Workflow for a single node energy system¶

In this application of the ETHOS.FINE framework, a single region energy system is modeled and optimized.

All classes which are available to the user are utilized and examples of the selection of different parameters within these classes are given.

The workflow is structures as follows:

Required packages are imported and the input data path is set
An energy system model instance is created
Commodity sources are added to the energy system model
Commodity conversion components are added to the energy system model
Commodity storages are added to the energy system model
Commodity sinks are added to the energy system model
The energy system model is optimized
Selected optimization results are presented

1. Import required packages and set input data path¶

The ETHOS.FINE framework is imported which provides the required classes and functions for modeling the energy system.

In [2]:

Copied!





import fine as fn
from getData import getData
from pathlib import Path


cwd = Path.cwd()
data = getData()

%matplotlib inline
%load_ext autoreload
%autoreload 2
import fine as fn
from getData import getData
from pathlib import Path


cwd = Path.cwd()
data = getData()

%matplotlib inline
%load_ext autoreload
%autoreload 2

2. Create an energy system model instance¶

The structure of the energy system model is given by the considered locations, commodities, the number of time steps as well as the hours per time step.

The commodities are specified by a unit (i.e. 'GW_electric', 'GW_H2lowerHeatingValue', 'Mio. t CO2/h') which can be given as an energy or mass unit per hour. Furthermore, the cost unit and length unit are specified.

In [3]:

Copied!





locations = {"GermanyRegion"}
commodityUnitDict = {"electricity": r"GW$_{el}$", "hydrogen": r"GW$_{H_{2},LHV}$"}
commodities = {"electricity", "hydrogen"}
numberOfTimeSteps = 8760
hoursPerTimeStep = 1
locations = {"GermanyRegion"}
commodityUnitDict = {"electricity": r"GW$_{el}$", "hydrogen": r"GW$_{H_{2},LHV}$"}
commodities = {"electricity", "hydrogen"}
numberOfTimeSteps = 8760
hoursPerTimeStep = 1

In [4]:

Copied!





esM = fn.EnergySystemModel(
    locations=locations,
    commodities=commodities,
    numberOfTimeSteps=8760,
    commodityUnitsDict=commodityUnitDict,
    hoursPerTimeStep=1,
    costUnit="1e9 Euro",
    lengthUnit="km",
    verboseLogLevel=0,
)
esM = fn.EnergySystemModel(
    locations=locations,
    commodities=commodities,
    numberOfTimeSteps=8760,
    commodityUnitsDict=commodityUnitDict,
    hoursPerTimeStep=1,
    costUnit="1e9 Euro",
    lengthUnit="km",
    verboseLogLevel=0,
)

3. Add commodity sources to the energy system model¶

3.1. Electricity sources¶

Wind onshore¶

In [5]:

Copied!





esM.add(
    fn.Source(
        esM=esM,
        name="Wind (onshore)",
        commodity="electricity",
        hasCapacityVariable=True,
        operationRateMax=data["Wind (onshore), operationRateMax"],
        capacityMax=data["Wind (onshore), capacityMax"],
        investPerCapacity=1.1,
        opexPerCapacity=1.1 * 0.02,
        interestRate=0.08,
        economicLifetime=20,
    )
)
esM.add(
    fn.Source(
        esM=esM,
        name="Wind (onshore)",
        commodity="electricity",
        hasCapacityVariable=True,
        operationRateMax=data["Wind (onshore), operationRateMax"],
        capacityMax=data["Wind (onshore), capacityMax"],
        investPerCapacity=1.1,
        opexPerCapacity=1.1 * 0.02,
        interestRate=0.08,
        economicLifetime=20,
    )
)

Full load hours:

In [25]:

nbval-check-output

Copied!

print(data["Wind (onshore), operationRateMax"].sum())
print(data["Wind (onshore), operationRateMax"].sum())

2300.4069071646272

4. Add conversion components to the energy system model¶

New combined cycly gas turbines for hydrogen¶

In [7]:

Copied!





