System Perspective on Biogas Use for Transport and Electricity Production

energies

Article System Perspective on Biogas Use for Transport and Electricity Production

Tommy Rosén * and Louise Ödlund Division of Energy Systems, Linköping University, SE-581 83 Linköping, Sweden; [email protected] * Correspondence: [email protected]

Received: 26 September 2019; Accepted: 29 October 2019; Published: 31 October 2019

Abstract: Linköping municipality has managed biogas driven buses in the regional transport system since 1997 and all buses in the municipality have run on biogas since 2015. Biogas is a renewable fuel and by replacing fossil fuels it can help to lower net CO2 emissions. However, Internal Combustion Engines (ICE) in buses still have a rather low efficiency, in the range of 15–30%. If the combustion of biogas instead takes place in a combined cycle gas turbine (CCGT) efficiency could be higher and heat losses reduced. This could be a feasible solution if the transport system instead used electric buses charged with electricity generated by the CCGT. This article has a top-down perspective on the regional transport system and the regional district heating system (DHS) in Linköping municipality. Two alternative systems are compared regarding CO2 emissions, electricity production and component efficiencies. The first system that is studied is in operation today and uses locally produced biogas in the ICE buses. In parallel the combined heat and power (CHP) system delivers electricity and heat to households in the region. The second system that is studied is a system with electric buses and a CHP system that uses biogas in the CCGT to deliver electricity and heat to the regional power grid and DHS. The study shows that emissions would be reduced if biogas use is changed from use in ICE buses to use in the CCGT in the CHP-DHS. Improved biogas use could lower CO2-eq emissions by 2.4 million kg annually by using a better fuel-energy pathway.

Keywords: District heating; system perspective; electric buses; biogas; smart energy systems

1. Introduction Biogas is an energy carrier that is produced by anaerobic digestion of organic material, often various kinds of food waste. Biogas can be used in several different ways to replace fossil fuels and hence reduce the emission of greenhouse gases (GHG). The reduction in GHG emissions compared to fossil fuels is estimated to be between 67% and 148%, these reduction values include system expansion consequences such as changed land use and fertilizer substitution (organic fertilizer can be a residue from biogas production) [1]. In 2016, biogas production in Sweden amounted to 2 TWh [2], which corresponds to 0.3% of the total energy used in Sweden that year [3]. The most common use of biogas in 2016 was as vehicle fuel at 1145 GWh and the second most common use was for heating at 400 GWh [2]. In a modelling assessment of cost-effective biogas utilization, Börjesson et al. [4] conclude that only about 10–16% of the technical biogas potential will be utilized without subsidies. However, biogas utilization of about 90% is predicted with subsidies of EUR 40–60/MWh. At the end of 2017, there were 55,117 gas vehicles in Sweden, of which 2533 were buses, 854 other heavy vehicles and the rest were cars [5]. When gas vehicles were first introduced, almost all the gas came from fossil natural gas. Since then there has been a steady trend whereby biogas has increased its share and in 2017 it accounted for 86% of the gas fuels in Sweden [5]. Fallde et al. [6] describe the biogas bus development in Linköping municipality during 1976–2015, a transition from a small niche to a new socio-technical regime with all city buses powered by biogas.

Energies 2019, 12, 4159; doi:10.3390/en12214159 www.mdpi.com/journal/energies Energies 2019, 12, 4159 2 of 13 Energies 2019, 12, x FOR PEER REVIEW 2 of 14

45 LargeLarge truck truck companies companies such such as Scania as Scania and and Volvo Volvo are areconstantly constantly trying trying to improve to improve the thegas gas engine engine 46 butbut it is it still is still not notas efficient as efficient as the as thediesel diesel engine, engine, although although the thedifference difference is now is now just just a few a few percent percent [5]. [5]. 47 EngineEngine development development will will continue continue but but the the laws laws of of thermodynamics thermodynamics also also set set an an absolute absolute upper upper limit limit to 48 to allall heat-based engines. HeatHeat losseslosses throughthrough the the cooling cooling system, system, the the exhaust exhaust pipe pipe and and the the engine engine block 49 blockwill will continue continue to be to abe large a large part part of the of the energy energy flow flow from from gas andgas and diesel diesel engines. engines. A useful A useful energy energy service 50 servicederived derived from from these these heat heat losses losses is hard is hard to envision. to envision. 51 AfterAfter real-world real-world testing testing of electric of electric buses in buses eight inSwedish eight municipalities Swedish municipalities in 2014–2015, in Borén 2014–2015, et 52 al. [7]Bor concludedén et al. [7 that] concluded the tested that buses the had tested low buses energy had use low and energy low external use and noise low levels. external The noise average levels. 53 energyThe averageuse in the energy tests was use between in the tests 0.86 was and between 1.02 kWh/km 0.86 and and 1.02 passengers kWh/km and and drivers passengers gave andpositive drivers 54 feedbackgave positive about the feedback electric about buses the [7]. electric At the buses begi [nning7]. At theof 2017, beginning Sweden of 2017,had Sweden41 electric had buses 41 electric in 55 commercialbuses in commercialtraffic [8] and tra ffibyc [June8] and 2019 by Junethat figure 2019 that had figure risen hadto 172 risen [9]. to With 172 [ 9low]. With energy low use, energy low use, 56 externallow external noise and noise no exhaust and no exhaustpipe emissions, pipe emissions, a rapid increase a rapid increasein electric in buses electric in commercial buses in commercial traffic 57 is traa probableffic is a probable scenario scenario once the once technology the technology becomes becomes more moremature. mature. Since Since 2016, 2016, the theSwedish Swedish 58 GovernmentGovernment has has paid paid a a20% 20% subsidy forfor thethe purchase purchase of of electric electric buses buses in order in order to support to support low emission low 59 emissionvehicles vehicles [10]. In [10]. a lifecycle In a lifecycle cost assessment cost assessment Lajunen Lajunen et al. [ 11et ]al. found [11] found that electric that electric buses stillbuses have still not 60 havereached not reached competitive competitive levels levels regarding regarding cost but cost the but authors the authors also concluded also concluded that electric that electric buses buses have the 61 havepotential the potential to significantly to significantly reduce reduce carbon carbon dioxide dioxide emissions, emissions, by up toby 75%.up to 75%. 62 A technologicalA technological change change from from biogas biogas ICE ICE buses buses to electric to electric buses buses implies implies more more than than just just a fuel a fuel 63 change.change. Figure Figure 1 illustrates1 illustrates the the more more complex complex energy energy flow flow that that must must be be considered considered if if biogas biogas is used in 64 in aa CHP CHP system system and and transport transport is is carried carried out out with with electric electric buses.

