Case Study

The SWSN superstructure combines a heat-integrated biorefinery supply network, a renewable electricity supply network and a food supply network—see Figure 1. A case study of supply networks spread over Central Europe is considered, the chosen countries being Austria, Czech Republic. Germany, Hungary, Poland, Slovakia and Slovenia. Com and wheat are feedstocks for food supply; solar pliotovoltaics, wind turbines and geothermal plants are used for electricity generation; and com stover, wheat straw, miscanthus, algae oil, waste cooking oil and forest residue are feedstocks for biofuel supply. The main products are food, bioethanol and green gasoline as gasoline substitutes, biodiesel and Fisclier-Tropsch (FT) diesel as diesel substitutes, as well as hydrogen and electricity. Possible technology routes are presented in Figure 4. The superstructure, which is an extension of the one shown in Figure 2, is based on four layers, L1-L4, including harvesting sites at LI, storage, pre-processing of raw materials to intermediate products and production of electricity at L2, bio-refineries at L3. and demand locations at L4 (Zore et al.. 2018b). Each layer is divided into 33 zones across Central Europe. It is assumed that locations of sites are at zone centers. Transportation is calculated within and between zones. To reduce the size of the model, it is assumed that the distances for the transportation of biomass and waste, energy and products are limited (Lam et al., 2011). For solar, wind and geothermal energy, it is assumed that they cannot be transported until they have been transformed into electrical energy. Up to 10% of the total area of each zone is assumed to be devoted to satisfying food and biofuel demand, and up to 1% to producing

Integration possibilities in a renewable energy supply network (after Zore et al., 2018b)

Figure 4. Integration possibilities in a renewable energy supply network (after Zore et al., 2018b).

electricity from renewable sources, and when all areas are combined, a total of up to 11% of the area is assumed. Note that almost 8% of the land in Europe is already used for com and wheat production (FAOSTAT, 2018). For the additional laud area (2%), an eco-cost of 0.5 €/m2 (Hendriks and Vogtlauder,

2004) is assumed, except for afforestation; in this case, an eco-benefit of 0.5 €/'m2 is considered.

The supply network model is a multi-period, mixed-integer linear programming (MILP) model, which considers monthly time periods for biomass, food and biofuels, and hourly time periods for solar, wind and geothermal energy and electricity (Cucek et al., 2016). In addition, 20 yearly periods are defined over the 20-year time horizon for replacing fossil-based fuels with biofuels, and electricity with renewable electricity. Owing to the large computational times, the number of periods has been reduced to 6 periods per month and 4 periods per day (Zore et al., 2018b). All the patterns of demand and yields are the same as in the model used for the continental-size supply network by Zore et al. (2018b).

The demand for food is assumed to remain constant throughout the year, while the demand for fuel and electricity is assumed to increase over tune. It is also assumed that all the zones in a specific country have the same demand for food, biofuel and electricity. The demand for food (com and wheat) should be satisfied in each coimtiy, and the share of demand for biofuels and renewable electricity is increased over the years, while total demand for liquid fuels is decreased (by up to 50% of current consumption in 20 years) because of the transport sector switching from liquid fuel to plug-in vehicles, and so increasing electricity consumption. Note that, on the 11% total of laud in Central Europe, it is not possible to achieve entire replacement of liquid fuel and electricity by biofuel and renewable electricity (100% of demand coming from renewable sources); thus, lower substitution levels were chosen (up to 30% for gasoline, 20% for diesel and 50% for electricity in the final year).

Data for the biorefinery and food supply network are taken from Cucek et al. (2014a, b), while for the electricity supply network, they are obtained from the U.S. Energy Information Administration (2016), and for the sake of price change projections, also from the U.S. Energy Information Administration (2013). The data related to sustainability NPY calculation are obtained from Zore et al. (2018b). All the results are based on a 3% interest rate, a 7% discount rate and a 20-year lifetime. The selected time horizon of 20 years is due to typical relatively shorter-term planning and assumed plant life of 20 years (Caputo et al., 2005). All the prices except those for fuel and electricity are kept constant, while higher redemption product prices (or subsidies) for fuels and electricity are needed in order to get some extra money, AR, in equation (3), so as to achieve positive economic viability (Ar.Pl/Ecoonuc) and sustainability (SNPV) of such large-scale production of renewably-somced electricity and biofuel. Therefore, a 2% price rise in fuel and electricity per year is introduced.

