def test_industrialSite():
# # Model an energy system
# Input parameters
locations = {'industry_0'}
commodityUnitDict = {
'electricity': r'MW$_{el}$',
'hydrogen': r'MW$_{H_{2},LHV}$',
'heat': r'MW$_{heat}$}'
}
commodities = {'electricity', 'hydrogen', 'heat'}
numberOfTimeSteps, hoursPerTimeStep = 24 * 6, 1 / 6 #8760, 1 # 52560, 1/6
costUnit, lengthUnit = '1e3 Euro', 'km'
# Code
esM = fn.EnergySystemModel(locations=locations,
commodities=commodities,
numberOfTimeSteps=numberOfTimeSteps,
commodityUnitsDict=commodityUnitDict,
hoursPerTimeStep=hoursPerTimeStep,
costUnit=costUnit,
lengthUnit=lengthUnit,
verboseLogLevel=0)
# ## Add source component
data = pd.read_excel(
os.path.join(os.path.dirname(__file__), '_testInputFiles',
'generationTimeSeries_e825103.xlsx'))
data = data.iloc[0:numberOfTimeSteps]
operationRateFix = pd.DataFrame(
data['e825103_2017_2.3MW_faults9'],
index=range(numberOfTimeSteps)) # Dataset with least missing data
operationRateFix.columns = ['industry_0']
# Input parameters
name, commodity = 'Wind turbines', 'electricity'
hasCapacityVariable = True
capacityMax = pd.Series([10],
index=['industry_0']) # 10 MW_el = 0.01 GW_el
investPerCapacity, opexPerCapacity = 0, 30 # 30 €/kW = 30 1e6€/GW = 30 1e3€/MW
interestRate, economicLifetime = 0.08, 20
esM.add(
fn.Source(esM=esM,
name=name,
commodity=commodity,
hasCapacityVariable=hasCapacityVariable,
operationRateFix=operationRateFix,
capacityMax=capacityMax,
investPerCapacity=investPerCapacity,
opexPerCapacity=opexPerCapacity,
interestRate=interestRate,
economicLifetime=economicLifetime))
# ## Add conversion components
esM.add(
fn.Conversion(
esM=esM,
name='PEMEC',
physicalUnit=r'MW$_{el}$',
commodityConversionFactors={
'electricity': -1,
'hydrogen': 0.67
},
hasCapacityVariable=True,
investPerCapacity=2300,
opexPerCapacity=12.5,
interestRate=0.08, # for 2018 CAPEX
economicLifetime=5))
func = lambda x: 0.5 * (x - 2)**3 + (x - 2)**2 + 0.0001
esM.add(
fn.ConversionPartLoad(
esM=esM,
name='AEC',
physicalUnit=r'MW$_{el}$',
commodityConversionFactors={
'electricity': -1,
'hydrogen': 0.64
},
commodityConversionFactorsPartLoad={
'electricity': -1,
'hydrogen': func
},
# commodityConversionFactorsPartLoad=pwl,
nSegments=2,
hasCapacityVariable=True,
bigM=99,
investPerCapacity=1300,
opexPerCapacity=18,
interestRate=0.08, # for 2018 CAPEX
economicLifetime=9))
# ## Add storage components
esM.add(
fn.Storage(esM=esM,
name='Hydrogen tank (gaseous)',
commodity='hydrogen',
hasCapacityVariable=True,
capacityVariableDomain='continuous',
capacityPerPlantUnit=1,
chargeRate=1,
dischargeRate=1,
sharedPotentialID=None,
stateOfChargeMin=0.06,
stateOfChargeMax=1,
investPerCapacity=0.004,
opexPerCapacity=0.004 * 0.02,
interestRate=0.08,
economicLifetime=20))
### Industrial hydrogen demand
operationRateFix = pd.DataFrame(
2 * np.ones(numberOfTimeSteps) * (hoursPerTimeStep),
columns=['industry_0']) # constant hydrogen demand of 2 MW_GH2:
esM.add(
fn.Sink(esM=esM,
name='Hydrogen demand',
commodity='hydrogen',
hasCapacityVariable=False,
operationRateFix=operationRateFix))
# Heat output
esM.add(
fn.Sink(esM=esM,
