New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Feature/nr_pf_solver #54

Merged

GabrielKS merged 24 commits into hrgks/psse_exporter_psy4 from feature/nr_pf_solver

Jan 16, 2025

rbolgaryn commented Dec 6, 2024

Integrate the implementation of the Newton-Raphson Power Flow calculation leveraging KLU factorization

Roman Bolgaryn added 4 commits

December 5, 2024 08:55


          WIP: implement NR power flow function, rename ACPowerFlow to NLSolveA…

198a4fe

…CPowerFlow, introduce KLUACPowerFlow for the use in the new NR power flow function


          WIP: some bugfixes in the PF function

d4a443a


          small changes

f2ad409


          KLU PF: use correct V0; return x in expected format; write results fu…

251f1b3

…nction supports KLU PF; implement tests for KLUACPowerFlow

rbolgaryn added the enhancement label

rbolgaryn requested a review from GabrielKS

December 6, 2024 21:54


          reformat code

210937c

github-actions bot reviewed

View reviewed changes

Contributor

github-actions bot left a comment

Remaining comments which cannot be posted as a review comment to avoid GitHub Rate Limit

JuliaFormatter

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 191 to 192 in 210937c

    
           if isapprox(Psources, P_gen; atol=0.001) && 
        
              isapprox(Qsources, Q_gen; atol=0.001)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 207 in 210937c

sort(collect(devices_); by=x -> _get_limits_for_power_distribution(x).max)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 229 in 210937c

if !isapprox(p_residual, 0.0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 236 in 210937c

while !isapprox(p_residual, 0.0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 258 in 210937c

if isapprox(p_residual, 0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 266 in 210937c

if !isapprox(p_residual, 0.0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 301 in 210937c

if isapprox(Qsources, Q_gen; atol=0.001)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 314 in 210937c

devices = sort(collect(devices_); by=x -> PSY.get_max_reactive_power(x))

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 321 in 210937c

if isapprox(total_active_power, 0.0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 336 to 337 in 210937c

    
           if isapprox(q_limits.max, 0.0; atol=BOUNDS_TOLERANCE) && 
        
              isapprox(q_limits.min, 0.0; atol=BOUNDS_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 364 in 210937c

if isapprox(q_residual, 0.0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 369 in 210937c

if !isapprox(q_residual, 0.0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 371 in 210937c

while !isapprox(q_residual, 0.0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 410 in 210937c

if isapprox(q_residual, 0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 422 in 210937c

if !isapprox(q_residual, 0.0; atol=ISAPPROX_ZERO_TOLERANCE)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 439 in 210937c

atol=ISAPPROX_ZERO_TOLERANCE,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 454 in 210937c

sort!(collect(PSY.get_components(PSY.Bus, sys)); by=x -> PSY.get_number(x)),

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 459 to 460 in 210937c

    
           P_gen = result[2*ix-1] 
        
           Q_gen = result[2*ix]

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 463 to 464 in 210937c

    
           Q_gen = result[2*ix-1] 
        
           bus.angle = result[2*ix]

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 467 to 468 in 210937c

    
           Vm = result[2*ix-1] 
        
           θ = result[2*ix]

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 484 in 210937c

function _get_branches_buses(data::Union{PTDFPowerFlowData,vPTDFPowerFlowData})

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 516 to 524 in 210937c

    
           bus_number=buses, 
        
           Vm=bus_magnitude, 
        
           θ=bus_angles, 
        
           P_gen=P_gen_vect, 
        
           P_load=P_load_vect, 
        
           P_net=P_gen_vect - P_load_vect, 
        
           Q_gen=Q_gen_vect, 
        
           Q_load=Q_load_vect, 
        
           Q_net=Q_gen_vect - Q_load_vect,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 529 to 537 in 210937c

    
           line_name=branches, 
        
           bus_from=from_bus, 
        
           bus_to=to_bus, 
        
           P_from_to=P_from_to_vect, 
        
           Q_from_to=Q_from_to_vect, 
        
           P_to_from=P_to_from_vect, 
        
           Q_to_from=Q_to_from_vect, 
        
           P_losses=zeros(length(branches)), 
        
           Q_losses=zeros(length(branches)),

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 556 in 210937c

data::Union{PTDFPowerFlowData,vPTDFPowerFlowData,ABAPowerFlowData},

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 575 in 210937c

result_dict = Dict{Union{String,Char},Dict{String,DataFrames.DataFrame}}()

