result in incorrect solutions. To help with this, UC.jl includes a utility function that verifies if a given solution is feasible, and, if not, prints all the validation errors it found. The implementation of this function is completely independent from the implementation of the optimization model, and therefore can be used to validate it. The function can also be used to verify solutions produced by other optimization packages, as long as they follow the [UC.jl data format](format.md).
-
-```julia
-using JSON
-using UnitCommitment
-
-# Read instance
-instance = UnitCommitment.read("instance.json")
-
-# Read solution (potentially produced by other packages) 
-solution = JSON.parsefile("solution.json")
-
-# Validate solution and print validation errors
-UnitCommitment.validate(instance, solution)
-```

instances/README.md

@@ -1,53 +0,0 @@
-Instances
-=========
-
-UnitCommitment.jl provides a large collection of benchmark instances collected
-from the literature and converted to a common data format. If you use these instances in your research, we request that you cite UnitCommitment.jl, as well as the original sources, as listed below. [See documentation for more details](https://anl-ceeesa.github.io/UnitCommitment.jl/).
-
-References
-----------
-
-### UnitCommitment.jl
-
-* [UCJL] **Alinson S. Xavier, Feng Qiu.** "UnitCommitment.jl: A Julia/JuMP Optimization Package for Security-Constrained Unit Commitment". Zenodo (2020). [DOI: 10.5281/zenodo.4269874](https://doi.org/10.5281/zenodo.4269874)
-
-
-### MATPOWER
-
-* [MTPWR] **D. Zimmerman, C. E. Murillo-Sandnchez and R. J. Thomas.** "Matpower:  Steady-state  operations,  planning,  and  analysis  tools  forpower systems research and education", IEEE Transactions on PowerSystems, vol. 26, no. 1, pp. 12 –19, Feb. 2011. [DOI: 10.1109/TPWRS.2010.2051168](https://doi.org/10.1109/TPWRS.2010.2051168)
-
-* [PSTCA] **University of Washington, Dept. of Electrical Engineering.** "Power Systems Test Case Archive". Available at: <http://www.ee.washington.edu/research/pstca/> (Accessed: Nov 14, 2020)
-
-* [JoFlMa16] **C. Josz, S. Fliscounakis, J. Maeght, and P. Panciatici.** "AC Power Flow
-Data in MATPOWER and QCQP Format: iTesla, RTE Snapshots, and PEGASE". [ArXiv (2016)](https://arxiv.org/abs/1603.01533).
-
-* [FlPaCa13] **S. Fliscounakis, P. Panciatici, F. Capitanescu, and L. Wehenkel.**
-"Contingency ranking with respect to overloads in very large power
-systems taking into account uncertainty, preventive and corrective
-actions", Power Systems, IEEE Trans. on, (28)4:4909-4917, 2013.
-[DOI: 10.1109/TPWRS.2013.2251015](https://doi.org/10.1109/TPWRS.2013.2251015)
-
-
-### PGLIB-UC
-
-* [PGLIB] **Carleton Coffrin and Bernard Knueven.** "Power Grid Lib - Unit Commitment". Available at: <https://github.com/power-grid-lib/pglib-uc> (Accessed: Nov 14, 2020)
-
-* [KrHiOn12] **Eric Krall, Michael Higgins and Richard P. O’Neill.** "RTO unit commitment test system." Federal Energy Regulatory Commission. Available at: <https://www.ferc.gov/industries-data/electric/power-sales-and-markets/increasing-efficiency-through-improved-software-1> (Accessed: Nov 14, 2020)
-
-* [KnOsWa20] **Bernard Knueven, James Ostrowski and Jean-Paul Watson.** "On Mixed-Integer Programming Formulations for the Unit Commitment Problem". INFORMS Journal on Computing (2020). [DOI: 10.1287/ijoc.2019.0944](https://doi.org/10.1287/ijoc.2019.0944)
-
-### RTS-GMLC
-
-* https://github.com/GridMod/RTS-GMLC
-
-* [BaBlEh19] **Clayton Barrows, Aaron Bloom, Ali Ehlen, Jussi Ikaheimo, Jennie Jorgenson, Dheepak Krishnamurthy, Jessica Lau et al.** "The IEEE Reliability Test System: A Proposed 2019 Update." IEEE Transactions on Power Systems (2019). [DOI: 10.1109/TPWRS.2019.2925557](https://doi.org/10.1109/TPWRS.2019.2925557)
-
-### OR-LIB
-
-* [ORLIB] **J.E.Beasley.** "OR-Library: distributing test problems by electronic mail", Journal of the Operational Research Society 41(11) (1990). [DOI: 10.2307/2582903](https://doi.org/10.2307/2582903)
-
-* [FrGe06] **A. Frangioni, C. Gentile.** "Solving nonlinear single-unit commitment problems with ramping constraints" Operations Research 54(4), p. 767 - 775, 2006. [DOI: 10.1287/opre.1060.0309](https://doi.org/10.1287/opre.1060.0309)
-
-### Tejada19
-
-* [TeLuSa19] **D. A. Tejada-Arango, S. Lumbreras, P. Sanchez-Martin and A. Ramos.** "Which Unit-Commitment Formulation is Best? A Systematic Comparison," in IEEE Transactions on Power Systems. [DOI: 10.1109/TPWRS.2019.2962024](https://ieeexplore.ieee.org/document/8941313/).