name='Heat output',
commodity='heat',
hasCapacityVariable=False))
# Input parameters
timeSeriesAggregation = False
solver = 'glpk'
# Optimize
esM.optimize(timeSeriesAggregation=timeSeriesAggregation, solver=solver)
def test_conversionPartLoad():
# Set up energy system model instance
locations = {'GlassProductionSite'}
commodities = {'electricity', 'heat', 'hydrogen', 'O2', 'CO2','rawMaterial'}
commodityUnitDict = {'electricity': r'kW$_{el}$', 'heat':r'kW$_{heat}$}', 'hydrogen':r'kW$_{H2}$',
'O2': r'kg$_{O_{2}}$/h', 'CO2': r'kg$_{CO_{2}}$/h', 'rawMaterial': r'kg$_{R}}$/h'}
numberOfTimeSteps=80
hoursPerTimeStep=0.25
esM = fn.EnergySystemModel(locations=locations, commodities=commodities, numberOfTimeSteps=numberOfTimeSteps,
commodityUnitsDict=commodityUnitDict,
hoursPerTimeStep=hoursPerTimeStep, costUnit='1 Euro', lengthUnit='km', verboseLogLevel=0)
### Sources ###
# Electricity from grid
esM.add(fn.Source(esM=esM, name='ElectricityGrid', commodity='electricity', hasCapacityVariable=False,
commodityCost=0.070))
# Oxygen source from trailer
esM.add(fn.Source(esM=esM, name='oxygenSource', commodity='O2', hasCapacityVariable=False,
commodityCost=0.070))
# Raw material for glass furnace
esM.add(fn.Source(esM=esM, name='rawMaterialSource', commodity='rawMaterial', hasCapacityVariable=False,
commodityCost=0.20))
### Conversion ###
# PEM Electrolyzer
# Get partLoadData from EC Campus Mainz
Utilization = [0.0208023774145616,0.0222882615156017,0.0222882615156017,0.0222882615156017,0.025260029717682,
0.025260029717682,0.0267459138187221,0.0282317979197622,0.0297176820208024,0.0312035661218424,
0.0326894502228826,0.0341753343239227,0.0356612184249628,0.0371471025260029,0.038632986627043,
0.0416047548291233,0.0430906389301634,0.0445765230312035,0.0475482912332838,0.0430906389301634,
0.0490341753343239,0.050520059435364,0.0520059435364041,0.0549777117384844,0.0579494799405646,
0.0609212481426448,0.0638930163447251,0.0683506686478455,0.0713224368499256,0.075780089153046,
0.0787518573551263,0.0832095096582466,0.087667161961367,0.0921248142644873,0.0980683506686478,
0.104011887072808,0.108469539375928,0.114413075780089,0.120356612184249,0.12778603268945,
0.13521545319465,0.142644873699851,0.151560178306092,0.157503714710252,0.166419019316493,
0.173848439821693,0.181277860326894,0.190193164933135,0.196136701337295,0.202080237741456,
0.209509658246656,0.216939078751857,0.224368499257057,0.233283803863298,0.240713224368499,
0.246656760772659,0.25408618127786,0.26151560178306,0.270430906389301,0.276374442793462,
0.283803863298662,0.292719167904903,0.298662704309063,0.306092124814264,0.313521545319465,
0.319465081723625,0.328380386329866,0.335809806835066,0.343239227340267,0.349182763744427,
0.356612184249628,0.364041604754829,0.371471025260029,0.38038632986627,0.389301634472511,
0.398216939078751,0.407132243684992,0.416047548291233,0.423476968796433,0.430906389301634,
0.439821693907875,0.448736998514115,0.457652303120356,0.465081723625557,0.472511144130757,
0.481426448736998,0.490341753343239,0.497771173848439,0.50668647845468,0.514115898959881,