esM.add(
    fn.Conversion(
        esM=esM,
        name="New CCGT plants (hydrogen)",
        physicalUnit=r"GW$_{el}$",
        commodityConversionFactors={"electricity": 1, "hydrogen": -1 / 0.6},
        hasCapacityVariable=True,
        investPerCapacity=0.7,
        opexPerCapacity=0.021,
        interestRate=0.08,
        economicLifetime=33,
    )
)
esM.add(
    fn.Conversion(
        esM=esM,
        name="New CCGT plants (hydrogen)",
        physicalUnit=r"GW$_{el}$",
        commodityConversionFactors={"electricity": 1, "hydrogen": -1 / 0.6},
        hasCapacityVariable=True,
        investPerCapacity=0.7,
        opexPerCapacity=0.021,
        interestRate=0.08,
        economicLifetime=33,
    )
)

Electrolyzers¶

In [8]:

Copied!





esM.add(
    fn.Conversion(
        esM=esM,
        name="Electroylzers",
        physicalUnit=r"GW$_{el}$",
        commodityConversionFactors={"electricity": -1, "hydrogen": 0.7},
        hasCapacityVariable=True,
        investPerCapacity=0.5,
        opexPerCapacity=0.5 * 0.025,
        interestRate=0.08,
        economicLifetime=10,
    )
)
esM.add(
    fn.Conversion(
        esM=esM,
        name="Electroylzers",
        physicalUnit=r"GW$_{el}$",
        commodityConversionFactors={"electricity": -1, "hydrogen": 0.7},
        hasCapacityVariable=True,
        investPerCapacity=0.5,
        opexPerCapacity=0.5 * 0.025,
        interestRate=0.08,
        economicLifetime=10,
    )
)

5. Add commodity storages to the energy system model¶

5.1. Electricity storage¶

Lithium ion batteries¶

The self discharge of a lithium ion battery is here described as 3% per month. The self discharge per hours is obtained using the equation (1-$\text{selfDischarge}_\text{hour})^{30*24\text{h}} = 1-\text{selfDischarge}_\text{month}$.

In [9]:

Copied!





esM.add(
    fn.Storage(
        esM=esM,
        name="Li-ion batteries",
        commodity="electricity",
        hasCapacityVariable=True,
        chargeEfficiency=0.95,
        cyclicLifetime=10000,
        dischargeEfficiency=0.95,
        selfDischarge=1 - (1 - 0.03) ** (1 / (30 * 24)),
        chargeRate=1,
        dischargeRate=1,
        doPreciseTsaModeling=False,
        investPerCapacity=0.151,
        opexPerCapacity=0.002,
        interestRate=0.08,
        economicLifetime=22,
    )
)
esM.add(
    fn.Storage(
        esM=esM,
        name="Li-ion batteries",
        commodity="electricity",
        hasCapacityVariable=True,
        chargeEfficiency=0.95,
        cyclicLifetime=10000,
        dischargeEfficiency=0.95,
        selfDischarge=1 - (1 - 0.03) ** (1 / (30 * 24)),
        chargeRate=1,
        dischargeRate=1,
        doPreciseTsaModeling=False,
        investPerCapacity=0.151,
        opexPerCapacity=0.002,
        interestRate=0.08,
        economicLifetime=22,
    )
)

5.2. Hydrogen storage¶

Hydrogen filled salt caverns¶

The maximum capacity is here obtained by: dividing the given capacity (which is given for methane) by the lower heating value of methane and then multiplying it with the lower heating value of hydrogen.

In [10]:

Copied!





esM.add(
    fn.Storage(
        esM=esM,
        name="Salt caverns (hydrogen)",
        commodity="hydrogen",
        hasCapacityVariable=True,
        capacityVariableDomain="continuous",
        capacityPerPlantUnit=133,
        chargeRate=1 / 470.37,
        dischargeRate=1 / 470.37,
        sharedPotentialID="Existing salt caverns",
        stateOfChargeMin=0.33,
        stateOfChargeMax=1,
        capacityMax=data["Salt caverns (hydrogen), capacityMax"],
        investPerCapacity=0.00011,
        opexPerCapacity=0.00057,
        interestRate=0.08,
        economicLifetime=30,
    )
)
esM.add(
    fn.Storage(
        esM=esM,
        name="Salt caverns (hydrogen)",
        commodity="hydrogen",
        hasCapacityVariable=True,
        capacityVariableDomain="continuous",
        capacityPerPlantUnit=133,
        chargeRate=1 / 470.37,
        dischargeRate=1 / 470.37,
        sharedPotentialID="Existing salt caverns",
        stateOfChargeMin=0.33,
        stateOfChargeMax=1,
        capacityMax=data["Salt caverns (hydrogen), capacityMax"],
        investPerCapacity=0.00011,
        opexPerCapacity=0.00057,
        interestRate=0.08,
        economicLifetime=30,
    )
)