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 617 in 210937c

    
           buses = sort!(collect(PSY.get_components(PSY.Bus, sys)); by=x -> PSY.get_number(x))

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 637 to 638 in 210937c

    
           P_gen_vect[ix] = result[2*ix-1] * sys_basepower 
        
           Q_gen_vect[ix] = result[2*ix] * sys_basepower

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 641 in 210937c

θ_vect[ix] = result[2*ix]

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Line 643 in 210937c

Q_gen_vect[ix] = result[2*ix-1] * sys_basepower

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 645 to 646 in 210937c

    
           Vm_vect[ix] = result[2*ix-1] 
        
           θ_vect[ix] = result[2*ix]

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 669 to 677 in 210937c

    
           bus_number=PSY.get_number.(buses), 
        
           Vm=Vm_vect, 
        
           θ=θ_vect, 
        
           P_gen=P_gen_vect, 
        
           P_load=P_load_vect, 
        
           P_net=P_gen_vect - P_load_vect, 
        
           Q_gen=Q_gen_vect, 
        
           Q_load=Q_load_vect, 
        
           Q_net=Q_gen_vect - Q_load_vect,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/src/post_processing.jl

Lines 681 to 689 in 210937c

    
           line_name=PSY.get_name.(branches), 
        
           bus_from=PSY.get_number.(PSY.get_from.(PSY.get_arc.(branches))), 
        
           bus_to=PSY.get_number.(PSY.get_to.(PSY.get_arc.(branches))), 
        
           P_from_to=P_from_to_vect, 
        
           Q_from_to=Q_from_to_vect, 
        
           P_to_from=P_to_from_vect, 
        
           Q_to_from=Q_to_from_vect, 
        
           P_losses=P_from_to_vect + P_to_from_vect, 
        
           Q_losses=Q_from_to_vect + Q_to_from_vect,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 33 to 34 in 210937c

    
           sys = PSB.build_system(PSB.PSITestSystems, "c_sys14"; add_forecasts=false) 
        
           data = PowerFlows.PowerFlowData(ACPowerFlow(), sys; check_connectivity=true)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 38 in 210937c

@test solve_ac_powerflow!(ACPowerFlow(), sys; method=:newton)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 42 in 210937c

data = PowerFlows.PowerFlowData(ACPowerFlow(), sys; check_connectivity=true)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 48 to 49 in 210937c

    
           @testset "AC Power Flow 14-Bus Line Configurations" for ACPowerFlow in (NLSolveACPowerFlow, KLUACPowerFlow) 
        
               sys = PSB.build_system(PSB.PSITestSystems, "c_sys14"; add_forecasts=false)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 58 in 210937c

sys = PSB.build_system(PSB.PSITestSystems, "c_sys14"; add_forecasts=false)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 64 in 210937c

@test isapprox(PSY.get_magnitude(test_bus), 1.002; atol=1e-3, rtol=0)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 66 in 210937c

sys = PSB.build_system(PSB.PSITestSystems, "c_sys14"; add_forecasts=false)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 82 to 83 in 210937c

    
           @test isapprox(df["bus_results"].P_gen, p_gen_matpower_3bus, atol=1e-4) 
        
           @test isapprox(df["bus_results"].Q_gen, q_gen_matpower_3bus, atol=1e-4)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 86 to 87 in 210937c