65 Figure 1. Schematic illustration of alternative energy flows. At the top, the energy flow for a biogas 66 Figurebus 1. with Schematic internal illustration combustion of engine altern (ICE).ative Atenergy the bottom, flows. At the the more top, complex the energy energy flow flow for whena biogas biogas 67 busis with used internal in a combined combustion heat engi andne power (ICE). (CHP) At the system bottom, and the transportmore complex is carried energy out flow with when electric biogas buses. 68 is used in a combined heat and power (CHP) system and transport is carried out with electric buses. Alternative uses of the produced biogas must be included in the analysis to accurately evaluate 69 theAlternative energy system uses performance.of the produced In the biogas worst must case, be there included are no in alternative the analysis uses to for accurately the produced evaluate biogas 70 theand energy hence system the biogas performance. will be burntIn the in worst a gas case, flare there without are creating no alternative a useful uses energy for the service. produced In 2016, 71 biogasbiogas and burnt hence in the gas biogas flares will was be the burnt third in most a gas common flare without use for creati biogas,ng a correspondinguseful energy service. to 9% ofIn the 72 2016,biogas biogas produced; burnt in this gas was flares partly was due the tothird a lack mo ofst biogascommon customers use for [biogas,2]. Hedegaard corresponding et al. [12 to] examined 9% of 73 thebiomass biogas produced; use and consequences this was partly of constrained due to a lack biomass of biogas availability, customers they [2 concluded]. Hedegaard that et technology al. [12] 74 examinedpathways biomass involving use heatand andconsequences power production of constrained and/or biomass biogas, natural availability, gas or they electricity concluded for transport that 75 technologyare advantageous. pathways involving heat and power production and/or biogas, natural gas or electricity 76 for transport are advantageous.

Energies 2019, 12, 4159 3 of 13

In a Swedish context, a probable alternative use for biogas is as fuel in a CHP plant in a DHS. During 2013, the most common energy carrier for heating and hot water was district heating, at 47 TWh [3]. DH is available in most Swedish cities but the fuel mix for each system varies considerably. The introduction of biogas in a DHS would therefore have a different impact depending on which fuels are being replaced. Some DHSs only deliver industrial waste heat to their heat customers and in those cases the introduction of biogas would not create a beneficial energy service. However, in other cases, replacing a fuel mix containing fossil fuels, it would be beneficial. The technical solution to enable biogas use in a DHS can be implemented in several different ways. The biogas could be burnt in a heat only boiler, in a boiler connected to a steam turbine or in a CCGT (Combined Cycle Gas Turbine). The most energy efficient solution is to use a CCGT, which has the highest electrical efficiency of the three alternatives. However, CCGT is the most expensive hardware and is therefore not always a feasible option. Börjesson et al. [13] examined the cost of oil use reduction in the Swedish transport sector and concluded that biomass gasification pathways show high potential for low cost oil reduction. Use of CCGTs is central in those pathways if electric vehicles are prioritized before vehicles with ICEs. It is crucial to consider alternative fuels, conversion efficiencies and the alternative production for the produced services. There is also a practical need to set boundaries for the analytical study. In this study the three categories: bus transport, electricity production and heat production are considered to be final products. The aim of this study is to examine and compare two alternatives for biogas use in Linköping municipality. The first alternative that is examined is in operation today. In this system, the locally produced biogas is used as fuel in the ICE biogas buses. In parallel, the CHP system delivers electricity and heat to households in the region. The second alternative that is studied is a system with electric buses and a CHP system that uses biogas in a CCGT to deliver electricity and heat to the regional power grid and DHS. In this study CCGT is assumed to be fuelled with upgraded biogas (97% methane) that is, the same fuel quality as the buses use today. A whole year analysis has been conducted to summarize the accumulated system impact.

2. Methodology This study used the optimization software MODEST (Model for Optimization of Dynamic Energy Systems with Time-dependent components and boundary conditions) (version—August 2015, Linköping, Sweden). The combined system and the separate transport system and DHS system were represented in the model by four different kinds of nodes: fuel nodes, conversion nodes (boilers, turbines, buses and flue gas condensers), demand nodes (heat load and transport demand) and waste nodes (electricity production and heat losses). The power plants were described in terms of efficiencies, maximum capacity and fuel type. The heat load profile was taken from measurements from 2015. The MODEST model is described in detail by Henning [14]. In recent years, several authors have used MODEST to analyse DHS, for example Lidberg et al. [15], Gebremedhin [16] and Blomqvist et al. [17]. MODEST is a top-down tool that can be used to represent the largest flows in an energy system. Figure2 illustrates the workflow used in this study. Two models based on power plant data, bus transport data and measurements were created. The models were optimized to cover the yearly heat load and transport demand at the lowest cost. The optimization result shows which power plants are used for each time step in the model, which in turn gives used fuels and the amount of electricity produced. After creating and optimizing the parallel models, their results were compared and evaluated regarding fuel consumption and GHG emissions. Energies 2019, 12, 4159 4 of 13 Energies 2019, 12, x FOR PEER REVIEW 4 of 14