7. Add commodity sinks to the energy system model¶

7.1. Electricity sinks¶

Electricity demand¶

In [11]:

Copied!





esM.add(
    fn.Sink(
        esM=esM,
        name="Electricity demand",
        commodity="electricity",
        hasCapacityVariable=False,
        operationRateFix=data["Electricity demand, operationRateFix"],
    )
)
esM.add(
    fn.Sink(
        esM=esM,
        name="Electricity demand",
        commodity="electricity",
        hasCapacityVariable=False,
        operationRateFix=data["Electricity demand, operationRateFix"],
    )
)

7.2. Hydrogen sinks¶

Fuel cell electric vehicle (FCEV) demand¶

In [12]:

Copied!





FCEV_penetration = 0.5
esM.add(
    fn.Sink(
        esM=esM,
        name="Hydrogen demand",
        commodity="hydrogen",
        hasCapacityVariable=False,
        operationRateFix=data["Hydrogen demand, operationRateFix"] * FCEV_penetration,
    )
)
FCEV_penetration = 0.5
esM.add(
    fn.Sink(
        esM=esM,
        name="Hydrogen demand",
        commodity="hydrogen",
        hasCapacityVariable=False,
        operationRateFix=data["Hydrogen demand, operationRateFix"] * FCEV_penetration,
    )
)

8. Optimize energy system model¶

All components are now added to the model and the model can be optimized. If the computational complexity of the optimization should be reduced, the time series data of the specified components can be clustered before the optimization and the parameter timeSeriesAggregation is set to True in the optimize call.

In [13]:

Copied!

esM.aggregateTemporally(numberOfTypicalPeriods=30)
esM.aggregateTemporally(numberOfTypicalPeriods=30)

Clustering time series data with 30 typical periods and 24 time steps per period 
further clustered to 12 segments per period...
		(5.4925 sec)

In [14]:

Copied!

esM.optimize(
    timeSeriesAggregation=True, solver=fn.utils.ImplementedSolvers.STANDARD_SOLVER.value
)
esM.optimize(
    timeSeriesAggregation=True, solver=fn.utils.ImplementedSolvers.STANDARD_SOLVER.value
)

Time series aggregation specifications:
Number of typical periods:30, number of time steps per period:24, number of segments per period:12

Declaring sets, variables and constraints for SourceSinkModel
	declaring sets... 
	declaring variables... 
	declaring constraints... 
		(0.1647 sec)

Declaring sets, variables and constraints for ConversionModel
	declaring sets... 
	declaring variables... 
	declaring constraints... 
		(0.0488 sec)

Declaring sets, variables and constraints for StorageModel
	declaring sets... 
	declaring variables... 
	declaring constraints... 
		(0.3974 sec)

Declaring shared potential constraint...
		(0.0003 sec)

Declaring linked component quantity constraint...
		(0.0000 sec)

Declaring commodity balances...
		(0.0161 sec)

		(0.0000 sec)

Declaring objective function...
		(0.1446 sec)