0.523031203566121,0.531946508172362,0.540861812778603,0.551263001485884,0.560178306092124,
0.567607726597325,0.576523031203566,0.585438335809806,0.595839524517087,0.604754829123328,
0.613670133729569,0.619613670133729,0.62852897473997,0.638930163447251,0.646359583952451,
0.656760772659732,0.664190193164933,0.673105497771173,0.684992570579494,0.692421991084695,
0.699851411589896,0.710252600297176,0.717682020802377,0.726597325408618,0.735512630014858,
0.744427934621099,0.75334323922734,0.762258543833581,0.769687964338781,0.780089153046062,
0.786032689450223,0.793462109955423,0.802377414561664,0.811292719167904,0.820208023774145,
0.829123328380386,0.838038632986627,0.846953937592867,0.858841010401188,0.867756315007429,
0.87667161961367,0.88707280832095,0.895988112927191,0.904903417533432,0.913818722139673,
0.924219910846954,0.933135215453194,0.942050520059435,0.947994056463595,0.958395245170876,
0.965824665676077,0.974739970282318,0.985141158989598,0.994056463595839]
Efficiency = [0.03449362655834178,0.05655553999159559,0.04920156884717763,0.07616612971004356,0.09332539571368476,
0.115387309146939,0.13254657515058135,0.1497058411542229,0.1693164308726712,0.1864756968763127,
0.2036349628799551,0.2256968763132085,0.2404048186020449,0.25511276089088053,0.27472335060932884,
0.2918826166129703,0.30904188261661275,0.33110379604986695,0.34581173833870255,0.3212985011906424,
0.3629710043423449,0.38013027034598645,0.3923868889200161,0.4095461549236585,0.42425409721249496,
0.4365107157865246,0.44876733436055427,0.45857262921977887,0.4683779240790026,0.4830858663678382,
0.4953424849418686,0.5075991035158983,0.517404398375122,0.5247583695195407,0.5321123406639585,
0.5419176355231823,0.5517229303824059,0.5639795489564365,0.5713335201008543,0.5811388149600779,
0.5884927861044958,0.5933954335341085,0.5982980809637204,0.6007494046785262,0.6007494046785262,
0.6007494046785262,0.5982980809637204,0.5982980809637204,0.5958467572489144,0.5958467572489144,
0.5958467572489144,0.5933954335341085,0.5933954335341085,0.5933954335341085,0.5909441098193018,
0.5909441098193018,0.5909441098193018,0.5884927861044958,0.5884927861044958,0.5884927861044958,
0.5884927861044958,0.5884927861044958,0.5884927861044958,0.5860414623896899,0.5860414623896899,
0.5860414623896899,0.583590138674884,0.583590138674884,0.583590138674884,0.5811388149600779,
0.5811388149600779,0.5786874912452721,0.5762361675304661,0.5762361675304661,0.5737848438156602,
0.5737848438156602,0.5713335201008543,0.5713335201008543,0.5688821963860483,0.5688821963860483,
0.5664308726712425,0.5664308726712425,0.5639795489564365,0.5639795489564365,0.5615282252416306,
0.5590769015268245,0.5590769015268245,0.5590769015268245,0.5566255778120178,0.5566255778120178,
0.5541742540972119,0.5541742540972119,0.5517229303824059,0.5517229303824059,0.5517229303824059,
0.5492716066676,0.5492716066676,0.5492716066676,0.5492716066676,0.5468202829527942,
0.5443689592379882,0.5443689592379882,0.5443689592379882,0.5443689592379882,0.5419176355231823,
0.5394663118083762,0.5394663118083762,0.5370149880935703,0.5370149880935703,0.5345636643787645,