    
           @testset "AC Power Flow convergence fail testing" for ACPowerFlow in (NLSolveACPowerFlow, KLUACPowerFlow) 
        
               pf_sys5_re = PSB.build_system(PSB.PSITestSystems, "c_sys5_re"; add_forecasts=false)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 110 to 111 in 210937c

    
           bus_name_formatter=x -> strip(string(x["name"])) * "-" * string(x["index"]), 
        
           runchecks=false,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 138 to 144 in 210937c

    
           number=1, 
        
           name="01", 
        
           bustype=ACBusTypes.REF, 
        
           angle=0.0, 
        
           magnitude=1.1, 
        
           voltage_limits=(0.0, 2.0), 
        
           base_voltage=230,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 150 to 156 in 210937c

    
           name="source_1", 
        
           available=true, 
        
           bus=b, 
        
           active_power=0.5, 
        
           reactive_power=0.1, 
        
           R_th=1e-5, 
        
           X_th=1e-5,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 160 to 166 in 210937c

    
           name="source_2", 
        
           available=true, 
        
           bus=b, 
        
           active_power=-0.5, 
        
           reactive_power=-0.1, 
        
           R_th=1e-5, 
        
           X_th=1e-5,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 179 to 185 in 210937c

    
           number=1, 
        
           name="01", 
        
           bustype=ACBusTypes.REF, 
        
           angle=0.0, 
        
           magnitude=1.1, 
        
           voltage_limits=(0.0, 2.0), 
        
           base_voltage=230,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 189 to 195 in 210937c

    
           number=2, 
        
           name="02", 
        
           bustype=ACBusTypes.PV, 
        
           angle=0.0, 
        
           magnitude=1.1, 
        
           voltage_limits=(0.0, 2.0), 
        
           base_voltage=230,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 198 in 210937c

a = Arc(; from=b1, to=b2)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 201 to 210 in 210937c

    
           name="l1", 
        
           available=true, 
        
           active_power_flow=0.0, 
        
           reactive_power_flow=0.0, 
        
           arc=a, 
        
           r=1e-3, 
        
           x=1e-3, 
        
           b=(from=0.0, to=0.0), 
        
           rating=0.0, 
        
           angle_limits=(min=-pi / 2, max=pi / 2),

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 216 to 222 in 210937c

    
           name="source_1", 
        
           available=true, 
        
           bus=b1, 
        
           active_power=0.5, 
        
           reactive_power=0.1, 
        
           R_th=1e-5, 
        
           X_th=1e-5,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 226 to 232 in 210937c

    
           name="source_2", 
        
           available=true, 
        
           bus=b2, 
        
           active_power=0.5, 
        
           reactive_power=1.1, 
        
           R_th=1e-5, 
        
           X_th=1e-5,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 236 to 242 in 210937c

    
           name="source_3", 
        
           available=true, 
        
           bus=b2, 
        
           active_power=-0.5, 
        
           reactive_power=-1.1, 
        
           R_th=1e-5, 
        
           X_th=1e-5,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 256 to 262 in 210937c

    
           number=1, 
        
           name="01", 
        
           bustype=ACBusTypes.REF, 
        
           angle=0.0, 
        
           magnitude=1.1, 
        
           voltage_limits=(0.0, 2.0), 
        
           base_voltage=230,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 268 to 274 in 210937c

    
           name="source_1", 
        
           available=true, 
        
           bus=b, 
        
           active_power=0.5, 
        
           reactive_power=0.1, 
        
           R_th=1e-5, 
        
           X_th=1e-5,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 278 to 295 in 210937c

    
           name="init", 
        
           available=true, 
        
           status=false, 
        
           bus=b, 
        
           active_power=0.1, 
        
           reactive_power=0.1, 
        
           rating=0.0, 
        
           active_power_limits=(min=0.0, max=0.0), 
        
           reactive_power_limits=nothing, 
        
           ramp_limits=nothing, 
        
           operation_cost=ThermalGenerationCost(nothing), 
        
           base_power=100.0, 
        
           time_limits=nothing, 
        
           prime_mover_type=PrimeMovers.OT, 
        
           fuel=ThermalFuels.OTHER, 
        
           services=Device[], 
        
           dynamic_injector=nothing, 
        
           ext=Dict{String,Any}(),

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 303 in 210937c

atol=0.001,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 308 in 210937c

atol=0.001,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 315 to 321 in 210937c

    
           number=1, 
        
           name="01", 
        
           bustype=ACBusTypes.REF, 
        
           angle=0.0, 
        
           magnitude=1.1, 
        
           voltage_limits=(0.0, 2.0), 
        
           base_voltage=230,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 325 to 331 in 210937c