116 Figure 2. Schematic illustration of the workflow for the study. Horizontal lines show borders for the 117 Figure 2. Schematic illustration of the workflow for the study. Horizontal lines show borders for the MODEST model. 118 MODEST model. Methodological Difficulties and Boundaries 119 Two models based on power plant data, bus transport data and measurements were created. 120This The method models is a were top-down optimized method, to cover which the means yearly that heat the load study and starts transport at the demand top, in this at the case lowest a cost. regional121 energyThe optimization system and result then movesshows downwhich inpower finer plants and finer are detail. used for At each a certain time degree step in of the system model, which detail,122 onein needs turn gives to stop used but fuels there and is no the distinct amount way of to electricity choose this produced. level. In the time domain, this study starts123 with year,After then creating moves toand months, optimizing weeks the and parallel days andmodels, then their stops. results In the were physical compared domain, and this evaluated study124 startsregarding with the fuel CHP-DHS consumption and the and regional GHG transportemissions. system in a municipality, then moves to the supply side, power plants, power plant components, fuels, buses, bus engines and then stops. It is the125 opinion Methodological of the authors Difficulties that more detail and Boundaries in this model would not be beneficial for the calculations but would rather add uncertainties. There are many details in the physical domain that are omitted, 126 This method is a top-down method, which means that the study starts at the top, in this case a in some cases by choice and in some cases by necessity. The top-down method used here is to be 127 regional energy system and then moves down in finer and finer detail. At a certain degree of system considered as a strategic tool to compare and evaluate possible technological pathways, not a tool to 128 detail, one needs to stop but there is no distinct way to choose this level. In the time domain, this give an exact forecast of future system emissions or energy flows. 129 study starts with year, then moves to months, weeks and days and then stops. In the physical domain, 3.130 Case Studythis study starts with the CHP-DHS and the regional transport system in a municipality, then moves 131 to the supply side, power plants, power plant components, fuels, buses, bus engines and then stops. Linköping is a municipality with about 150,000 inhabitants situated in the southeast of Sweden. 132 It is the opinion of the authors that more detail in this model would not be beneficial for the Fallde et al. [6] describe how in the late 1980s LITA, the municipally owned public transport company, 133 calculations but would rather add uncertainties. There are many details in the physical domain that began to investigate a fuel change, from diesel to biogas. After a test period between 1992 and 1994, 134 are omitted, in some cases by choice and in some cases by necessity. The top-down method used here a gradual change from diesel buses to biogas was initiated. Linköping municipality has managed 135 is to be considered as a strategic tool to compare and evaluate possible technological pathways, not biogas driven buses in the regional transport system ever since, with all the buses in the municipality 136 a tool to give an exact forecast of future system emissions or energy flows. running on biogas from 2015. There are 65 city buses and in 2017 they delivered 5.3 million km of 3 transport137 3. service. Case Study The fuel consumption was 0.6 Nm /km, corresponding to 32 GWh annually [18]. The biogas is locally produced and the annual production exceeds the consumption of the bus fleet. During138 2017, productionLinköping stoodis a municipality at 10.7 million with Nm about3 of upgraded 150,000 inhabitants biogas, corresponding situated in tothe 104 southeast GWh [19 of]. Sweden. There139 is notFallde a natural et al. gas[6] griddescribe in this how region in the where late the 1980s produced LITA, gasthe can municipally be injected. owned However, public the transport biogas140 productioncompany, site began is situated to investigate in proximity a fuel change, to the city’s from CHP diesel plants, to biogas. which After could a usetest thisperiod biogas. between 1992 141Tekniska and 1994, verken a gradual AB is the change regional from power diesel company buses to and biogas has awas yearly initiated. production Linköping of 1500 municipality GWh has heat142 andmanaged 260 GWh biogas electricity driven at two buses production in the regional sites. In transp the urbanort system area, DHever is since, the dominant with all the way buses of in the supplying143 municipality heat to multi-family running buildings,on biogas from small 2015. houses There and commercialare 65 city buses buildings. and in The 2017 base they load delivered in 5.3 3 the144 DHS ismillion covered km by of wastetransport incineration service. atThe CHP fuel plants. consumption With increasing was 0.6 Nm load,/km, other corresponding CHP plants are to 32 GWh starting145 toannually burn other [18]. fuels The such biogas as wood,is locally rubber, produced coal and and oil. the Heat annual only production boilers are exceeds used to coverthe consumption peak of 3 loads.146 Thethe fuel bus merit fleet. order During is waste, 2017, produc wood, coaltion /stoodrubber, at oil. 10.7 million Nm of upgraded biogas, corresponding 147 to 104 GWh [19]. There is not a natural gas grid in this region where the produced gas can be injected.

Energies 2019, 12, 4159 5 of 13

The optimization objective was to minimize the annual system cost while satisfying the given heat load demand and the transport demand. Fuel costs were ﬁxed during the examined year and were mainly set to force the model optimization algorithm to choose plants in the correct merit order. Details of the model are shown in Tables1–5 and Figures3 and4.

Table 1. Description of studied cases.

Case Description Model of the current system. The local transport service provided by 65 biogas driven Case 1 buses. CHP and DHS provide power and heat to the region. Model of an integrated system. Biogas is used in CCGT, which is integrated in the CHP Case 2 and DHS. The local transport service is provided by electric buses.

Table 2. Modelled production units in Linköping’s CHP system. Source: TvAB.

Flue Gas Turbine Eﬃciency Maximum Condenser (Electricity Power Plant Output Fuel Eﬃciency (% of Production/Input (MW) Boiler Power) Heat) KV 50 72 Waste incineration 15–40 0.10–0.15 KV 61 66 Waste incineration 0–22 0.17–0.27 KV 81 83 Waste incineration 0–22 0.17–0.27 CCGT 150 (Oil or Gas) + Waste FGC not available 0.33 KV 1—wood 60 Wood 20–40 0.18–0.26 KV 1—coal 60 Coal/rubber FGC not available 0.18–0.26 KV 1—oil 120 Oil FGC not available 0.18–0.26 Heat only boilers (several distributed in 240 Oil FGC not available No turbine DH network)

Table 3. Bus data in the model.

Transport Bus Type Fuel Heating and AC Engine Eﬃciency Intensity 605 km bus Upgraded biogas Included in annual Biogas ICE 0.6 Nm3 biogas/km transport/hour (97% methane) fuel use. 605 km bus Electricity from Time dependent Electric 1.4 kWh/km transport/hour regional network (see Section 4.1)

Table 4. Time division in the model.

Period Time Resolution January–April (ﬁrst 10 days in monthly duration diagram) Average heat load for 48 h period January–April (remaining days in monthly duration diagram) Average heat load for 96 h period May-August Average heat load for one week September–December (ﬁrst 10 days in monthly duration diagram) Average heat load for 48 h period September–December (remaining days in monthly Average heat load for 96 h period duration diagram)

Table 5. Fuel prices in Model.