GLPSOL--GLPK LP/MIP Solver 5.0
Parameter(s) specified in the command line:
 --write C:\Users\JA60A~1.BEL\AppData\Local\Temp\tmppz54exqh.glpk.raw --wglp
 C:\Users\JA60A~1.BEL\AppData\Local\Temp\tmp5047l8ia.glpk.glp --cpxlp C:\Users\JA60A~1.BEL\AppData\Local\Temp\tmp5rg4_8uf.pyomo.lp
Reading problem data from 'C:\Users\JA60A~1.BEL\AppData\Local\Temp\tmp5rg4_8uf.pyomo.lp'...
8339 rows, 4892 columns, 22821 non-zeros
52744 lines were read
Writing problem data to 'C:\Users\JA60A~1.BEL\AppData\Local\Temp\tmp5047l8ia.glpk.glp'...
43104 lines were written
GLPK Simplex Optimizer 5.0
8339 rows, 4892 columns, 22821 non-zeros
Preprocessing...
8203 rows, 4097 columns, 21770 non-zeros
Scaling...
 A: min|aij| =  6.507e-04  max|aij| =  4.545e+02  ratio =  6.986e+05
GM: min|aij| =  2.576e-01  max|aij| =  3.883e+00  ratio =  1.508e+01
EQ: min|aij| =  6.633e-02  max|aij| =  1.000e+00  ratio =  1.508e+01
Constructing initial basis...
Size of triangular part is 7111
      0: obj =   0.000000000e+00 inf =   1.583e+03 (745)
   1913: obj =   1.175060694e+01 inf =   1.847e-13 (0) 16
*  3297: obj =   7.820770238e+00 inf =   5.285e-13 (0) 13
OPTIMAL LP SOLUTION FOUND
Time used:   0.8 secs
Memory used: 10.0 Mb (10533487 bytes)
Writing basic solution to 'C:\Users\JA60A~1.BEL\AppData\Local\Temp\tmppz54exqh.glpk.raw'...
13240 lines were written

Status: ok
Termination condition: optimal
Statistics: 
  Branch and bound: 
    Number of bounded subproblems: 0
    Number of created subproblems: 0
Error rc: 0
Time: 0.9601352214813232


Name: unknown
Lower bound: 7.82077023780424
Upper bound: 7.82077023780424
Number of objectives: 1
Number of constraints: 8339
Number of variables: 4892
Number of nonzeros: 22821
Sense: minimize

Solve time: 1.6836094856262207 sec.

Processing optimization output...
for SourceSinkModel ...(0.3352sec)
for ConversionModel ...(0.2209sec)

C:\Programming\FINE\fine\storage.py:2056: UserWarning: Charge and discharge at the same time for component Salt caverns (hydrogen)
  warnings.warn(

for StorageModel ...  (0.7136sec)
		(1.2779 sec)

9. Selected results output¶

Sources and Sink¶

Show optimization summary

In [15]:

Copied!

esM.getOptimizationSummary("SourceSinkModel", outputLevel=2)
esM.getOptimizationSummary("SourceSinkModel", outputLevel=2)

Out[15]:

			GermanyRegion
Component	Property	Unit
Electricity demand	operation	[GW$_{el}$*h/a]	30957.888055
Electricity demand	operation	[GW$_{el}$*h]	30957.888055
Hydrogen demand	operation	[GW$_{H_{2},LHV}$*h/a]	4765.074877
Hydrogen demand	operation	[GW$_{H_{2},LHV}$*h]	4765.074877
Wind (onshore)	NPVcontribution	[1e9 Euro]	5.545377
	TAC	[1e9 Euro/a]	5.545377
	capacity	[GW$_{el}$]	41.371855
	capexCap	[1e9 Euro/a]	4.635196
	commissioning	[GW$_{el}$]	41.371855
	invest	[1e9 Euro]	45.50904
	operation	[GW$_{el}$*h/a]	44223.602817
	operation	[GW$_{el}$*h]	44223.602817
	opexCap	[1e9 Euro/a]	0.910181

Plot operation time series (either one or two dimensional)

In [16]:

Copied!

fig, ax = fn.plotOperation(esM, "Electricity demand", "GermanyRegion")
fig, ax = fn.plotOperation(esM, "Electricity demand", "GermanyRegion")

No description has been provided for this image

In [17]:

Copied!

fig, ax = fn.plotOperationColorMap(esM, "Electricity demand", "GermanyRegion")
fig, ax = fn.plotOperationColorMap(esM, "Electricity demand", "GermanyRegion")

Conversion¶

Show optimization summary

In [18]:

Copied!