    
           number=2, 
        
           name="02", 
        
           bustype=ACBusTypes.PV, 
        
           angle=0.0, 
        
           magnitude=1.1, 
        
           voltage_limits=(0.0, 2.0), 
        
           base_voltage=230,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 334 in 210937c

a = Arc(; from=b1, to=b2)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 337 to 346 in 210937c

    
           name="l1", 
        
           available=true, 
        
           active_power_flow=0.0, 
        
           reactive_power_flow=0.0, 
        
           arc=a, 
        
           r=1e-3, 
        
           x=1e-3, 
        
           b=(from=0.0, to=0.0), 
        
           rating=0.0, 
        
           angle_limits=(min=-pi / 2, max=pi / 2),

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 352 to 358 in 210937c

    
           name="source_1", 
        
           available=true, 
        
           bus=b1, 
        
           active_power=0.5, 
        
           reactive_power=0.1, 
        
           R_th=1e-5, 
        
           X_th=1e-5,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 362 to 368 in 210937c

    
           name="source_2", 
        
           available=true, 
        
           bus=b2, 
        
           active_power=0.5, 
        
           reactive_power=1.1, 
        
           R_th=1e-5, 
        
           X_th=1e-5,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Lines 372 to 389 in 210937c

    
           name="init", 
        
           available=true, 
        
           status=false, 
        
           bus=b2, 
        
           active_power=0.1, 
        
           reactive_power=0.1, 
        
           rating=0.0, 
        
           active_power_limits=(min=0.0, max=0.0), 
        
           reactive_power_limits=nothing, 
        
           ramp_limits=nothing, 
        
           operation_cost=ThermalGenerationCost(nothing), 
        
           base_power=100.0, 
        
           time_limits=nothing, 
        
           prime_mover_type=PrimeMovers.OT, 
        
           fuel=ThermalFuels.OTHER, 
        
           services=Device[], 
        
           dynamic_injector=nothing, 
        
           ext=Dict{String,Any}(),

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 397 in 210937c

atol=0.001,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_nlsolve_powerflow.jl

Line 402 in 210937c

atol=0.001,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_powerflow_data.jl

Line 2 in 210937c

sys = PSB.build_system(PSB.PSITestSystems, "c_sys14"; add_forecasts=false)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_powerflow_data.jl

Line 10 in 210937c

sys = PSB.build_system(PSB.PSITestSystems, "c_sys14"; add_forecasts=false)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_powerflow_data.jl

Lines 13 to 14 in 210937c

    
           @test PowerFlowData(DCPowerFlow(), sys; time_steps=time_steps) isa PF.ABAPowerFlowData 
        
           @test PowerFlowData(PTDFDCPowerFlow(), sys; time_steps=time_steps) isa

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_powerflow_data.jl

Line 16 in 210937c

@test PowerFlowData(vPTDFDCPowerFlow(), sys; time_steps=time_steps) isa

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 2 in 210937c

isdir(test_psse_export_dir) && rm(test_psse_export_dir; recursive=true)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 13 in 210937c

macro log_assert(ex, comparison_name=nothing)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 27 in 210937c

default_tol=SYSTEM_REIMPORT_COMPARISON_TOLERANCE;

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 41 in 210937c

all(isapprox.(col1, col2; atol=my_tol))

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 55 in 210937c

default_tol=SYSTEM_REIMPORT_COMPARISON_TOLERANCE;

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 67 in 210937c

    
           isapprox(a, b; atol=SYSTEM_REIMPORT_COMPARISON_TOLERANCE) || IS.isequivalent(a, b)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 71 in 210937c

bus_name_mapping=Dict{String,String}(),

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 73 in 210937c

include_types=[

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 86 in 210937c

exclude_fields=Set([

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 93 in 210937c

exclude_fields_for_type=Dict(

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 114 in 210937c

generator_comparison_fns=[ # TODO rating

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Lines 121 to 123 in 210937c

    
           ignore_name_order=true, 
        
           ignore_extra_of_type=Union{PSY.ThermalStandard,PSY.StaticLoad}, 
        
           exclude_reactive_power=false)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 132 in 210937c

result &= IS.compare_values(sys1, sys2; exclude=[:data])

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 181 in 210937c

exclude=my_excludes,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 192 in 210937c