Fuel Price in Model Waste 0 SEK/MWh Wood 70 SEK/MWh Coal/Rubber 180 SEK/MWh Oil CHP 320 SEK/MWh Oil—Heat only boiler 600 SEK/MWh Energies 12 Energies 20192019,, 12,, x 4159 FOR PEER REVIEW 6 6of of 14 13 Energies 2019, 12, x FOR PEER REVIEW 6 of 14

162 162 163 FigureFigure 3. 3. IllustrationIllustration of of nodes nodes and and flows ﬂows in in the the Case Case 1 1 model. model. 163 Figure 3. Illustration of nodes and flows in the Case 1 model.

164 164 165 FigureFigure 4. 4. IllustrationIllustration of of nodes nodes and and flows ﬂows in in the the Case Case 2 2 model. model. 165 Figure 4. Illustration of nodes and flows in the Case 2 model.

166 166

Energies 2019, 12, 4159 7 of 13

In the model the bus traﬃc is evenly distributed during the year. The smallest time step in the model is 2 days, which means that the transport demand node needs fuel for 48 h 605 km/h = × 29,040 km to cover that period. Time periods used in the model are presented in Table4. The heat load was modelled with the monthly duration diagram in Figure5. The shortest time step used was 48 h. All fast transients (less than 48 h) are handled with the power plant heat accumulator, so the 48 h time step is reasonable. It is also important to note that too short a time step, for example, an hour, will create false power peaks in the model that do not exist in the real CHP plant; these power

peaksEnergies are2019 handled, 12, x FOR by PEER the REVIEW accumulator. 8 of 14

179

180 Figure 5. Heat load in the model, presented with monthly duration diagram, i.e., days sorted by heat 181 load in each month. Days Days with with higher heat lo loadsads are modelled in greater detail, see Table4 4..

182 3.1. Electric Bus Engine Efficiency Efficiency 183 The electric busbus engineengine eefficiencyfficiency for for the the future future system system in in Case Case 2is 2 unknown.is unknown. However, However, there there are 184 areelectric electric buses buses in operation in operation today today and a and calculation a calculat ofion electricity of electricity use based use on based existing on existing buses is possible.buses is 185 Thepossible. electric The bus electric fleet inbus Case fleet 2 in is Case assumed 2 is assumed to be of equalto be sizeof equal to Case size 1,to that Case is 1, 65 that buses: is 65 53 buses: units 53 of 186 18units m busesof 18 m and buses 12 units and of12 12units m buses.of 12 m The buses. average The energyaverage use energy in the use test in ranged the test between ranged 0.86between and 187 1.020.86 kWhand 1.02/km kWh/km in [7] for in 12 [7] m buses.for 12 m In buses. this case In this study, case energy study, use energy is assumed use is assumed to be 1 kWh to be/km 1 kWh/km for 12 m 188 forbuses 12 andm buses 1.5 kWh and/ km1.5 kWh/km for 18 m buses.for 18 m With buses. a linear With calculation, a linear calculation, energy use energy for the use entire for the bus entire fleet 189 busbecomes fleet becomes 1.4 kWh/ km.1.4 kWh/ Thiskm. energy This use energy is consistent use is consistent with other with sources, other seesources, for example see for example [20–22]. [20– 190 22]. 3.2. Bus Heating and Air-Conditioning in the Model 191 3.2. BusThe Heating introduction and Air-conditioning of an alternative in the Model drive train, without ICE, also affects the heating and 192 air-conditioningThe introduction for the of bus. an Thealternative required drive energy train, for heatingwithout andICE, air-conditioning also affects the must heating beprovided and air- 193 conditioningby the onboard for energy the bus. system, The required instead ofenergy utilizing for excessheating heat and from air-conditioning the combustion. must During be provided cold days by 194 theheating onboard can be energy a considerable system, instead part of of the utilizing total energy excess use heat for anfrom electric the combustion. bus. Borén etDuring al. [7] measuredcold days 195 heatingthe energy can use be a for considerable heating as 0.67 part kWh of the/km total in January.energy use That for measurement an electric bus. was Borén done et in al. a region[7] measured with a 196 thesimilar energy climate use for to Linköping heating as municipality.0.67 kWh/km in January. That measurement was done in a region with 197 a similarIn the climate MODEST to Linköping model used municipality. in this case study, heating and AC were assumed to have an energy 198 demandIn the corresponding MODEST model to values used in in this Figure case6. study, Temperatures heating areand average AC were values assumed in the to regionhave an derived energy 199 demandfrom SMHI corresponding [23]. Heating to values was assumed in Figure to 6. be Temper generatedatures by are an average onboard values heat-pump in the region with COP derived 3.0. 200 Thisfrom assumptionSMHI [23]. Heating is in accordance was assumed with Göhlichto be generated et al. [24 by]. an onboard heat-pump with COP 3.0. This 201 assumption is in accordance with Göhlich et al. [24].

202

203 Figure 6. Temperature and assumed energy use for heating and AC for electric buses.

Energies 2019, 12, x FOR PEER REVIEW 8 of 14

179 180 Figure 5. Heat load in the model, presented with monthly duration diagram, i.e., days sorted by heat 181 load in each month. Days with higher heat loads are modelled in greater detail, see Table 4.

182 3.1. Electric Bus Engine Efficiency 183 The electric bus engine efficiency for the future system in Case 2 is unknown. However, there 184 are electric buses in operation today and a calculation of electricity use based on existing buses is 185 possible. The electric bus fleet in Case 2 is assumed to be of equal size to Case 1, that is 65 buses: 53 186 units of 18 m buses and 12 units of 12 m buses. The average energy use in the test ranged between 187 0.86 and 1.02 kWh/km in [7] for 12 m buses. In this case study, energy use is assumed to be 1 kWh/km 188 for 12 m buses and 1.5 kWh/km for 18 m buses. With a linear calculation, energy use for the entire 189 bus fleet becomes 1.4 kWh/km. This energy use is consistent with other sources, see for example [20– 190 22].

191 3.2. Bus Heating and Air-conditioning in the Model 192 The introduction of an alternative drive train, without ICE, also affects the heating and air- 193 conditioning for the bus. The required energy for heating and air-conditioning must be provided by 194 the onboard energy system, instead of utilizing excess heat from the combustion. During cold days 195 heating can be a considerable part of the total energy use for an electric bus. Borén et al. [7] measured 196 the energy use for heating as 0.67 kWh/km in January. That measurement was done in a region with 197 a similar climate to Linköping municipality. 198 In the MODEST model used in this case study, heating and AC were assumed to have an energy 199 demand corresponding to values in Figure 6. Temperatures are average values in the region derived 200 Energiesfrom SMHI2019, 12 [23]., 4159 Heating was assumed to be generated by an onboard heat-pump with COP 3.0.8 This of 13 201 assumption is in accordance with Göhlich et al. [24].