0.5345636643787645,0.5321123406639585,0.5321123406639585,0.5296610169491526,0.5296610169491526,
0.5272096932343466,0.5272096932343466,0.5272096932343466,0.5247583695195399,0.5247583695195399,
0.522307045804734,0.522307045804734,0.522307045804734,0.522307045804734,0.519855722089928,
0.519855722089928,0.519855722089928,0.517404398375122,0.5149530746603161,0.5125017509455102,
0.5125017509455102,0.5100504272307043,0.5100504272307043,0.5075991035158983,0.5051477798010924,
0.5051477798010924,0.5051477798010924,0.5026964560862865,0.5026964560862865,0.5026964560862865,
0.5026964560862865,0.5002451323714806,0.5002451323714806,0.49779380865667455]
d = {'x':Utilization, 'y':Efficiency}
partLoadData = pd.DataFrame(d)
partLoadData
nSegments = 4
esM.add(fn.ConversionPartLoad(esM=esM, name='PEMEC', physicalUnit=r'kW$_{el}$',
commodityConversionFactors={'electricity':-1, 'hydrogen':1},
commodityConversionFactorsPartLoad={'electricity':-1, 'hydrogen': partLoadData},
nSegments=nSegments,
hasCapacityVariable=True,
bigM=99999,
investPerCapacity=900, opexPerCapacity=900*0.01, interestRate=0.08,
economicLifetime=10))
# Glass production - Hydrogen gas furnace
capacityFix = 4985
annualLoss = 0.03 #After one year a glass melting furnace needs 3% more energy to maintain process quality due to increasing thermal losses
operationRateFix = pd.DataFrame(np.linspace(capacityFix*(1-annualLoss/2),capacityFix*(1+annualLoss/2),num=numberOfTimeSteps),columns=['GlassProductionSite'])
operationRateFix.mean()
operationRateFix = operationRateFix/operationRateFix.mean()
esM.add(fn.Conversion(esM=esM, name='hydrogenGasFurnace', physicalUnit= r'kW$_{H2}$',
commodityConversionFactors={'hydrogen':-1, 'electricity':-0.020,'O2': -0.137, # stochiometric combustion: -0.238; lambda: -0.274
'rawMaterial': -0.209, 'heat':0.070, 'CO2':0.020}, # 'CO2':0.040 for H2 with german grid; 'CO2':0.021 for 100% renewable electricity
hasCapacityVariable=True, capacityFix = capacityFix,
operationRateFix=operationRateFix,
investPerCapacity=1103.5, opexPerCapacity=652, interestRate=0.08,
economicLifetime=10))
### Sinks ###
# Heat output
esM.add(fn.Sink(esM=esM, name='Heat output', commodity='heat', hasCapacityVariable=False))
# CO2 output
esM.add(fn.Sink(esM=esM, name='CO2 output', commodity='CO2', hasCapacityVariable=False))
# O2 output
# We include this sink to enable feasibility of the optimization problem in case the electrolyzer produces slightly more oxygen than required in the hydrogen furnace (round-off error < 1%) - we only that for stochiometric combustion -> for lambda > 1 that shouldn't be necessary
esM.add(fn.Sink(esM=esM, name='O2 output', commodity='O2', hasCapacityVariable=False))
### Optimization ###
# Input parameters
timeSeriesAggregation=False
solver='glpk'
# Code
esM.optimize(timeSeriesAggregation=timeSeriesAggregation, solver=solver)
### Test ###
# Overall TAC
srcSnkSummary = esM.getOptimizationSummary("SourceSinkModel", outputLevel=1)
convSummary = esM.getOptimizationSummary("ConversionModel", outputLevel=1)
convPartloadSummary = esM.getOptimizationSummary("ConversionPartLoadModel", outputLevel=1)