GenSource = Union{Generator,Source}

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 208 in 210937c

any(Union{typeof(gen1),typeof(gen2)} .<: include_types) && continue

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 214 in 210937c

exclude=exclude_fields,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 225 in 210937c

    
           function test_power_flow(sys1::System, sys2::System; exclude_reactive_flow=false)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Lines 235 to 236 in 210937c

    
           POWERFLOW_COMPARISON_TOLERANCE; line_name=nothing, Q_to_from=reactive_power_tol, 
        
           Q_from_to=reactive_power_tol, Q_losses=reactive_power_tol)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Lines 253 to 254 in 210937c

    
           do_power_flow_test=true, 
        
           exclude_reactive_flow=false,

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 268 in 210937c

test_power_flow(sys, sys2; exclude_reactive_flow=exclude_reactive_flow)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 277 in 210937c

exclude_metadata_keys=["case_name"],

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 332 in 210937c

exclude_reactive_flow=true) # TODO why is reactive flow not matching?

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 363 in 210937c

    
           test_power_flow(sys2, reread_sys2; exclude_reactive_flow=true)  # TODO why is reactive flow not matching?

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 377 in 210937c

exclude_reactive_flow=true) # TODO why is reactive flow not matching?

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 407 in 210937c

    
           test_power_flow(sys2, reread_sys2; exclude_reactive_flow=true)  # TODO why is reactive flow not matching?

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 420 in 210937c

exclude_reactive_power=true) # TODO why is reactive power not matching?

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 424 in 210937c

    
           test_power_flow(sys2, reread_sys3; exclude_reactive_flow=true)  # TODO why is reactive flow not matching?

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 427 in 210937c

exporter = PSSEExporter(sys, :v33, export_location; write_comments=true)

[JuliaFormatter] _{reported by reviewdog 🐶}

PowerFlows.jl/test/test_psse_export.jl

Line 429 in 210937c

exclude_reactive_flow=true) # TODO why is reactive flow not matching?


          formatting

c699556

GabrielKS requested changes

View reviewed changes

Contributor

GabrielKS left a comment

A few initial comments. I'll dig into the math more later. Thanks!

Project.toml Outdated Show resolved Hide resolved

src/PowerFlows.jl Show resolved Hide resolved

src/nlsolve_ac_powerflow.jl Show resolved Hide resolved

src/nlsolve_ac_powerflow.jl Outdated Show resolved Hide resolved

src/nlsolve_ac_powerflow.jl Outdated

-                      write_powerflow_solution!(system, res.zero, max_iterations)
+                  converged, x = _solve_powerflow!(pf, data, check_reactive_power_limits; kwargs...)
+                  if converged
+                      write_powerflow_solution!(system, x, max_iterations)
                       @info("PowerFlow solve converged, the results have been stored in the system")
                       #Restore original per unit base
                       PSY.set_units_base_system!(system, settings_unit_cache)

Contributor

GabrielKS Dec 10, 2024 •

edited

Loading

Not necessarily your responsibility but just flagging that this is a good use case for

PowerFlows.jl/src/psse_export.jl

Line 47 in 465e92f

    
           function with_units(f::Function, sys::System, units::Union{PSY.UnitSystem, String})

(and a good argument for moving it someplace more central)

Contributor

GabrielKS Jan 16, 2025

This can be done now — with_units is now with_units_base in PSY

src/nlsolve_ac_powerflow.jl Outdated Show resolved Hide resolved

src/nlsolve_ac_powerflow.jl Outdated Show resolved Hide resolved

src/nlsolve_ac_powerflow.jl Outdated

+                  # Find indices for each bus type
+                  ref = findall(x -> x == PowerSystems.ACBusTypesModule.ACBusTypes.REF, data.bus_type)
+                  pv = findall(x -> x == PowerSystems.ACBusTypesModule.ACBusTypes.PV, data.bus_type)
+                  pq = findall(x -> x == PowerSystems.ACBusTypesModule.ACBusTypes.PQ, data.bus_type)