202 Figure 6. Temperature and assumed energy use for heating and AC for electric buses. 203 Figure 6. Temperature and assumed energy use for heating and AC for electric buses. 3.3. Emissions in the Model

The emission values in Table6 are derived from The Environmental Fact Book [ 25]. Emissions for electricity use are calculated as average CO2-eq in the Nordic region. If excess electricity is produced by the examined system, then 131.2 kg CO2-eq/MWh is subtracted from the system’s local emissions.

Table 6. Emissions of greenhouse gas (GHG) CO2-eq for each fuel, The Environmental Fact Book [25].

Fuel Emissions kg CO2-eq/MWh Wood 14 Oil 288 Coal/rubber 360 Waste 136 Biogas 39.6 Electricity use 131.2 * * Emissions for electricity use in the Nordic region, source: IVL—Swedish Environmental Research Institute [26].

3.4. CCGT Operation Linköping municipality’s CHP-DHS has one CCGT unit. It is a 150 MW unit running on 75 MW of gas or oil and 75 MW of waste incineration, where the steam part of the turbine uses waste as fuel. When comparing annual biogas production and the power of this CCGT unit, it is clear that this unit is oversized. However, the amount of biogas used by the buses today corresponds to running this unit at full power approximately 8 h every week and that is a feasible operation time. Using all locally produced biogas in the CCGT turbine would correspond to 26 h of full power every week. Detailed CCGT operation is not included in this study but there are several options available to handle the diﬀerence in biogas fuel access and CCGT power. Turbine retroﬁt is one option and dual fuel use another (burning both oil and biogas).

4. Model Results This section presents the model results. The full model results for each case consist of a 38 96 × matrix with all energy flows between different nodes in the model for each time step. The results presented here are selected results to illustrate the most important differences between the studied cases. Energies 2019, 12, 4159 9 of 13

4.1. Efficiencies and Fuel Consumption A comparison of biogas use in the two studied systems is not easy. The change from Case 1 to Case 2 is not just a bus engine change; it is also a change in the delivered service from the biogas use. In Case 2 the biogas use is involved in a more complex energy flow, as illustrated in Figures7 and8. In Case 2 the biogas use affects waste incineration, DH heat production, electricity production and fuel supply to cover the bus transport demand. The services produced by the biogas use are a bus transport service, a heating service and an electricity production service. The biogas use in Case 2 is Energies 2019, 12, x FOR PEER REVIEW 10 of 14 Energiesalso directly2019, 12, x integrated FOR PEER REVIEW with a waste incineration fuel-energy flow. 10 of 14

235 235 236 FigureFigure 7. Fuel-energy7. Fuel-energy pathway pathway for for the the regionally regionally produced produced biogas biogas when when the combined the combined cycle cycle gas turbine gas 236 Figure 7. Fuel-energy pathway for the regionally produced biogas when the combined cycle gas 237 (CCGT) is used. turbine (CCGT) is used. 237 turbine (CCGT) is used.

238 238 239 Figure 8. Fuels and services when biogas is used in the CCGT at the CHP plant. 239 Figure 8. Fuels and services when biogas is used in the CCGT at the CHP plant. 240 InIn CaseCase 11 thethe fuel-energyfuel-energy flowflow is a simplesimple flow,flow, wherewhere 3232 GWhGWh biogasbiogas fuelfuel resultsresults inin 5.35.3 millionmillion 240 In Case 1 the fuel-energy flow is a simple flow, where 32 GWh biogas fuel results in 5.3 million 241 km busbus transporttransport service. The more complex energy system service produced in Case 2 is shown in 241 km bus transport service. The more complex energy system service produced in Case 2 is shown in 242 Figure8 8.,., with with two two fuels fuels giving giving three three services. services. One One way way to to assess assess the the value value of theof the system system integration integration in 242 Figure 8., with two fuels giving three services. One way to assess the value of the system integration 243 Casein Case 2, compared2, compared to Caseto Case 1, is1, tois examineto examine the the added added system system service. service. Moving Moving from from Case Case 1 to1 to Case Case 2, 243 in Case 2, compared to Case 1, is to examine the added system service. Moving from Case 1 to Case 244 382, GWh38 GWh waste waste incineration incineration is added is added on the on fuel the side fuel and side at theand same at the time same 43 GWhtime of43 “heatGWh delivered of “heat 244 2, 38 GWh waste incineration is added on the fuel side and at the same time 43 GWh of “heat 245 todelivered DHS” is to added DHS” and is added 8 GWh and “excess 8 GWh electricity “excess production”electricity production” is added. Hence, is added. the systemHence, integrationthe system 245 delivered to DHS” is added and 8 GWh “excess electricity production” is added. Hence, the system 246 isintegration beneficial is because beneficial more because services more are services produced are with produced less fuel. with This less becomes fuel. This easier becomes to see easier when to thesee 246 integration is beneficial because more services are produced with less fuel. This becomes easier to see 247 entirewhen systemthe entire is studied.system is studied. 247 when the entire system is studied. 248 4.2. Entire System Results 248 4.2. Entire System Results 249 Table7 7 shows shows the the model model results results for for the the entire entire systems, systems, Case Case 1 1 and and Case Case 2. 2. OnlyOnly CaseCase 22 usesuses 249 Table 7 shows the model results for the entire systems, Case 1 and Case 2. Only Case 2 uses 250 electric buses and the CCGT, apart from that the greatestgreatest didifferencefference betweenbetween thethe cases is in electricity 250 electric buses and the CCGT, apart from that the greatest difference between the cases is in electricity 251 production,production, where therethere isis aa 4%4% increaseincrease comparingcomparing CaseCase 22 withwith CaseCase 1.1. The greatergreater electricityelectricity 251 production, where there is a 4% increase comparing Case 2 with Case 1. The greater electricity 252 productionproduction isis aa consequence consequence of of the the use use of of CCGT, CCGT, which which has has the highestthe highest electrical electrical efficiency efficiency in the in CHP the 252 production is a consequence of the use of CCGT, which has the highest electrical efficiency in the 253 systemCHP system (see Table (see2 ).Table Case 2). 2 also Case has 2 lower also has fuel uselower for fuel waste, use wood, for waste, coal /rubber wood, and coal/rubber oil. Emissions and areoil. 253 CHP system (see Table 2). Case 2 also has lower fuel use for waste, wood, coal/rubber and oil. 254 Emissions are 1% lower for Case 2, if excess electricity production is considered. In relation to the 254 Emissions are 1% lower for Case 2, if excess electricity production is considered. In relation to the 255 annual load for the integrated system, the differences between Case 1 and Case 2 then become small. 255 annual load for the integrated system, the differences between Case 1 and Case 2 then become small. 256 However, in relation to the biogas flow, the differences are substantial. 256 However, in relation to the biogas flow, the differences are substantial. 257 257

Energies 2019, 12, 4159 10 of 13 Energies 2019, 12, x FOR PEER REVIEW 11 of 14

258 1% lower forTable Case 7. Model 2, if excess results, electricity annual flows production for all fuels is considered. and some selected In relation production to the annualunits. load for the integrated system, the differences between Case 1 and Case 2 then become small. However, in relation Case 1 Case 2 to the biogas flow, the differences are substantial. Waste incineration (GWh) 1560 1553 Table 7. Wood (GWh) Model results, annual flows for all fuels 144.7 and some selected production 142.6 units. Coal/Rubber (GWh) 38 Case 1 37.1 Case 2 Oil (GWh) Waste incineration (GWh) 6.9 1560 6.6 1553 Wood (GWh) 144.7 142.6 Biogas (GWh) Coal/Rubber (GWh) 31.7* 38 32.2* 37.1 Total heat productionOil (GWh) (GWh) 1459 6.9 1459 6.6 Biogas (GWh) 31.7 * 32.2 * Total electricityTotal production heat production (GWh) (GWh) 256.6 1459 264.6 1459 CCGT electricityTotal production electricity (GWh) production (GWh) 0 256.6 21.5 264.6 CCGT electricity production (GWh) 0 21.5 Electric bus electricityElectric use bus (GWh) electricity use (GWh) 0 0 8.0 8.0 Local CO2-eq emissionsLocal CO 2(million-eq emissions kg) (million kg) 232.5 232.5 231.1 231.1 CO -eq emissions with subtraction for excess 2 232.5 230.1 CO2-eq emissionselectricity with subtraction production for (million excess kg)electricity 232.5 230.1 production (million kg) * A small difference in biogas use is caused by the numerical method. 259 *a small difference in biogas use is caused by the numerical method. Figure9. shows the di fferences in fuel used and electricity produced between Case 1 and Case 2. 260 TheFigure use of 9. wood shows as the fuel differences is reduced in by fuel 2 GWh, used coaland/ rubberelectricity is reduced produced by between 0.9 GWh, Case oil is1 reducedand Case by 261 2. 0.3The GWh use of and wood waste as incineration fuel is reduced is reduced by 2 GWh, by 6.4 coal/rubber GWh. The is model reduced also by shows 0.9 GWh, an 8 GWhoil is increasereduced in 262 byelectricity 0.3 GWh and production. waste incineration These system is reduced changes by can 6.4 alsoGWh. be The calculated model also as CO shows2-eq emissionan 8 GWh reductions, increase 263 inas electricity shown in production. Figure 10. These system changes can also be calculated as CO2-eq emission reductions, 264 as shown in Figure 10.

265

266 FigureFigure 9. 9.DifferencesDiﬀerences in infuel fuel used used and and electricity electricity production production between between Case Case 1 and 1 and Case Case 2. 2.

Increased electricity production from the CHP-DHS will replace other electricity production in the electricity market. Here, the average emission value for electricity production in the Nordic region is used to calculate the eﬀect of that replacement. With 8 GWh increased electricity production, the CO2 reduction is greatest for this system change. Lowered waste incineration has the second largest reduction in CO2 emissions and coal/rubber combustion the third largest. The uncertainties in the top-down method should be emphasized regarding exact emission values; the interesting result is not the exact emission values but rather the overall tendency of a general fuel saving and an eﬃciency gain.

267

268 Figure 10. Differences in CO2-eq emissions (1000 kg) between Case 1 and Case 2. Emission values for 269 each fuel from Table 5.

Energies 2019, 12, x FOR PEER REVIEW 11 of 14

258 Table 7. Model results, annual flows for all fuels and some selected production units.

Case 1 Case 2 Waste incineration (GWh) 1560 1553 Wood (GWh) 144.7 142.6 Coal/Rubber (GWh) 38 37.1 Oil (GWh) 6.9 6.6 Biogas (GWh) 31.7* 32.2* Total heat production (GWh) 1459 1459 Total electricity production (GWh) 256.6 264.6 CCGT electricity production (GWh) 0 21.5 Electric bus electricity use (GWh) 0 8.0 Local CO2-eq emissions (million kg) 232.5 231.1 CO2-eq emissions with subtraction for excess electricity 232.5 230.1 production (million kg) 259 *a small difference in biogas use is caused by the numerical method.

260 Figure 9. shows the differences in fuel used and electricity produced between Case 1 and Case 261 2. The use of wood as fuel is reduced by 2 GWh, coal/rubber is reduced by 0.9 GWh, oil is reduced 262 by 0.3 GWh and waste incineration is reduced by 6.4 GWh. The model also shows an 8 GWh increase 263 in electricity production. These system changes can also be calculated as CO2-eq emission reductions, 264 as shown in Figure 10.

265 Energies 2019, 12, 4159 11 of 13 266 Figure 9. Differences in fuel used and electricity production between Case 1 and Case 2.

Energies 2019, 12, x FOR PEER REVIEW 12 of 14

270 Increased electricity production from the CHP-DHS will replace other electricity production in 271 the electricity market. Here, the average emission value for electricity production in the Nordic region 272 is used to calculate the effect of that replacement. With 8 GWh increased electricity production, the 273 CO2 reduction is greatest for this system change. Lowered waste incineration has the second largest 274 reduction in CO2 emissions and coal/rubber combustion the third largest. The uncertainties in the 267 275 top-down method should be emphasized regarding exact emission values; the interesting result is Figure 10. Differences in CO2-eq emissions (1000 kg) between Case 1 and Case 2. Emission values for 276268 not the exactFigure emission 10. Differences values in but CO 2rather-eq emissions the overall (1000 te kg)ndency between of a Case general 1 and fuel Case saving 2. Emission and an values efficiency for each fuel from Table5. 277269 gain. each fuel from Table 5. 4.3. Sensitivity Study Electric Bus Efficiency 278 4.3. Sensitivity Study Electric Bus Efficiency The use of electric buses for inner city transport is a new technology and this implies greater 279 uncertaintiesThe use of aboutelectric effi busesciencies for than inner for city more transport established is a alternatives.new technology To address and this this implies issue, thegreater model 280 uncertaintieswith electric about buses efficiencies was also calculated than for with more lower established efficiencies alternatives. for the electric To address bus drive this train. issue, Figure the 11 281 modelshows with the electric excess electricitybuses was production also calculated from with Case lower 2—the efficiencies integrated for system the electric with CCGT. bus drive train. 282 Figure 11 shows the excess electricity production from Case 2—the integrated system with CCGT.

283 284 FigureFigure 11. 11.ExcessExcess electricity electricity from from Case Case 2 with 2 with different diﬀerent values values for for electric electric bus bus efficiency. eﬃciency.

285 TheThe trend trend is isobvious obvious and and expected, expected, lower lower efficienc eﬃciencyy for for the the electric electric bus bus will will give give lower lower excess excess 286 productionproduction of ofelectricity electricity from from the the CHP-DHS. CHP-DHS. However, However, there there is isexcess excess electricity electricity production production for for the the 287 integratedintegrated system, system, even even with with an an electric electric bus bus effic eﬃiencyciency of of1.82 1.82 kWh/km. kWh/km. The The value value of of1.82 1.82 kWh/km kWh/km 288 correspondscorresponds to toa 30% a 30% increase increase in inelectricity electricity use use compared compared with with the the calculated calculated value value used used in the in the model. model.

289 5. 5.Discussion Discussion 290 TheThe biogas biogas fuel-energy fuel-energy flow flow is isa small a small part part of ofthe the studied studied regional regional energy energy system system in inLinköping Linköping 291 municipality.municipality. Biogas Biogas energy energy use usein today’s in today’s ICE ICEbuse busess is only is only2%, compared 2%, compared to the to energy the energy use in use the in 292 CHP-DHSthe CHP-DHS in the municipality. in the municipality. This difference This di infference size between in size the between bus transport the bus system transport and systemthe CHP- and 293 DHSthe can CHP-DHS be a problem can be when a problem evaluating when a system evaluating integration a system between integration these between two systems. these Comparing two systems. 294 theComparing studied cases, the studiedCase 1 and cases, Case Case 2 and 1 and looking Case at 2 andthe whole looking system, at the wholethere is system, only a 1% there reduction is only ain 1% reduction in CO emissions. However, from another perspective, looking at the fuel-energy pathway 295 CO2 emissions. However,2 from another perspective, looking at the fuel-energy pathway for the 296 biogas,for the the biogas, improvement the improvement is substantial. is substantial. The change The changefrom Case from 1 Caseto Case 1 to 2 Case eliminates 2 eliminates 65 sources 65 sources of 297 heatof heatloss (the loss (thebuses) buses) and andredirects redirects fuel-energy fuel-energy flows flows in more in more efficient efficient pathways. pathways. The The study study indicates indicates that improved biogas use could lower CO -eq emissions by 2.4 million kg annually by using a better 298 that improved biogas use could lower CO2-eq2 emissions by 2.4 million kg annually by using a better 299 fuel-energyfuel-energy pathway. pathway. 300 ThereThere is isan an asymmetrical asymmetrical information information situation regardingregarding the the two two compared compared cases cases in thisin this case case study. 301 study.The firstThe casefirst is case well is known well andknown the inputand the data input forthe data model for isthe derived model fromis derived reliable from measurements. reliable 302 measurements. Input data for the second case model has more uncertainties. Both electricity use for 303 the electric bus engine and the electricity use for AC and heating in the buses are based on theoretical 304 assumptions guided by literature and experiments. There is also uncertainty about CCGT efficiency 305 for the suggested operation. However, these three uncertainties all affect the same part of the model 306 result, the excess electricity production. 307 With an overall evaluation of the uncertainties regarding the differences between Case 1 and 308 Case 2, the following can be stated. Increased electricity production is likely. Elimination of heat

Energies 2019, 12, 4159 12 of 13

Input data for the second case model has more uncertainties. Both electricity use for the electric bus engine and the electricity use for AC and heating in the buses are based on theoretical assumptions guided by literature and experiments. There is also uncertainty about CCGT efficiency for the suggested operation. However, these three uncertainties all affect the same part of the model result, the excess electricity production. With an overall evaluation of the uncertainties regarding the differences between Case 1 and Case 2, the following can be stated. Increased electricity production is likely. Elimination of heat losses is certain. A net reduction in fuel use is certain. However, there is less certainty about which specific fuel use (waste, wood, oil, coal/rubber) will be reduced. This case study can to some extent be generalized to other CHP-DHSs and regional bus transport systems. A changed fuel-energy pathway for biogas can be beneficial for CHP-DHS with CCGT and the key issue is whether the change results in higher efficiencies in the fuel-energy pathway. The higher efficiency could be both a reduction in heat losses and lower losses of energy quality, that is, higher electrical efficiency. However, each specific energy system must be analysed separately in order to evaluate a change in the fuel-energy pathway for biogas.

6. Conclusions Moving from Case 1 to Case 2, 38 GWh waste incineration is added to the fuel side, along with 43 GWh of “heat delivered to DHS” and 8 GWh “excess electricity production.” Hence, the system integration is beneﬁcial because more services are produced with less fuel. The study indicates that improved biogas use could lower CO2-eq emissions by 2.4 million kg annually by using a better fuel-energy pathway. Implementation of new technology is complicated and certainly not only a matter of theoretical calculations; several practical issues and economic conditions must also be addressed. However, this theoretical case study does not reject a development where biogas use is altered from combustion in bus engines to combustion in combined cycle gas turbines. On the contrary, the regional energy system perspective is in favour of such a development, because of the elimination of heat losses and the improvement to the biogas fuel-energy pathway.

Author Contributions: All the parts of the manuscript were discussed among the two authors. T.R. was the main author and wrote all the parts. Funding: This research received no external funding. Conflicts of Interest: One of the authors, T.R., has in his private stock portfolio stocks in two companies which manufacture buses. One company manufactures both electric and biogas driven buses, the other company only electric buses. Due to the stock portfolio, one might consider that a conflict of interest could be present in this research work. The author T.R. could have an economic benefit in the case of an increasing electric bus market. However, no data, measurements or other material in this article have been derive from any of these companies. The conclusion is therefore that the content in this article has not been affected by the private stock ownership.

References

1. Life Cycle Assessment of Biofuels in Sweden; Department of Technology and Society, Environmental and Energy Systems Studies, Lund University: Lund, Sweden, 2010. 2. Energimyndigheten. Produktion och Användning av Biogas och Rötrester år 2016; Energimyndigheten: Eskilstuna, Sweden, 2016. 3. Energimyndigheten. Energiläget; Energimyndigheten: Eskilstuna, Sweden, 2015; Available online: www. energimyndigheten.se (accessed on 1 July 2019). 4. Börjesson, M.; Ahlgren, E.O. Cost-eﬀective biogas utilisation—A modelling assessment of gas infrastructural options in a regional energy system. Energy 2012, 48, 212–226. [CrossRef] 5. Energigas Sverige. Available online: http://www.energigas.se/fakta-om-gas/fordonsgas-och-gasbilar/ statistik-om-fordonsgas/ (accessed on 8 December 2018). 6. Fallde, M.; Eklund, M. Towards a sustainable socio-technical system of biogas for transport: The case of the city of Linköping in Sweden. J. Clean. Prod. 2015, 98, 17–28. [CrossRef] Energies 2019, 12, 4159 13 of 13

7. Borén, S.; Nurhadi, L.; Ny, H. Preference of Electric Buses in Public Transport: Conclusions from Real Life Testing in Eight Swedish Municipalities. Int. J. Environ. Ecol. Eng. 2016, 10, 255. 8. Powercircle.org. Available online: http://powercircle.org/nyhet/hur-manga-elbussar-finns-sverige/ (accessed on 6 December 2018). 9. Antal Elbussar i Sverige, 2019. Available online: https://www.elbilsstatistik.se/elbilsstatistik (accessed on 13 October 2018). 10. Swedish Government. Förordning (2016:836) om Elbusspremie; Swedish Government: Stockholm, Sweden, 2016. 11. Lajunen, A.; Lipman, T. Lifecycle cost assessment and carbon dioxide emissions of diesel, natural gas, hybrid electric, fuel cell hybrid and electric transit buses. Energy 2016, 106, 329–342. [CrossRef] 12. Hedegaard, K.; Thyø, K.A.; Wenzel, H. Life Cycle Assessment of an Advanced Bioethanol Technology in the Perspective of Constrained Biomass Availability. Environ. Sci. Technol. 2008, 42, 7992–7999. [CrossRef] [PubMed] 13. Börjesson, M.; Ahlgren, E.O. Modelling transport fuel pathways: Achieving cost-effective oil use reduction in passenger cars in Sweden. Technol. Forecast. Soc. Chang. 2012, 79, 801–818. [CrossRef] 14. Henning, D. Optimisation of Local and National Energy Systems: Development and Use of the MODEST Model. Ph.D. Thesis, Linköping University, Linköping, Sweden, 1999. 15. Lidberg, T.; Olofsson, T.; Trygg, L. System impact of energy efficient building refurbishment within a district heated region. Energy 2016, 106, 45–53. [CrossRef] 16. Gebremedhin, A. Introducing District Heating in a Norwegian town—Potential for reduced Local and Global Emissions. Appl. Energy 2012, 95, 300–304. [CrossRef] 17. Blomqvist, S.; Amiri, S.; Rohdin, P.; Ödlund, L. Analyzing the Performance and Control of a Hydronic Pavement System in a District Heating Network. Energies 2019, 12, 2078. [CrossRef] 18. Ostgotatrafiken, 2018. Available online: www.ostgotatrafiken.se (accessed on 18 September 2018). 19. Tekniska_verken_AB, 2018. Available online: www.tekniskaverken.se (accessed on 27 August 2018). 20. Zhou, B.; Wu, Y.; Zhou, B.; Wang, R.; Ke, W.; Zhang, S.; Hao, J. Real-world performance of battery electric buses and their life-cycle benefits with respect to energy consumption and carbon dioxide emissions. Energy 2016, 96, 603–613. [CrossRef] 21. Topal, O.; Nakir, I. Total Cost of Ownership Based Economic Analysis of Diesel, CNG and Electric Bus Concepts for the Public Transport in Istanbul City. Energies 2018, 11, 2369. [CrossRef] 22. Ma, Y.; Ke, R.Y.; Han, R.; Tang, B.J. The analysis of the battery electric vehicle’s potentiality of environmental effect: A case study of Beijing from 2016 to 2020. J. Clean. Prod. 2017, 145, 395–406. [CrossRef] 23. Monthly Average Temperature in Sweden, SMHI. Available online: https://www.smhi.se/klimatdata/ meteorologi/temperatur (accessed on 3 December 2018). 24. Göhlich, D.; Ly, T.A.; Kunith, A.; Jefferies, D. Economic assessment of different air-conditioning and heating systems for electric city buses based on comprehensive energetic simulations. World Electr. Veh. J. 2015, 7, 398–406. [CrossRef] 25. Gode, J.M.F.; Hagberg, L.; Palm, D. Miljöfaktaboken 2011: Estimated Emission Factors for Fuels, Electricity, Heat and Transport in Sweden; 2011 VÄRMEFORSK Service AB 101 53 STOCKHOLM; VÄRMEFORSK: Stockholm, Sweden, 2011; ISSN 1653-1248. 26. Gode, J.M.F.; Arnell, J.; Höglund, J. Emissionsfaktorer för Nordisk Elproduktionsmix B2118; IVL Svenska Miljöinstitutet: Stockholm, Sweden, 2012.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).