Table 1 shows the basic economic and environmental data for raw materials and products. Environmental data is represented by eco-cost coefficients as obtained from Delft University of Technology (2019). More detailed data considered in the case study are available in the literature (Cucek et al., 2014a. b), and the data related to social sustainability in Zore et al. (2017).

Table 1. Prices and eco-cost coefficients for raw materials and products.

Raw

materials

Price at year 1 ($/kg)

Eco-cost coefficient ($/kg)

Products

Price at year 1 (S/kg)

Eco-cost coefficient

(S/kg)

Com gram

0.250

0.290

Com gram

0.300

0.290

Wheat

0.300

0.100

Wheat

0.330

0.100

Com stover

0.060

0.205

Ethanol

1.115* | 1,745**

0.378

Wheat straw

0.060

0.100

Green gasolme

1.775* | 2.778**

0.379

Miscanthus

0.030

0.050

MeOH-diesel

1.169* | 1.758**

0.430

Forest residue

0.040

0.041

EtOH-diesel

1.169* | 1.758**

0.430

Algae oil

0

0.000

FT-diesel

1.329* | 1.758**

0.437

Cookmg oil

0.200

0.282

Hydrogen

1.580

1,054

Water

2.5T0"J

0.378

Products

Price at year 1 (S/kWh)

Eco-cost coefficient (S/kWh)

Ethanol

1.920***

1.989

Hydrogen

2.200***

0.150

Electricity

0.100

Methanol

0.350

0.290

- from wind turbines

0.010

- from solar photovoltaics

0.028

- fi'om geothermal plants

0.002

- from coal power plants

0.165

  • * market price.
  • ** subsidized price.
  • *** assumed prices of products are slightly higher because of transport, additional storage and other cost.

The data for calculating the number of employees for the construction, operation and maintenance of processes are taken from the JEDI models (NREL, 2015), data on earnings from Wikipedia (2016), summarized from the national statistical offices, and the number of unemployed from Eurostat (2016).

The MILP models applied to a case study of Central Europe contain about 1,200,000 constraints,

8,000,000 continuous and 10,600 binary variables. The solutions presented in the following are obtained with an optimality gap of 5% in about 72 horns of CPU time, using a GAMS/CPLEX solver on an HPC server DL580 G9 CTO with 4 processors (32 core) Intel* Xeon* CPU E5-4627 v2 @ 3.30 GHz and 768 GB RAM.

Table 2 shows the main results when maximizing Economic NPV and Sustainability NPV from a wider macroeconomic perspective. Note that both objective values are significant, 194,517 M€ is maximal Economic NPV, and 378,438 M6 is maximal SNPV, of which Economic NPY is 130,589 M€, Eco NPY 106,359 M€ and Social NPV 141,491 M€. Note also that, in the calculation of Eco NPY, see equation (8), values of NPV of eco-benefit (EB), NPV of eco-cost (EC) and that from higher redemption prices for biofuels and electricity from alternative resources (AR) are 501,295 M€, 286,462 M€ and 108,474 M€. Rather than spending 501,295 M€ of eco-cost to avoid burdening the environment caused by fossil-based production plants (which corresponds to EB), new renewable-based production plants are proposed to be installed with eco-cost being significantly lower, 286,462 M€. The difference of the two corresponds to gross Eco NPV of 214,883 M€, from which 108,474 M€ are deducted, which are earned by higher redemption prices, thus obtaining the final value of Eco NPV (106,359 M€). Note that, by charging 286,462 M€ of eco-cost to new renewable alternatives, an assumption of zero waste operation is imposed (considering solely the waste inside the company gate; Zore et al., 2017). Without that, the final value of SNPV would be even higher. This assumption is important because it enables through maximization of SNPV identifying those unburdening alternatives that at the same time burden the environment the least. The final interpretation of the maximal SNPV value of 378,438 M€ is that in order to make modifications of this supply network producing food, fuel and electricity sustainable, instead of spending 501,295 M€ to avoid burdening the environment, owing to current, mainly fossil- based energy supply, 286,462 M€ would be spend for zero waste operation of renewable-based supply networks and 108,474 M€ would be charged through higher redemption prices (or subsidies) in order to make new production plants economically viable, while still achieving Economic NPV of 130,589 M€ and Social NPV of 141,491 M€, both calculated from a wider-macroeconomic perspective.

Different trade-offs between Economic NPV and eco-NPYs are obtained when optimization is performed with economic and sustainability criteria. When maximizing the economic criterion, higher дгррЕсомтис, kut significantly smaller NPV*-'0 and APE501 are obtained than in the case of maximizing the sustainability criterion. When maximizing SNPV, significant increases in NP1/£co (for 89,848 M€) and ATE”’1 (for 123,788 M€) are obtained, which together more than triple the increase, compared to the decrease in APEEcouomlc (decrease of 63,928 M€), giving rise to a significant increase in SNPV—from 228,731 M€ to 378,439 M€.

When maximizing SNPV, the results show that even with their lower economic NPV value, renewable technologies are still the more sustainable solution when producing biofuel and electricity. The main differences between the most economic and the most sustainable solutions for satisfying market demand occur in the selection of raw materials, the type of product and the locations selected. Afforestation instead of miscanthus cultivation is the main reason for higher NP1/*'°. Maximizing SNPV largely promotes a

Table 2. Results when maximising economic and sustainability net present values.

Economic items

Maximization criteria

дrp yr.Economic

SNPV

A7>FEconomi£ (M€)

194,517

130,589

NPV*" (M€)

16,511

106,359

Дгр J/Sodal (M€)

17,703

141,491

SNPV (M€)

228,731

378,439

circular economy, renewable electricity production and afforestation, and thus solutions that allow for gr eater unburdening, and create less of a burden on the environment.

Table 3 shows the main results from the optimization of Economic and Sustainability NPYs in terms of area used, food supply, demand for fuel substitutes and electricity, the raw materials and technology used to produce renewable fuel and electricity and the number of employees in renewable energy production.

All the available area (10%) is used for food and biofuel production or for additional afforestation. The sustainability perspective would suggest that up to about 1.5% of the area be afforested, given the eco-benefits related to afforestation, as explained above. However, in year 20, the percentage of afforested area is lower, by about 0.01%, owing to the higher demands. On the contrary, from an economic perspective, the afforested area suggested is negligible.

With both criteria, the demand assumed for food, fuel and electricity is satisfied in all the years. Wien maximizing jVPPEc““mK, gasoline substitute requirements in the first year are satisfied by ethanol alone, while when SNPV is maximized, green gasoline is also produced. The capacity of both ethanol and gr een gasoline mainly increases over the years. For the substitution of diesel fuel, both maximization criteria suggest producing FT-diesel in larger quantities than biodiesel from waste cooking oil using methanol as a catalyst.

The capacity of both waste cooking oil and biodiesel production remains constant over the years. The biggest difference in raw material use is obtained for miscanthus and forest residue. A'PPEcoomlc prefers miscanthus because of its higher yield and lower price, and thus good economic performance, while SNPV prefers forest residue (afforestation) because of its sustainability. In the final year, when the share of fuel substitutes is the highest, both optimization criteria mainly suggest miscanthus cultivation and use.

Various technologies have been selected to produce biofuel and electricity. Among the technologies producing biofuel are gasification of lignocellulosic biomass and further syngas fermentation, FT synthesis for production of bioethauol, hydrogen, green gasoline and FT diesel, and transesterificatiou of waste cooking oil with methanol as a catalyst. To produce renewable electricity, wind turbines are preferred over solar photovoltaics and geothermal power plants because of their better economic performance.

Table 3 shows that when maximizing SNPV, almost twice the number of employees is required in the final year as compared to the case when maximizing Л'РГ/Ес<шо,шс. The requirements to satisfy demand over the years are met with about 295,000 employees when maximizing NPV^com!m<: and with about

429.000 employees when maximizing SNPV. Most of the workers are required in the construction and manufacturing sectors for electricity production, specifically in building and operating wind turbines and photovoltaic panels. A higher number of employees is required for constructing and operating photovoltaic panels than for wind turbines and geothermal plants. From an economic perspective, only a small percentage of electricity demand is, thus, satisfied with solar photovoltaics.

Figure 5 shows suggested optimal production of gasoline substitutes over the next 20 years across Central Europe when maximizing SNPV. In year 1, production takes place mainly in Germany, western Austria and central Hungary. In year 10, when the demand for gasoline substitutes increases to 20%, higher amounts of biofuel are proposed in Germany and western Austria and in 2 new locations in Poland, but no longer in central Hungary. In the final year, with a further increase in production to 30%, an increase in the production capacity of already selected locations is suggested.

Figure 6 shows the distribution of diesel substitutes across Central Europe. At the start of the project, when 10% of diesel fuel should be replaced, the selected biorefineries for biodiesel production are allocated in the same locations as for gasoline substitutes. This is due to the lower transport and investment cost if biorefineries producing renewable ethanol and diesel are at the same location. In year 10, the increase in capacity is mainly observed in parts of Germany, western Austria and in 2 new locations in Poland. In the final year, when 20% of current diesel consumption should be replaced by renewable diesel fuel, the same locations are suggested but with increased capacity.

Figure 7 shows the distribution of hydrogen production over the years. Hydrogen is produced as a main product by gasification and hydrogen production from lignocellulosic biomass (Martin and Grossmann, 201 lb), and as a by-product by gasification and syngas fermentation, gasification and catalytic

Table 3. Main results of economic and sustainability optimisations.

Maximization criteria

Supply network items

fypyEcoMBfc

SNPV

Year 1

Year 10

Year 20

Year 1

Year 10

Year 20

Ar ea used (%)

11.00

11.00

11.00

11.00

11.00

11.00

afforested (%)

0.003

0.003

0.003

1.53

1.53

0.013

Food (kt/v)

com grain

24,769

24,769

24,769

24,769

24,769

24,769

wheat

37,226

37,226

37,226

37,226

37,226

37,226

Food demand satisfied (°/o)

100,00

100.00

100.00

100.00

100.00

100,00

Fuel demand satisfied (%)

gasoline

10,00

20.00

30.00

10.00

20,00

30.00

diesel

10,00

15.00

20.00

10.00

15,00

20.00

Electricity demand satisfied (%)

produced total

10,00

30.00

50.00

10.00

30,00

50.00

from wmd

9.22

26.82

30.16

5.52

15.67

24.70

from solar

0.78

3.18

19.84

4.48

14.33

25.30

from geothermal

0

0

0

0

0

0

Raw materials (kt/v)

com stover

13,235

14,405

14,862

14,862

14,034

13,164

wheat stiaw

37,970

37,970

37,970

37,970

37,970

37,970

miscanthus

27,520

30,163

33,663

0

0

32,556

forest residue

0

0

0.369

18.365

26.512

1.441

algae

0

0

0

0

0

0

cookmg oil

1,301

1,301

1,301

1,556

1,556

1,556

Technologies:[1]

hydrogen1

dry-grind process2

syngas fermentation2

catalytic synthesis4

FT synthesis5

WCO methanol4

WOO ethanol7

algae methanol8

algae ethanol”

photovoltaics (km2)

25

102

576

136

445

820

wind turbines

8,714

35,066

44,169

4,757

16,637

31,802

binary cycle geothermal plants

0

0

0

0

0

0

Biofuels (kt/y)

ethanol

4,469

4,680

9,575

2,917

6,332

9,747

green gasoline

2,673

2,406

974

1,636

2,297

et-diesel[2]

me-diesel***

1,248

1,248

1,248

1,493

1,493

1,493

FT-diesel

3,880

6,369

8,858

3,665

6,154

8,643

hydrogen

4,197

4,140

1,844

1,599

1,374

1,348

Number of employees

23,192

51,086

295,145

79,265

237,785

428,730

Distribution of gasoline substitute production in years 1,10 and 20 when maximizing SNPV

Figure 5. Distribution of gasoline substitute production in years 1,10 and 20 when maximizing SNPV.

Distribution of diesel substitute production in years 1,10 and 20 when maximizing SNPV

Figure 6. Distribution of diesel substitute production in years 1,10 and 20 when maximizing SNPV.

Distribution of hydrogen production in years 1,10 and 20 when maximizing SNPV

Figure 7. Distribution of hydrogen production in years 1,10 and 20 when maximizing SNPV.

Distr ibution of installed wind turbines in years 1, 10 and 20 when maximizing SNPV

Figure 8. Distr ibution of installed wind turbines in years 1, 10 and 20 when maximizing SNPV

synthesis (Martin and Grossmann, 2011a) and by gasification. FT synthesis and hydrocracking (Martin and Grossmann, 2011c) of lignocellulosic biomass. There were no requirements set for hydrogen itself; however, hydrogen could potentially play an important role in various sectors: electricity, heat, industry, transport and energy storage (Staffell et al., 2019). The results in terms of hydrogen production show that its production is reduced when the production of biofuel increases, owing to the lower production of hydrogen as a main product. The locations for hydrogen production are, in principle, the same as those for the production of biofuel.

Figure 8 shows the locations of wind turbines in years 1,10 and 20 when maximizing SNPV. The wind, as a renewable source of electricity, takes precedence over photovoltaics, since it enables electricity production over nights and in winters. In Central Europe, there are only moderate solar resources and low ambient temperatures, especially in winter, therefore, electricity potential from photovoltaics is limited (Suri et ah, 2007).

In year 1. when 10% of electricity must be provided from renewable sources, 4,757 wind turbines are proposed, located in northern Germany and in continental locations with higher wind speed (eastern Poland). When demand for electricity rises to 30% of the final demand (including increases due to energy shift from liquid fuel to electricity), several new locations in central Germany, western Slovakia, western Austria and Poland are selected. At the same time, the number of wind turbines at previously selected locations, is increased, mainly in eastern Poland. In year 20, when the demand for renewable electricity rises to 50% (including transportation fuel shift), a total of 31,802 wind turbines are installed. Besides all the previous locations, in this scenario, wind turbines are selected in most of the zones in Germany, in several zones in Poland and in the western Czech Republic.

The locations of photovoltaic panel installations in years 1,10 and 20 when maximizing .SAP Fare shown in Figure 9.

In year 1, when 10% of renewable electricity should be supplied, the selected locations for panels are mainly in southern Germany, with a few in northern Germany and eastern Poland, for an area of 136 knr in total. In year 10, with the increase in demand to 30%, the suggested locations for panels are in southern, central and eastern Germany, in two zones in Poland, western Austria and the western Czech Republic. To achieve a 50% share of renewable electricity (including additional demand due to the move towards electric vehicles) in year 20, it is suggested that 820 km1 of photovoltaic panels be installed. Photovoltaic installations are proposed for various zones in Germany. Poland, western Czech Republic, western Austria and western Slovakia. It was demonstrated that, in order to meet the high demand for renewable energy in the future, a combination of technologies should be used.

  • [1] technologies: 1 gasification and lignocellulosic hydrogen production (Martin and Grossmann, 2011a, b), 2 dry-gnnd process (Karuppiali et al., 2008), 3 gasification and syngas fermentation and 4 gasification and catalytic synthesis of lignocellulosic biomass (Martin andGrossmann, 2011a, b), 5 gasification, FT synthesis and hydrocracking (Martin and Grossmann, 201 lc), 5 biodiesel production fi'om waste cookmg oil with methanol, and7 ethanol, and from8 algal oil with methanol (Martin andGrossmann, 2012), and9 ethanol (Severson et al., 2013);
  • [2] biodiesel produced using ethanol and *** methanol as alcohol.
 
Source
< Prev   CONTENTS   Source   Next >