Contributor

GabrielKS Dec 10, 2024

Probably no significant difference, but note that these three lines could be combined to only iterate over the bus types once

src/nlsolve_ac_powerflow.jl Outdated Show resolved Hide resolved

src/nlsolve_ac_powerflow.jl Outdated Show resolved Hide resolved

GabrielKS reviewed

View reviewed changes

src/nlsolve_ac_powerflow.jl Outdated

+                      J = [j11 j12; j21 j22]
+                      factor_J = KLU.klu(J)
+                      dx = -(factor_J \ F)

Contributor

GabrielKS Dec 10, 2024 •

edited

Loading

Comment from José: do this and previous two lines in place using functor approach

GabrielKS reviewed

View reviewed changes

src/powerflow_types.jl Outdated Show resolved Hide resolved

Roman Bolgaryn added 2 commits

December 16, 2024 14:45


          Addressed review comments except for using functor

b0bc565


          formatter

42f02ab

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated Show resolved Hide resolved

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated Show resolved Hide resolved

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated Show resolved Hide resolved

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated Show resolved Hide resolved

Roman Bolgaryn added 4 commits

December 17, 2024 14:14


          small fix ACPowerFlow type; add test for results consistency for 2000…

e6a0364

… bus system


          small change

da200a1


          optimize NR code to reduce memory allocations

b385c65


          trying to reduce memory allocation

d7f7c88

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated

+                  F_imag = view(F, npvpq + 1:npvpq + npq)
+                  mis .= V .* conj.(Ybus * V) .- Sbus
+                  @. F_real = real(mis_pvpq)  # In-place assignment to the real part, using views

Member

jd-lara Dec 19, 2024

@rbolgaryn I think you are overwriting F_real here. You might want to try F_real[:] =

Roman Bolgaryn added 3 commits

January 8, 2025 15:29


          NR PF non-allocating

9d3b099


          vectorized update for dSbus_dV

d688b58


          fixes Add solve_powerflow! for AC power flow #49

11d86a1

rbolgaryn linked an issue

that may be closed by this pull request

Add solve_powerflow! for AC power flow #49

Open

Roman Bolgaryn added 2 commits

January 9, 2025 14:31


          formatting


          WIP: multiperiod AC PF

b700bc4

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated

+                  return converged, x
+              end
+              function _newton_powerflow(

Member

jd-lara Jan 10, 2025

Shouldn't this be nlsolve_powerflow?

Author

rbolgaryn Jan 10, 2025

We are relying on multiple dispatch here to go to KLU or NLSolve from solve_powerflow(...)

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated Show resolved Hide resolved

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated Show resolved Hide resolved

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated Show resolved Hide resolved

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Show resolved Hide resolved

jd-lara reviewed

View reviewed changes

src/newton_ac_powerflow.jl Outdated Show resolved Hide resolved

Roman Bolgaryn added 4 commits

January 10, 2025 11:40


          adrressing comments

0f1545e


          wip: multiperiod ac pf

d067275


          wip: solve_powerflow!

d198c2e


          Merge branch 'hrgks/psse_exporter_psy4' into feature/nr_pf_solver

a121ff1

GabrielKS requested changes

View reviewed changes

Contributor

GabrielKS left a comment •

edited

Loading

Various notes from a software engineering/code style perspective

src/PowerFlowData.jl

@@ @@ -173,6 +173,8 @@ function PowerFlowData( @@
                       end
                   end
+                  timestep_map = Dict(zip([i for i in 1:time_steps], timestep_names))

Contributor

GabrielKS Jan 15, 2025

Suggested change

      
                timestep_map = Dict(zip([i for i in 1:time_steps], timestep_names))
          
                Dict(enumerate(timestep_names))

src/common.jl

@@ @@ -162,7 +162,7 @@ function make_powerflowdata( @@
                   # TODO: bus_type might need to also be a Matrix since the type can change for a particular scenario
                   bus_type = Vector{PSY.ACBusTypes}(undef, n_buses)
                   bus_angles = zeros(Float64, n_buses)
-                  bus_magnitude = zeros(Float64, n_buses)
+                  bus_magnitude = ones(Float64, n_buses)

Contributor

GabrielKS Jan 15, 2025

Why this change?