esM.getOptimizationSummary("ConversionModel", outputLevel=2)
esM.getOptimizationSummary("ConversionModel", outputLevel=2)

Out[18]:

			GermanyRegion
Component	Property	Unit
Electroylzers	NPVcontribution	[1e9 Euro]	0.424378
	TAC	[1e9 Euro/a]	0.424378
	capacity	[GW$_{el}$]	4.877081
	capexCap	[1e9 Euro/a]	0.363414
	commissioning	[GW$_{el}$]	4.877081
	invest	[1e9 Euro]	2.438541
	operation	[GW$_{el}$*h/a]	16851.368489
	operation	[GW$_{el}$*h]	16851.368489
	opexCap	[1e9 Euro/a]	0.060964
New CCGT plants (hydrogen)	NPVcontribution	[1e9 Euro]	0.124811
	TAC	[1e9 Euro/a]	0.124811
	capacity	[GW$_{el}$]	1.525874
	capexCap	[1e9 Euro/a]	0.092767
	commissioning	[GW$_{el}$]	1.525874
	invest	[1e9 Euro]	1.068112
	operation	[GW$_{el}$*h/a]	4218.529839
	operation	[GW$_{el}$*h]	4218.529839
	opexCap	[1e9 Euro/a]	0.032043

In [19]:

Copied!

fig, ax = fn.plotOperationColorMap(esM, "New CCGT plants (hydrogen)", "GermanyRegion")
fig, ax = fn.plotOperationColorMap(esM, "New CCGT plants (hydrogen)", "GermanyRegion")

Storage¶

Show optimization summary

In [20]:

Copied!

esM.getOptimizationSummary("StorageModel", outputLevel=2)
esM.getOptimizationSummary("StorageModel", outputLevel=2)

Out[20]:

			GermanyRegion
Component	Property	Unit
Li-ion batteries	NPVcontribution	[1e9 Euro]	1.010062
	TAC	[1e9 Euro/a]	1.010062
	capacity	[GW$_{el}$*h]	60.112569
	capexCap	[1e9 Euro/a]	0.889837
	commissioning	[GW$_{el}$*h]	60.112569
	invest	[1e9 Euro]	9.076998
	operationCharge	[GW$_{el}$*h/a]	6384.378729
	operationCharge	[GW$_{el}$*h]	6384.378729
	operationDischarge	[GW$_{el}$*h/a]	5751.502616
	operationDischarge	[GW$_{el}$*h]	5751.502616
	opexCap	[1e9 Euro/a]	0.120225
Salt caverns (hydrogen)	NPVcontribution	[1e9 Euro]	0.716143
	TAC	[1e9 Euro/a]	0.716143
	capacity	[GW$_{H_{2},LHV}$*h]	1235.216115
	capexCap	[1e9 Euro/a]	0.012069
	commissioning	[GW$_{H_{2},LHV}$*h]	1235.216115
	invest	[1e9 Euro]	0.135874
	operationCharge	[GW$_{H_{2},LHV}$*h/a]	10272.356153
	operationCharge	[GW$_{H_{2},LHV}$*h]	10272.356153
	operationDischarge	[GW$_{H_{2},LHV}$*h/a]	10272.356153
	operationDischarge	[GW$_{H_{2},LHV}$*h]	10272.356153
	opexCap	[1e9 Euro/a]	0.704073

In [21]:

Copied!





fig, ax = fn.plotOperationColorMap(
    esM,
    "Li-ion batteries",
    "GermanyRegion",
    variableName="stateOfChargeOperationVariablesOptimum",
)
fig, ax = fn.plotOperationColorMap(
    esM,
    "Li-ion batteries",
    "GermanyRegion",
    variableName="stateOfChargeOperationVariablesOptimum",
)

In [22]:

Copied!





fig, ax = fn.plotOperationColorMap(
    esM,
    "Salt caverns (hydrogen)",
    "GermanyRegion",
    variableName="stateOfChargeOperationVariablesOptimum",
)
fig, ax = fn.plotOperationColorMap(
    esM,
    "Salt caverns (hydrogen)",
    "GermanyRegion",
    variableName="stateOfChargeOperationVariablesOptimum",
)