TAC=(srcSnkSummary.loc[('ElectricityGrid', 'TAC', '[1 Euro/a]'),'GlassProductionSite']+
srcSnkSummary.loc[('oxygenSource', 'TAC', '[1 Euro/a]'),'GlassProductionSite']+
srcSnkSummary.loc[('rawMaterialSource', 'TAC', '[1 Euro/a]'),'GlassProductionSite']+
convSummary.loc[('hydrogenGasFurnace', 'TAC', '[1 Euro/a]'),'GlassProductionSite']+
convPartloadSummary.loc[('PEMEC', 'TAC', '[1 Euro/a]'),'GlassProductionSite'])
np.testing.assert_allclose(TAC, 14016197.2088, rtol = 0.005) # relative toerlance < 0.5%
# Electricity TAC
np.testing.assert_allclose(srcSnkSummary.loc[('ElectricityGrid', 'TAC', '[1 Euro/a]'),'GlassProductionSite'], 6.23855e+06, rtol = 0.01) # relative toerlance < 1%
# PEMEC summary
np.testing.assert_allclose(convPartloadSummary.loc[('PEMEC', 'TAC', '[1 Euro/a]'),'GlassProductionSite'], 1.46349e+06, rtol = 0.002) # relative toerlance < 0.2%
np.testing.assert_allclose(convPartloadSummary.loc[('PEMEC', 'capacity', '[kW$_{el}$]'),'GlassProductionSite'], 10225.1729065, rtol = 0.002) # relative toerlance < 0.2%
np.testing.assert_allclose(convPartloadSummary.loc[('PEMEC', 'capexCap', '[1 Euro/a]'),'GlassProductionSite'], 1371467.06108539, rtol = 0.002) # relative toerlance < 0.2%
np.testing.assert_allclose(convPartloadSummary.loc[('PEMEC', 'invest', '[1 Euro]'),'GlassProductionSite'], 9202655.61585, rtol = 0.002) # relative toerlance < 0.2%
np.testing.assert_allclose(convPartloadSummary.loc[('PEMEC', 'opexCap', '[1 Euro/a]'),'GlassProductionSite'], 92026.5561585, rtol = 0.002) # relative toerlance < 0.2%
np.testing.assert_allclose(convPartloadSummary.loc[('PEMEC', 'operation', '[kW$_{el}$*h/a]'),'GlassProductionSite'], 8.82488e+07, rtol = 0.01) # relative toerlance < 1%
# PEM operation results
opVarOptPartLoad = esM.componentModelingDict["ConversionPartLoadModel"].operationVariablesOptimum.loc[('PEMEC', 'GlassProductionSite')]
opVarOptConstLoad = [2480.73776181,2481.6941601,2482.65055839,2483.60695668,2484.56335497,2485.51975325,2486.47615154,
2487.43254983,2488.38894812,2489.34534641,2490.3017447,2491.25814299,2492.21454128,2493.17093957,2494.12733785,
2495.08373614,2496.04013443,2496.99653272,2497.95293101,2498.9093293,2499.86572759,2500.82212588,
2501.77852417,2502.73492245,2503.69132074,2504.64771903,2505.60411732,2506.56051561,2507.5169139,
2508.47331219,2509.42971048,2510.38610877,2511.34250706,2512.29890534,2513.25530363,2514.21170192,
2515.16810021,2516.1244985,2517.08089679,2518.03729508,2518.99369337,2519.95009166,2520.90648994,
2521.86288823,2522.81928652,2523.77568481,2524.7320831,2525.68848139,2526.64487968,2527.60127797,
2528.55767626,2529.51407455,2530.47047283,2531.42687112,2532.38326941,2533.3396677,2534.29606599,
2535.25246428,2536.20886257,2537.16526086,2538.12165915,2539.07805743,2540.03445572,2540.99085401,
2541.9472523,2542.90365059,2543.86004888,2544.81644717,2545.77284546,2546.72924375,2547.68564204,
2548.64204032,2549.59843861,2550.5548369,2551.51123519,2552.46763348,2553.42403177,2554.38043006,
2555.33682835,2556.29322664]
np.testing.assert_allclose(opVarOptPartLoad, opVarOptConstLoad,rtol=0.01)