src/common.jl Outdated

-                  init_1 = zeros(n_buses, time_steps)
-                  init_2 = zeros(n_branches, time_steps)
+                  init_1 = zeros(Float64, n_buses, time_steps)
+                  init_1m = ones(Float64, n_buses, time_steps)

Contributor

GabrielKS Jan 15, 2025

Why this change?

src/newton_ac_powerflow.jl

+                  kwargs...,
+              )
+                  #Save per-unit flag
+                  settings_unit_cache = deepcopy(system.units_settings.unit_system)

Contributor

GabrielKS Jan 15, 2025

Use with_units_base

Contributor

GabrielKS Jan 15, 2025

(also do this in solve_powerflow, etc.)

src/newton_ac_powerflow.jl

		end


		function _ac_powereflow(

Contributor

GabrielKS Jan 15, 2025

Suggested change

      
            function _ac_powereflow(
          
            function _ac_powerflow(

test/runtests.jl

@@ @@ -33,6 +35,9 @@ MAIN_DIR = dirname(@__DIR__) @@
               include("test_utils/common.jl")
               include("test_utils/psse_results_compare.jl")
+              #include("test_utils/legacy_pf.jl")
+              #Base.eval(PF, Meta.parse("include(\"test/test_utils/legacy_pf.jl\")"))
+              Base.eval(PowerFlows, :(include("./test_utils/legacy_pf.jl")))

Contributor

GabrielKS Jan 15, 2025

The @eval macro would probably make this syntax nicer. But also, this is kind of hacky — why do you need to evaluate it as if it were a part of PowerFlows? Shouldn't it suffice to define function PowerFlows._newton_powerflow, etc. in the current module?

test/test_multiperiod_ac_powerflow.jl

+                  data.bus_activepower_withdrawals .= deepcopy(withs)
+                  # get power flows with NR KLU method and write results
+                  results = solve_powerflow(pf, data, sys)

Contributor

GabrielKS Jan 15, 2025

TODO actually run some tests…

test/test_newton_ac_powerflow.jl Outdated

+                  converged1, V1, S1 = PowerFlows._solve_powerflow!(pf, data, false)
+                  x1 = PowerFlows._calc_x(data, V1, S1)
+                  @test LinearAlgebra.norm(result_14 - x1, Inf) <= 1e-6
+                  @test solve_powerflow!(pf, sys; method = :newton)
                   # Test enforcing the reactive power Limits

Contributor

GabrielKS Jan 15, 2025

TODO verify that the updates from #76 make it into here

test/test_utils/psse_results_compare.jl

		@@ -1,4 +1,4 @@
		function psse_bus_results_compare(file_name, pf_results)
		function psse_bus_results_compare(file_name::String, pf_results::Dict)

Contributor

GabrielKS Jan 15, 2025

nitpick:

Suggested change

      
            function psse_bus_results_compare(file_name::String, pf_results::Dict)
          
            function psse_bus_results_compare(file_name::AbstractString, pf_results::AbstractDict)

for flexibility

Contributor

GabrielKS Jan 15, 2025

(same below)

src/newton_ac_powerflow.jl

+                      @info("PowerFlow solve converged, the results are exported in DataFrames")
+                      df_results = write_results(pf, system, data, x)
+                  else
+                      df_results = missing

Contributor

GabrielKS Jan 15, 2025

I'm not completely sure, but I think this might be better as

Suggested change

      
                    df_results = missing
          
                    df_results = nothing

. See https://docs.julialang.org/en/v1/manual/faq/#Nothingness-and-missing-values-1 and https://discourse.julialang.org/t/diff-missing-vs-nothing/35806.

Roman Bolgaryn added 3 commits

January 15, 2025 16:46


          integrate previous changes to powersystemdata

994123b


          remove a println

09189c3


          small change to _update_system!

691a73a

Contributor

GabrielKS commented Jan 16, 2025

Merging now so we can have a consolidated dev branch to work off of — @rbolgaryn will continue to address the above feedback in a new pull request.

GabrielKS merged commit 3c0ccd7 into hrgks/psse_exporter_psy4

3 of 6 checks passed

GabrielKS mentioned this pull request

PowerFlows.jl unified develop branch #53

Draft

jd-lara deleted the feature/nr_pf_solver branch

January 16, 2025 18:33

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels