ABSTRACT This study of the district heating system of Aalborg (Denmark) analyses how fiscal instruments affect the extent excess heat recovery helps reduce the carbon footprint of heat. It builds on a supply-and-demand framework and characterizes the changes in excess heat supply with consequential life cycle assessment in reference to one gigajoule distributed. The heat supply curve is defined through ten scenarios, which represent incremental shares of excess heat as the constraints of the said legal instruments are lifted. The heat demand curve follows the end-users’ response to price changes. The most ambitious scenario doubles the amount of excess heat supplied and reduces the heat carbon footprint by 90% compared to current level, for an end-user price increase of 41%. The price increase results from a higher supply of excess heat at a higher price and an unchanged purchase cost from the coal-fired CHP plant despite a lower supply. This highlights the necessity of a flexible supplier when the share of recovered excess heat is high. 1. Introduction In the 2020 Energy Strategy of the European Commission, the European Union (EU) defined specific climate targets to lower GHG emissions and energy use by 20% and increase alternative energy sources by 20% [1]. A series of more stringent targets are soon to be formulated for 2030 and 2050. A suggested way to reach these targets is to reduce the use of fossil fuels and increase the energy efficiency of specific energy pathways. Industrial symbiosis (IS), a concept defined as the exchange of residual material and energy flows between otherwise unrelated industrial activities within a geographically defined scope [2], is a possible solution for industries and countries to achieve the above-mentioned targets. It is a concept that can potentially improve industrial sustainability and be “an important strategy for low- carbon development” [3]. International Journal of Sustainable Energy Planning and Management Vol. 14 2017 39 However, the development of IS is not without obstacles, as described by various scholars such as Harris (2007), Lehtoranta et al. (2011) and Desrochers (2001) [4–6]. Both Chertow (2004) and Bojsen & Ulhøi (2000) present obstacles of economic, legislative, organizational or physical nature that often prevent the full deployment of an IS system, if not its emergence at all [2, 7]. For instance, a repressive legal framework can limit incentives for companies to utilize and transform waste materials and excess heat [7]. Also, the development of inter-industrial collaborations on residual materials requires time and resources for the participating parties. It is critical for the development of IS that firms have an economic drive (i.e. lower transaction costs) and political support (i.e. specific goals for lowering emission levels and promoting closed-loop systems) [8]. However, the literature documenting the impacts of such obstacles on the development of IS remains broad and theoretical. To our 1 Corresponding author e-mail: sacchi@plan.aau.dk International Journal of Sustainable Energy Planning and Management Vol. 14 2017 39–56 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating !"#$%&'($))*%+ ,'-$&$'."&/0$&0%&"1$'!$#/*21$ !"#$%&'"(&)*+),-$((.(/0)1$-2*%/)3(.4"%5.&60)7"(8529%//$8"):;0)<===)1$-2*%/0)!"('$%> Keywords District heating; Heat recovery; Industrial symbiosis; Excess heat; Life cycle assessment; URL: dx.doi.org/10.5278/ijsepm.2017.14.4 40 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating knowledge, there exist no case studies demonstrating the extent to which such obstacles can prevent IS from developing or provide indication on the performance level one could have hoped to see, had they not existed. The relevance of IS in reducing emissions of greenhouse gases (GHG), notably through the recovery of excess heat for district heating (DH) purposes, has been recognized by governments and scholars alike [9–11]. Heat recovery in the context of DH is understood as the process of retrieving excess heat released by various industrial activities without additional use of energy [10]. The heat is either directly distributed in the DH network for domestic heating purposes or complements the production of a dedicated heat plant. If not recovered, the excess heat usually dissipates into an air or water compartment and is lost. Furthermore, academic studies point at the importance of conducting system studies exemplifying the benefits of excess heat recovery [9]. Therefore, this study precisely intends to model and describe the influence of legal and economic constraints that prevent the full exploitation of the advantages that IS can deliver in terms of carbon footprint reduction. In this case, IS is based on the context of excess heat delivery in the receiving DH network of the city of Aalborg in Denmark. This paper is structured as follows. The next sub- section describes the concept of IS. Section two introduces the case study of Aalborg in details. The case is central for this study as it provides a basis for modelling DH supply scenarios with varying shares of excess heat as explained in section 3, namely “Description of the method”. Section 4 highlights the most relevant results. Main conclusions are drawn in section 5, followed by a discussion in the last section. 1.1. Industrial symbiosis Chertow [12] defines the concept of IS as a network of unrelated firms that share diverse resources such as heat, energy, water, waste materials and even information. It results in economic and environmental benefits for the engaged parties as it leads to reduced production and purchase costs while it also decreases the consumption of virgin resources [13]. There exist numerous examples of self-organized IS networks, which, taken to a larger scale, are called eco-industrial parks. The cases of Kalundborg (Denmark), Guayama (Puerto Rico), Styria (Austria) or Rotterdam (the Netherlands) are only few of the many successful examples documented in the literature [14]. In Kalundborg only, fifty unique synergies take place between the industries in the area. A third of the synergies are concerned with the exchange of water and heat [14]. 2. Case study: Industrial symbiosis and district heating in the city of Aalborg 2.1. Aalborg energy strategy The city of Aalborg is situated in the Region of Northern Denmark and is part of the Municipality of Aalborg. The municipality has translated the Danish energy ambitions of minimal dependence on non-renewable energy sources into a local strategy for fossil-free heat production. An important message is the emphasis on the need for diversified energy sources which include excess heat from industries, heat pumps, solar and wind power and geothermal energy. One of the short-term goals the strategy contains is the increased use of excess heat in the DH system, delivered at both high and low temperature [15]. There is emphasis on striving for cost- effective technologies for sustainable energy production [16]: cost-effectiveness and price for end-users are important factors at the governance level when considering alternative energy sources. 2.2. Aalborg IS synergies Not as known as the IS case of Kalundborg, the city of Aalborg has for the past few years multiplied the cases of inter-industrial collaborations among local companies. Figure 1 illustrates the current synergies in Aalborg. The names of the public and private stakeholders remain undisclosed. A fair share of the synergies is related to the local Portland cement producer. The cement producer is strategically located in the industrial area of Aalborg with a direct access to transportation by water and land. The location eases the receiving of waste materials from the coal-fired combined heat and power plant (CHP plant), the harbour and other neighbouring activities. Most of the exchanges documented cover the processing and exchange of alternative fuels and energy flows. The recovery and exchange of excess heat from industries is one of the main “resources” delivered from industries to the local heat distribution company. Some exchanges are constrained in supply by the demand for the primary product they derive from, e.g. the supply of heat from the waste incineration plant is constrained by the supply of municipal solid waste (MSW). It is itself constrained by the level of consumption activity of the local households. In practice, however, the import of MSW from other regions is always possible. International Journal of Sustainable Energy Planning and Management Vol. 14 2017 41 Romain Sacchi & Yana Konstantinova Ramsheva As this paper focuses on DH, the following paragraphs detail IS synergies in the context of DH. Currently, more than 60% of all Danish households, both urban and suburban, are connected to the DH grid [17]. In the urban area of Aalborg, the DH network supplies 80% all of the building stock heated area [18]. As the Figure 2 [19] depicts, the DH grid is sustained by the following suppliers, listed by ascending order of supply priority: • a waste CHP plant, that co-produces heat and electricity as a result of the thermal treatment of MSW, • the Portland cement producer, that co-produces heat as a result of white Portland clinker production, • the municipal crematorium and wastewater treatment plant, • the coal-fired CHP plant [20], that, when in co- generation mode, co-produces electricity and other co-products along with the supply of heat, • and several small-scale decentralized natural gas-/biomass-fired heat plants in the outskirt of the city that act as a back-up capacity. The share of excess heat in the DH grid in 2016 represents almost 40% of the net amount of heat distributed. It is mainly provided by the waste treatment plant and the Portland cement producer. They deliver a steady amount of heat all year through as the latter is aligned on the supply level of the primary activity they derive from, namely waste volume reduction and white Portland clinker. The coal-fired CHP plant is the only unconstrained and dedicated heat supplier that covers for the short-term variations in heat demand throughout the year. Indeed, while the suppliers of excess heat can vary the amount of heat recovered and supplied through investment in heat recovery equipment, they are not flexible regarding the overall amount of heat produced. Hence, the latter is conditioned by the stochiometric requirements of their main activity. In other words, the amount of excess heat produced depends on the demand for cement or the availability of household waste to treat, for the cement producer and the waste CHP plant respectively. For that reason, excess heat does not bear any environmental burden since a marginal increase in the Meat and bone meal Chalk Gypsum Fly ash Cleaned sand Dredged sand Cooling CEMENT Dried sewage sludge Coal CHP Biodiesel production Aalborg District Heat Supply energy and fuel-based exchange material-based exchange upcoming exchange constrained in supply Waste water Wastewater Crematory Boiler producer Waste CHP Harbour Ship HOSPITAL Human remains Municipal solid waste Ethylene glycol current configuration A nim al residues Livestock w asteAnimal fodder Sea food processing Cooling equipment market market Figure 1: Material and energy flow exchanges in the industrial area of Aalborg 42 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating heat supply via investment in additional heat recovery does not lead to any noticeable increase of the plant emissions or activity. Thus, this study considers excess heat carbon neutral [21] at the margin, as the environmental burden is entirely associated to the primary activity the recovered heat derives from. Unlike recovered excess heat, the production of heat at the coal-fired CHP plant bears all the environmental burden of the plant activity, as the total emissions largely fluctuate according to the demand for heat, not electricity or other co-products when the plant operates in co-generation mode. As such, any additional amount of excess heat in the network displaces demand for heat from the coal-fired CHP plant, which reduces the carbon footprint of a gigajoule (GJ) of heat delivered. It helps the city to achieve its long-term commitment of a GHG- neutral DH system by 2050, set by the Danish government [22]. 2.3. Aalborg IS constraints A number of academic studies look into the advantages of excess heat recovery and conclude that there are still unexploited benefits [9, 10, 23]. Unfortunately, many initiatives to engage in IS synergies often find initial investments in infrastructure or transaction costs very high [8]. It is the case notably with large-scale flue gas condensation units needed to enter the market. Additionally, according to our knowledge, there is currently no sufficient documentation on the importance of economic profitability in the context of IS, apart from a Swedish study, where the authors analyse the unexploited potential for excess heat in the context of Sweden [24]. In Denmark, two national fiscal instruments seem to play a role in the decision of investing in excess heat recovery, both stemming from the Heat Supply Act (Lov om varmeforsyning, in Danish): a tax on recovered energy (usually passed on to the end-users) as well as a price ceiling principle (or price cap), which limits by law the return on investment for heat recovery infrastructures using the so-called substitution principle. This substitution principle sets a price limit for heat that usually corresponds to the lowest purchase price paid to the average conventional heat supplier of the country, namely large- scale coal-fired power plants. The former instrument aims % Coal CHP CEMENT Waste CHP Boilers Crematorium Wastewater Waste water Human remains Municipal solid waste Aalborg District Heat Supply current configuration regular heat supply percentage of heat demand fulfillment excess heat recovery contrained supply potential supplier market market Nonwoven fabric Printers CHP 2 % < 1 % < 1 % < 1 % 21 % 17 % 60 % Figure 2: District heat suppliers in Aalborg in 2016. Source: based on [19] International Journal of Sustainable Energy Planning and Management Vol. 14 2017 43 Romain Sacchi & Yana Konstantinova Ramsheva at avoiding the undesirable production of false excess heat (where the excess heat is purposely produced with the additional use of fuel), while the latter aims at keeping the end-users heat price as low as possible, with the objective to maximize the social and economic welfare of the end- users. The authors consider the latter instrument may lead to a payback time for investment in true excess heat recovery that lays beyond the acceptable time horizon for many private actors in the industrial area of Aalborg. A paradoxical situation arises where the different stakeholders are caught between the will of the government to maximize the social wellbeing of the DH recipients by offering heat at the lowest possible price and the long-term environmental agenda of the city to gradually decarbonize the DH system. The authors wish to study the effects of the fiscal instruments set in place by the government destined to enforce the former agenda on the ability of the city to pursue their environmental objectives. In this context, the present study wishes to answer the three following questions: • What is the effect of the price ceiling principle on the environmental performances of the IS system at delivering low-carbon DH with the current installed capacity? • To what extent can excess heat recovery help the city fulfil its objective of a GHG-neutral DH system? • What would be the economic impact on the end- users? These questions are answered with reference to the distribution of one gigajoule of heat to the end-user at the margin. 3. Description of the method This study applies both qualitative and quantitative research approaches to provide sufficient data and answer the three questions formulated above. This study relies on a supply-and-demand framework for DH in Aalborg in 2016. It helps to capture the changes in the environmental footprint of the distributed heat as more excess heat is introduced in the network. It follows a seven-step approach. Step 1. The current DH supply — identification of the current heat supply capacity of the system Step 2. The current DH demand — collection of data regarding the current demand for heat for the built environment in Aalborg Step 3. Changes in supply of excess heat — cost and energy modelling of different scenarios with varying shares of excess heat in the system Step 4. Changes in demand for DH — evaluation of the response of the end-users to varying shares of excess heat (via the price elasticity of demand) Step 5. Identification of the substitution effect on the marginal heat supplier because of a change in demand Step 6. Market equilibriums — for each scenario, the market equilibrium is calculated Step 7. Carbon footprint analysis — the changes in the system because of the incremental supply of excess heat are characterized with the help of consequential Life Cycle Assessment (LCA) The next sections describe the approaches this study follows to gather the necessary data for each step. 3.1. Step 1. The current DH supply On the supply side, the current capacity of individual heat producers (the coal-fired CHP plant, the waste treatment plant and the Portland cement producer) are modelled based on technology and cost information provided by the Ministry of Energy, published environmental and financial reports as well as direct communications with company representatives during the spring of 2017. A supplier cut-off criterion of 1% is applied: minor heat suppliers such as the local crematorium, the waste water treatment plant and potential newcomers are excluded as the benefits in terms of results completeness or accuracy would not justify the time spent on modelling them. The coal-fired CHP plant and the waste treatment plant illustrate a challenging case of cost allocation (between the production costs of heat and electricity). For the coal-fired CHP plant, an additional task was to distinguish the inputs associated to the co-generation mode, as opposed to those associated to the condensation mode, where only electricity is produced. Indeed, the environmental report only gives the aggregated annual use of inputs and outputs. Hence, with a heat-to-power coefficient of 0.78 (given by the ratio between the heat and electricity nominal power output) and an overall reported conversion efficiency of 91% in co-generation mode, it was possibly to obtain the needed inventory. The V and E allocation method suggested by the Ministry of Treasure [25] is used to split the investment and maintenance costs as well as the fuel inputs between the co-products, to estimate unitary heat production costs 44 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating for heat and electricity, as depicted in Figure 3. The company is free to choose between both V and E allocation methods when reporting fuel use for taxation purpose. Since the share of fuel destined to produce electricity is exempt of taxation, in all logic, the reporting energy company chooses the method that allocates as much resources to electricity production as possible. The same logic is followed in the present model. A 5% profit margin is added on top of the production cost to obtain the per-gigajoule purchase price of heat. It is important to note that, although it is needed at this stage to perform a cost allocation to determine unitary production costs, the emissions of the plant are entirely associated to the production of heat in the LCA model, as the demand for the latter determines the production of both heat and electricity when in co-generation mode. 3.2. Step 2. The current DH needs On the demand side, the current need for heating and the overall heat footprint of each square meter of the built environment in Aalborg are calculated based on the method followed by Kragh and Wittchen (2014) [26] and presented in Figure 4. The Danish Ministry of Statistics provides the detailed distribution of the heated area in Aalborg in 2016 per building type, heating technology and building age intervals [27]. Episcope’s TABULA model, a European harmonized model for measuring and comparing building thermal efficiency across types and locations, is used to estimate the current energy footprint per heated square meter for the 300 different building typologies in Aalborg [28]. Pre-existing Danish building typologies in Episcope were adapted to the context of Aalborg, under the assumption that the Danish building stock is homogenous enough to do so. Some parameters were adjusted to the context of Aalborg, such as weather- related parameters. Additional building typologies were created on top of those existing in the Episcope database. Presumably, reducing the complexity and variety of heat transfers of the whole building stock in Aalborg down to a few dozens of parameters re-arranged into 300 building typologies is done at the expense of accuracy. This is confirmed when the overall heating demand obtained overestimates the real reported heating demand for 2016 by 10%. Nevertheless, the authors believe it provides a solid base for estimating the price elasticity of demand discussed in the next sections. Annual financial reports Annual environmental reports Technology data catalogue for energy plants (Energinet) Investment Fixed and var. operation costs Fixes and var. maintenance costs Mass and energy balance Allocation of total production cost and emissions Unitary production cost and emissions Figure 3: Cross checking of data sources for the approximation of unitary cost and emissions for heat International Journal of Sustainable Energy Planning and Management Vol. 14 2017 45 Romain Sacchi & Yana Konstantinova Ramsheva 3.3. Step 3. Changes in supply of excess heat It is hypothesized in the ‘Introduction’ section that the current legislative framework hinders the recovery of additional excess heat in the DH system of Aalborg. As confirmed through written and face-to-face communications with the current excess heat suppliers [29–31], there is a current price ceiling on heat recovery which strongly limits the overall profitability of heat recovery operations. Thus, it does not allow a payback time short enough for the present investors. In this study, such constraint is lifted to deduct its effect. A triangulation method is used to ensure the validity of the data and the conclusions drawn upon it [32]. First, individual semi-structured interviews with existing excess heat suppliers are conducted. They allow to estimate potential supply of additional excess heat. Additional supply capacity from existing suppliers is mainly achieved by means of investment in: • the extraction of latent heat from the condensation contained in the flue gases, • the enhanced extraction of latent heat from the flue gases below the dew point with marine scrubbers, further aided by large-scale heat pumps, • the recovery of radiative heat on rotary kilns. Second, the financial investments required for retrieving the additional excess heat in question are estimated, completed by the Ministry of Energy’s Technology Data for Energy plants documentation [33]. With knowledge on the potential additional heat supply, the associated financial investment and the acceptable investment payback time of the suppliers, an excess heat supply curve in relation to the price level offered to the different suppliers is obtained. The heat supply curve is defined throughout a total of 10 scenarios presented in Table 1. All scenarios supply the demanded amount of heat in 2016 of 6.715 TJ (minus the demand change because of the price elasticity of demand described in the next section), in addition to a 17.5% network loss. They are listed with an incremental share of excess heat, to draw a supply curve for excess heat. • Scenario 0 corresponds to a supply mix without the presence of excess heat, but only the coal- fired CHP plant; • Scenario 1 adds the heat supply from the small- scale CHP plants located at the outskirt of the city, i.e. biomass and natural gas heat plants; • Scenario 2 includes the excess heat delivery from the waste treatment plant; • Scenario 3 adds excess heat from one of the recovery units located at the cement factory; • Scenario 4 adds a second heat recovery unit located at the cement factory and represents the Country Denmark Denmark Denmark Denmark Denmark Age interval 2011- 2007-2010 1999-2006 1979-1998 1973-1978 kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year kWh/m2/year Single family house Terraced house Building ... ... ... ... ... ... ... ... ... ... ... ... Age interval District heat demand Aalborg Aalborg Aalborg Aalborg Aalborg 2010-2014 2005-2009 2000-2004 1995-1999 1990-1994 kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year kWh/year ... ... ... ... ... ... ... ... ... ... ... Single family house Terraced house Building ... Aalborg-specific parameters adjusted Heating degree-days Average number of storeys per building type Age interval: 1995-1999 Aalborg District heating Gas boiler Biomass boiler Electrical appliances Age interval: 2000-2004 Aalborg District heating Gas boiler Biomass boiler Electrical appliances Age interval: 2005-2009 Aalborg District heating Gas boiler Biomass boiler Electrical appliances Single family house Age interval: 2010-2014 Aalborg District heating Gas boiler ... M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 M2 Biomass boiler Terraced house Buildings Public buildings Sport facilities ... Electrical appliances Figure 4: Illustration of the method followed to define the demand for district heat in Aalborg [27, 28] 46 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating T ab le 1 : E xc es s he at s up pl ie rs m ix f or e ac h of t he 1 0 sc en ar io s Sc en . Sc en . Sc en . Sc en . Sc en . 4 Sc en . Sc en . Sc en . Sc en . Sc en . Su pp lie r D es cr ip ti on 0 1 2 3 (c ur re nt ) 5 6 7 8 9 C oa l-f ir ed C H P pl an t • C oa l- fi re d C H P p la nt x x x x x x x x x x • H ea t p ow er o ut pu t o f 42 0 M W a t 9 0° C • L H V o f co al : 2 3 G J/ t • 85 % n et e ff ic ie nc y ra te in co -g en er at io n m od e in 2 01 6 (9 1% g ro ss e ff ic ie nc y ra te ) • 0. 78 G J of e le ct ri ci ty c o- pr od uc ed pe r G J of h ea t d el iv er ed • A nn ua l o pe ra tin g tim e: 8 76 0 ho ur s O ut sk ir t C H P pl an ts S m al l- sc al e bi om as s an d na tu ra l g as h ea t p la nt s. x x x x x x x x x W as te C H P pl an t • 94 % e ff ic ie nc y (h ea t a nd e le ct ri ci ty ) x x x x x x x x • L H V o f M S W : a ro un d 11 G J/ t • P ow er o ut o f 50 M W a t 8 0° C • 0. 3 G J of e le ct ri ci ty p er G J of h ea t d el iv er ed • A nn ua l o pe ra tin g tim e: 8 14 2 ho ur s C em en t p ro du ce r — R ec ov er y un it 1 • D ou bl e st ag e he at e xc ha ng er x x x x x x x • P ow er o ut pu t o f 37 M W a t 7 4° C • P ow er o ut pu t o f 54 M W a t 6 0° C • A nn ua l o pe ra tin g tim e: 7 44 0 ho ur s C em en t p ro du ce r — R ec ov er y un it 2 • D ou bl e st ag e he at e xc ha ng er x x x x x x • P ow er o ut pu t o f 15 M W a t 7 4° C • P ow er o ut pu t o f 24 M W a t 6 0° C • A nn ua l o pe ra tin g tim e: 7 44 0 ho ur s C em en t p ro du ce r — R ec ov er y un it 3 • D ou bl e st ag e he at e xc ha ng er x x x x x • P ow er o ut pu t o f 19 M W a t 7 4° C • P ow er o ut pu t o f 60 M W a t 6 0° C • A nn ua l o pe ra tin g tim e: 7 44 0 ho ur s • E st im at ed in ve st m en t: 8 M ! • E st im at ed a nn ua l m ai nt en an ce : 0 .4 4 M ! W as te C H P pl an t — E xt ra r ec ov er y un it • C on de ns in g he at r ec ov er y un it x x x x • po w er o ut pu t o f 3, 4 M W a t 8 0° C • A nn ua l o pe ra tin g tim e: 8 14 2 ho ur s • E st im at ed in ve st m en t: 9. 1 M ! C em en t p ro du ce r — R ec ov er y un it 4 • D ry f lu e ga s he at r ec ov er y un it x x x • P ow er o ut pu t o f 6 M W a t o ve r 10 0° C • A nn ua l o pe ra tin g tim e: 7 44 0 ho ur s • E st im at ed in ve st m en t: 4. 43 M ! C em en t p ro du ce r — R ec ov er y un it 5 • D ry f lu e ga s he at r ec ov er y un it x x • P ow er o ut pu t o f 15 M W a t o ve r 10 0° C • A nn ua l o pe ra tin g tim e: 7 44 0 ho ur s • E st im at ed in ve st m en t: 9. 4 M ! H ea t p um p • L ar ge s ca le h ea t p um ps x • N um be r of h ea t p um ps r eq ui re d: 38 a t 4 M W th • S ou rc e te m p. in : 6 0° C , s ou rc e te m p. o ut : 3 0° C • S ou rc e si nk in : 4 2° C , s ou rc e si nk o ut : 8 0° C • L or en tz C O P : 1 2. 6 • A nn ua l o pe ra tin g tim e: 7 44 0 ho ur s • E st im at ed a nn ua l c os t: 28 M ! E st im at ed s ha re o f ex ce ss h ea t in t he D H s ys te m (% ) 0 0 22 37 43 50 51 57 60 90 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 47 Romain Sacchi & Yana Konstantinova Ramsheva current situation in Aalborg (also previously presented in Figure 2); • Scenario 5 builds on Scenario 4 and adds the heat delivered by a third recovery unit at the cement factory; • Scenario 6 includes an additional heat recovery from the waste treatment plant; • Scenario 7 and 8 add radiative heat from the rotary kilns of the cement producer; • The additional excess heat in Scenario 9 is obtained by the further recovery of latent heat below the dew point of the flue gases of the cement kilns. Scenario 9 implies that the DH network delivery temperature of 74°C is reduced to 60°C. Recovering heat at a lower temperature allows to extract additional latent energy contained in the moisture of the flue gas. It results in more energy at a lower temperature. The recovered heat is then increased to 80°C with several large-scale 4MWth heat pumps and an auxiliary use of electricity, to be usable for residential heating. In earlier scenarios, where the excess heat is delivered at 74°C by the cement factory, the coal-fired CHP plant had the task to add 6°C to the recovered heat to reach an average temperature of 80°C. Recovering the heat at a lower temperature also entails the purchase and installation of water treatment infrastructures to handle and clean the additional condensate water. It also encompasses additional maintenance costs that result from the fouling effect of gypsum-rich condensate water on the heat exchangers. 3.4. Step 4. Changes in demand for DH This step concerns the modelling of the response of the demand to changes in the end-user DH price. Heating is a necessity good for which the demand is rather inelastic. While an increase in the DH price leads to a decrease in the heat amount demanded, it is necessarily compensated to keep a constant amount of indoor comfort. The possibility for the end-users to switch to alternative means of heating as a response to increments of the DH price is discarded. It is indeed legally and economically difficult to do so once the heated area is connected to the DH grid. A permission to stop using the DH network needs to be asked to the relevant authority. When granted, the annual fixed part of the DH end-user subscription, used to finance the connection to the DH sub-station, still needs to be paid, rendering the switch to other sources of heating uneconomical, albeit not impossible. The study makes instead the simplifying assumption that the reduced demand for DH, because of a price increase, is instead displaced on the thermal renovation of the building envelope. Sourcing from a database that contains updated prices on thermal renovation projects in Denmark (Molio pris database) [34], a solver is used to find the optimal combination of renovation works for each building typology in Aalborg in order to comply with the current regulation for renovated buildings (BR2015) [35] at a minimal cost. The need for additional insulation is calculated with reference to an indoor comfort temperature of 20°C, as indicated in the BR2015 regulation, with a 5-year average annual heating degree-days for the region of Aalborg. A series of constraints have been added to the solver. For example, buildings before 1930 cannot undergo façade walls renovation (for aesthetic preservation reasons), while only buildings built between 1900 and 1950 can undergo wall cavities filling with glass-blown granulates (buildings built after 1950 are assumed to be already insulated that way). Additionally, the economic cost of wall insulation from the inside includes the lost liveable indoor area multiplied by the current average square meter cost in Aalborg. Estimating the cost of thermal renovation for each building type allows to define the DH price level at which the building owner would rather invest in the insulation of the building envelope rather than accept the change in DH price (price elasticity of demand). The heavy assumption made here is that building owners follow a strictly economic rationale, which might not always be true in practice. The decision of investment happens when the return on investment over the lifetime of the renovation project (that is the ratio between the avoided heating cost over the project lifetime and the total cost of insulation) reaches a ratio of 1.33. While the ratio of 1.33 may seem arbitrary, it is the one considered by the BR2015 guidelines [35]. Such exercise allows to approximate a demand curve for DH in relation to its price, presented later in Section 4.1 ‘The DH demand curve’. 3.5. Step 5. Substitution effects and price elasticity of demand As the scenarios introduce an increasing share of excess heat in the DH grid, the price and quantities purchased 48 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating from each supplier by the utility company to satisfy the demand change too. Additionally, as the share of excess heat supply increases, the share of heat supplied by the dedicated coal-fired CHP plant reduces to keep a constant supply output on the market. Doing so increases the unitary heat price from that coal-fired CHP plant as fixed capital and investment costs allocated to the production of heat still run despite a lower production level. The relation between the amount of heat demanded from the coal-fired CHP plant and the purchase price level is depicted in Figure 5. In parallel, a reduced heat delivery from the coal-fired CHP plant also leads to a reduced co-delivery of electricity (as the coefficient of co-production of the coal-fired CHP plant between both outputs in co- generation mode is assumed constant). The missing delivery of electricity will be compensated by an equivalent production of electricity from a mix of marginal electricity-supplying technologies in Denmark, coming mostly from biomass and wind power [36]. It is assumed that the energy distribution company runs at marginal costs — which is confirmed by the two latest financial reports of the Aalborg utility company [37, 38]. This means that any increase in heat purchase costs translates in an increase in price on the side of the end-users. Considering the price elasticity of demand, an increase in price for the end-users translates in an overall reduced demand for heat. Such reduction is obtained from the demand curve calculated in Step 4. The model reduces the required heat supply by an equivalent amount from the marginally least-preferred heat supplier in the mix of suppliers, namely the coal-fired CHP plant. This returns a mid- to long-term market equilibrium (Step 6, discussed in Section 3.5) for which the carbon footprint of a distributed gigajoule of heat can be calculated (Step 7, discussed in Section 3.6). At the same time, as the demand reacts (decreases) and the supply from the coal-fired CHP plant reduces, the model considers the amount of transportation activity, the production of insulation materials and all other requirements necessary to support the renovation works needed to preserve the initial indoor comfort temperature of 20°C on the share of the building stock that reacts to the DH price change. Common thermal transmittance values are used for mineral wool and a local production is assumed for most materials (mineral wool, concrete, bricks, gypsum boards, windows, doors, ventilation systems, etc.) as well as an average transportation distance of 150 km from their production facility to the renovation site. 3.6. Step 6. Market equilibriums Knowing the calculated supply and demand preferences at any given price level allows to calculate the market equilibrium for each scenario. As the share of excess heat increases through the scenarios, it returns a new end-user price level. The latter induces a decrease in demand for DH and an increase in demand for thermal insulation. This affects in turn the supply of DH and its price level. This is the mid-term equilibrium at which a new DH quantity is supplied for a corresponding end- user price level. 3.7. Step 7. Carbon footprint analysis ‘Carbon footprint’, as defined by Wiedmann and Minx (2008) is the amount of GHG emitted through the life cycle of a product or service, supplied by an organization or by a process [39]. To consider the consequences of increasing the supply of excess heat on the carbon footprint of the heat produced, a consequential LCA is conducted and thus provides a comparison of the environmental impacts of each scenario [40]. The results from the LCA can give a good starting point for developing a discussion on the potential solutions for Aalborg in delivering GHG- neutral DH to its end-users with the current available resources. When the market equilibriums for the different scenarios are defined, the LCA model calculates the carbon footprint of one GJ of heat distributed to the end- user at the margin as an increased share of excess heat is introduced in the system. The material and energy 0 200 400 600 800 1000 2000 4000 6000 8000 10000 Amount of heat supplied [TJ] P ric e In de x Amount supplied function of price level Current price level (Index 100) Figure 5: Relation between the amount of heat purchased from the coal-fired CHP plant and the purchase price level International Journal of Sustainable Energy Planning and Management Vol. 14 2017 49 Romain Sacchi & Yana Konstantinova Ramsheva requirements to deliver one GJ of heat in each scenario are modelled in OpenLCA, the LCA tool used in this study. Background processes — e.g. transport activities, production of insulation material, supply of electricity — are modelled using the consequential life cycle inventory database ecoinvent version 3.3. The presence of large uncertainties in the underlying economic model of each supplier calls for the use of uncertainty-handling techniques, such as the Monte Carlo analysis. To do so, uncertain parameter inputs were identified, in accordance with the communications with the excess heat suppliers. An uncertainty distribution profile was associated to several of these parameters. Table 2 lists some of the uncertain parameters at the excess heat recovery level. The Monte Carlo algorithm iterates 1,000 times through each scenario. For each iteration, a random variable is picked within the uncertainty distribution of each model input for which uncertainty was defined. The algorithm then builds a technology matrix which is passed to the LCA solver class of OpenLCA. The solver multiplies the inverse of the technology matrix by an environmental matrix and a demand vector to return the total material and energy inventory of each scenario. Table 2: Non-exhaustive list of parameters and associated uncertainty distributions Probability distributionDistribution parameters Uniform Mean: 1.06 Min: 1.05 Max: 1.08 Triangular Mode: 0.175 Min: 0.15 Max: 0.2 Triangular Mode: 7440 Min: 7000 Max: 8000 Uniform Mean: 150 Min: 100 Max: 200 Triangular Mode: 77 Min: 72 Max: 82 Description GJ of heat in at 60°C for every GJ of heat out at 80°C delivered by the heat pumps. Heat loss ratio in the DH network. Annual number of hours of operation at the cement factory. Distance for the delivery of materials necessary to thermal renovation works. Purchase price of heat from supplier 1. Parameter name GJ_in_out_HP Network_loss Operating_time_ap Transport_distance Price_supplier_1 17 0 1.086 1.135 27 0 0.150 0.200 23 0 7.037E3 7.9 17 0 1.001E2 2.00 26 0 72.024 81.728 50 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating The inventory is then multiplied by the characterization factors provided by the IPCC Global Warming 100a method to obtain a carbon footprint expressed in kg of CO2-eq per GJ distributed with a time horizon of 100 years. 4. Results interpretation This section details the calculated demand (Section 4.1) and supply (Section 4.2) curves as well as the resulting changes in the carbon footprint of one gigajoule of heat (Section 4.3). 4.1. The DH demand curve Figure 6 illustrates the relation between demanded district heat and the heat price index in Aalborg. The red polynomial regression curve allows to approximate a demand curve for DH in relation to its price. The curve is later used to find market equilibriums for each heat supply mix scenario, presented in the next sub-section. Two groups of buildings are more prone than others to react to DH price increments: • Old buildings that would find a significant reduction of their energy footprint through insulation, • Relatively recent buildings, often public institutions, with a large heated area and a well- insulated envelope that would find interest in upgrading, at limited costs, to a newer ventilation system with heat recovery. The mild slope of the demand curve indicates a rather inelastic demand to DH price changes. It is because the price elasticity of demand in this study is defined in a context where the change to other means of heating is not permitted and where the building owners act rationally, as explained in the Section 3.4. Thermal renovation projects are expensive and ROI of 1.33 are only reached at high DH price increase. In practice, some building owners would switch to another source of heating, while others would simply not notice price changes. Real and measured data would likely differ with the demand curve illustrated below. 4.2. The new market equilibriums Figure 7 shows the market equilibriums reached for the 10 scenarios. From left to right, each scenario introduces an increasing amount of excess heat in the supply mix. Scenario 4 represents the current situation in Aalborg. Scenarios 5 to 9 represent market 10 0 11 9 13 2 14 5 15 8 17 1 18 3 19 6 20 9 22 2 23 5 24 7 26 0 27 3 28 6 29 9 31 2 32 4 33 7 35 0 36 3 37 6 38 8 40 1 41 4 42 7 44 0 45 3 46 5 47 8 49 1 50 4 51 7 52 9 54 2 55 5 56 8 58 1 59 4 60 6 61 9 63 2 64 5 65 8 67 1 68 3 69 6 70 9 72 2 73 5 DH price Index 100 (100 = current price) 0 1000 2000 3000 4000 5000 6000 7000 8000 H ea t d em an d [T J] Poly. regression of demand variation 0.00138× 2 -4.532× +8353 Farm houses Single family houses Terraced houses Multi-storey buildings Student accomodations Public institutions Other permanent residences Other services Offices/services Hotels Cultural buildings Teaching facilities Hospitals Daycare facilities Other institutions Sport facilities Other leisure facilities Figure 6: Relation between demanded district heat and the heat price level in Aalborg International Journal of Sustainable Energy Planning and Management Vol. 14 2017 51 Romain Sacchi & Yana Konstantinova Ramsheva Demanded amount Outskirt CHPs Waste CHP Cement producer - Recovery unit 1 Cement producer - Recovery unit 2 Cement producer - Recovery unit 3 Cement producer - Recovery unit 4 Cement producer - Recovery unit 5 Heat pump Coal-fired CHP (demand-adjusted) Avoided through insulation Waste CHP - Extra recovery unit 0 1000 2000 3000 4000 5000 6000 7000 Scenario 0 Scenario 1 Scenario 2 Scenario 3 Scenario 4 (current) Scenario 5 Scenario 6 Scenario 7 Scenario 8 Scenario 9 H ea t s up pl ie d [T J] 79 85 96 99 100 103 105 107 111 141 End-user price Index (Scenario 4 = 100) Figure 7: Market equilibriums for each supply mix scenario (red scatter = demand curve for heat). From left to right, increasing shares of excess heat are introduced equilibriums where the constraint from the price ceiling principle is lifted. Each scenario corresponds to a new price level. As the price level increases, the demand for heat, represented by the red demand curve, decreases. The least-preferred supplier, the coal-fired CHP plant, is affected by the decrease in demand. The missing heat output is compensated by means of heat preservation through insulation of the building stock. Scenario 9 introduces the use of heat pumps to boost the temperature level of the heat recovered at the cement producer recovery units. Hence, the share of the cement producer supply increases. Additionally, the red segment represents the additional supply of heat generated by the auxiliary input of electricity in the heat pumps (calculated as the difference between the amount of heat transferred to the sink and the amount of heat transferred from the source, for the calculated coefficient of operation). The evolution of the unitary price can be seen in Figure 8. Scenario 9 reaches an excess heat share of 90% for a price increase of 41% compared to Scenario 4. This price increase is the result of the combination of two cost-related aspects. There is an increased cost of purchase of excess heat on one hand and a constant cost of purchase of heat from the coal-fired CHP plant on the other hand. 65% of the purchase cost increase is associated with the investment and maintenance of 38 4MWth heat pumps. They represent an annual cost of 28 M", of which almost a third (or 20% of the additional purchase cost) is a tax applied on the use of electricity in the context of heat production. The remaining of the purchase cost increase (35%) comes from an increased volume of excess heat purchased from the cement producer and the waste CHP plant at a higher price (that reflects investments in heat recovery equipment and infrastructures to collect and treat condensate water). It is to note that a fourth of these 35% represents the tax on recovered energy discussed in Section 2.3. Despite its reduced supply throughout the scenarios, the coal-fired CHP plant needs to ensure the role of flexible heat supplier. It can adapt to short-termed seasonal demand fluctuations and complete the supply to meet demand peaks. At the same time, it needs to cover fixed and running expenses despite a lower heat production output level (reference to Figure 5). This creates a lock-in situation where the virtually unchanged 52 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating cost of purchase of heat from the coal-fired CHP plant prevents any savings that could be used to finance the above-mentioned investments. 4.3. Carbon footprint results The results of the Monte Carlo simulation analysis are presented in Figure 9. Uncertainty in the model inputs propagates throughout the outputs. For that reason, it is difficult to conclude on a clear carbon footprint improvement for Scenario 4 over Scenario 3. However, there is a clear statistical improvement trend as the share of excess heat in the DH system increases. For example, compared to the current estimated carbon footprint of about 153 kg of CO2-eq per GJ distributed (Scenario 4), Scenario 9 delivers a GJ of heat at almost a tenth of that value (about 11 kg of CO2-eq per GJ distributed on average). This shows that untapped potential in excess heat recovery can lead to substantially lower carbon footprint levels for the DH system, and bring the city closer to its GHG-neutral heat delivery objective. 5. Conclusions In the introduction of the paper, three research questions were raised. This section aims to provide answers to each of them, based on the results presented in the preceding section. Furthermore, it elaborates on some weaknesses of the model and what possible drawbacks those can have on the results of the study. Q1: What is the effect of the price ceiling principle on the environmental performance of the IS system at delivering low-carbon DH with the current installed capacity? The below country-average price for the end- users of the DH network in Aalborg is a result of a political decision [41]. Nevertheless, such a price ceiling on the DH suppliers’ side may restrict further capital- intensive investments in excess heat recovery. Indeed, this study indicates that the amount of excess heat supplied could be at most multiplied by two, had favourable economic conditions been in place. In other words, the effect of the price ceiling principle on the environmental performance of the DH system is to drastically limit the potential for carbon footprint reduction, aside from keeping the end-users price low. The LCA analysis conducted in this study indicates that increasing the supply of excess heat can substantially reduce the need for coal-based heat, and, altogether with a demand displacement effect, lead to a reduction of the heat carbon footprint by 93% compared to the current situation. As discussed in the Section 2.2, such answer S ce na rio 0 S ce na rio 1 S ce na rio 2 S ce na rio 3 S ce na rio 4 S ce na rio 5 S ce na rio 6 S ce na rio 7 S ce na rio 8 S ce na rio 9 0.0 5.0 10.0 15.0 20.0 H ea t p ric e [# p er G J di st rib ut ed ] Distribution costs Administration costs Other Production costs Other income Other expenses Outskirt CHPs Waste CHP Cement producer - recovery unit 1 Cement producer - recovery unit 2 Cement producer - recovery unit 3 Cement producer - recovery unit 4 Cement producer - recovery unit 5 Coal-fired CHP Heat pump Waste CHP - extra recovery unit Figure 8: Unitary price structure per GJ distributed International Journal of Sustainable Energy Planning and Management Vol. 14 2017 53 Romain Sacchi & Yana Konstantinova Ramsheva holds on the assumption that recovered excess heat does not bear any of the environmental burden associated to the industrial process. Q2: To what extent can excess heat recovery help the city fulfil its objective of a GHG-neutral DH system? The different scenarios built in this study follow the short- and longer-term strategies of Aalborg Municipality of moving away from fossil fuels for heating purposes, e.g. using high-, medium- and low temperature heat from industries. The most ambitious scenario (Scenario 9) results in a tenfold lowering of the carbon footprint of the heat compared to the current scenario (Scenario 4) i.e. from 153 kg of CO2-eq. per GJ distributed down to about 11 kg of CO2-eq., provided that the share of excess heat grows from 43% to 90% of the supply mix. Q3: What would be the economic impact on the end- users? As presented in Section 3.3 ‘Changes in supply of excess heat’, a share of excess heat as high as 90% of the gross supply mix can be achieved through capital- intensive investments in various equipment. It results in an increase of the end-user price of 41% compared to the current price level, ceteri paribus. The market equilibrium for each scenario is calculated after consideration of the demand elasticity and displacement of the demand for alternatives, i.e. thermal insulation of the building envelope. The results from section 4.1 ‘The DH demand curve’ show that insulation is a preferred strategy for old buildings, while more recent buildings with large heated area are rather upgraded with a new mechanical ventilation system with indoor heat recovery. But the reader should be aware that the conclusions of this study hold on the assumption that building owners act rationally and that the decision of insulating a house is taken as soon as it is economically viable to do so. This assumption is, without a doubt, weighting heavily on the calculation of the demand elasticity. Some building owners may decide to undergo building renovation well before the project reaches a ROI of 1.33, while others may be unaware of heating price changes. While such uncertainty may have an influence on the end-results, the authors assume the above-described market dynamics and the conclusions drawn from them would remain unchanged. 6. Discussion A potential drawback about the proposed scenarios in this study is the focus on excess heat as the only viable alternative to coal-based heat. Instead, diversifying the energy sources (e.g. geothermal, wind power-to-heat), an ambition very high on the agenda of Aalborg Municipality [16], could be a plausible alternative. Diversifying the energy mix secures against volatility of prices and supply levels [42]. Scenario 0 Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 5 Scenario 6 Scenario 7 Scenario 8 Scenario 9 0 100 200 300 400 500 600 kg o f C O 2- eq . p er G J di st rib ut ed 0 20 40 60 80 100 120 140 160 E nd-user price index (S cenario 4 = Index 100) Monte Carlo results (1,000 iterations) 254.61 255.52 248.23 191.5 153.34 138.34 119.47 103.71 97.29 10.84 79 85 96 99 100 103 105 107 111 141 Figure 9: Monte Carlo simulation results for each supply mix scenario (red values = carbon footprint median, blue scatter = end-user price level index). Boxes represent 50% of the distribution while the interval defined within the black horizontal lines represents 90% of the distribution. Cross-shaped points are outliers to the distribution 54 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating However, alternative (renewable) energy sources can be costly to implement compared to excess heat recovery. For illustrative purpose, Table 3 shows the average nominal investment per MW of heat for common district heating technologies given by the Danish Ministry of Energy and compares it to the average nominal investment associated to the capacity increment between Scenario 4 (current capacity) and Scenario 9 (most ambitious scenario). Between scenarios 4 and 9, an additional 125 MW of excess heat are installed, for an average nominal investment of 0.2 M"/MW. The low nominal investment figures are explained by the fact that infrastructures already exist and a substantial part of the economic burden is sustained by the activity the heat is a co-product of. The analysis of the price increase between Scenario 4 and 9 in Section 4.2 shows that the necessity to keep a flexible heat supplier prevents savings that could partly or entirely finance the needed investment in heat recovery equipment, leading to an overall price increase. The supply agreement between the city of Aalborg and the coal-fired CHP plant will last until 2027, after which a national directive phases out heat and electricity production from coal. Hence, should the excess heat recovery be taken to an extent similar to Scenario 9, it would be desirable to opt for a small- to medium-scale renewable-based heat- producing technology that has the ability to adjust to short- term demand variations (e.g. biomass or biogas). From a time perspective, the excess heat supply is constrained and hardly flexible. There is a risk that the system over-supplies in the summer, when the demand for heat is low, and under-supplies in winter, when the demand for heat is high. To consider the seasonal profile of the demand for heat would require heat storage solutions. This would certainly add an additional economic burden on the end-users. On a system level, it could also be argued that this study does not fully reflect the positive impacts that IS brings in a global perspective, but only relative to the carbon footprint of the district heat. The case of the DH system of Aalborg was selected due to the ongoing, upcoming and potential IS synergies. The study showcases the benefits that can potentially be drawn from a fully-deployed IS system for both surrounding industries and the society, with an angle on municipal excess heat delivery. Yet, the aim of this study is not to investigate an optimal energy mix for supplying DH to Aalborg, but rather to demonstrate that the city of Aalborg can achieve its ambition of providing a cleaner heat with the available, yet untapped, resources without investing in technologies that require heavy investments and without the use of additional fuel. The conclusions are relevant for an international audience with an interest in IS, since they provide general insights on how legal and economic instruments can hinder the full development of collaborative industrial projects. Acknowledgements The authors of this paper would like to thank several persons: Mads Kristian Ullitz and Michael Rosengreen Christensen at Aalborg Portland A/S for their continuous support, and Kim B. Wittchen, Senior Researcher at the Danish Building Research Institute, for his valued guidance and work behind the Episcope model. The Table 3: Comparison of nominal investment per MW between common district heating technologies and the additional excess heat supply capacity scenarized in this study Nominal Investment (M!/MW-heat) Source Solar 2 [34] Geothermal 2 Wave Power 7.8 Electrolysis 1.4 – 6 Centralized Biogas 3.4 – 5.8 Cement producer — Recover Unit 1 0.03 Present study Cement producer — Recover Unit 2 0.04 Cement producer — Recover Unit 3 0.1 Cement producer — Recover Unit 4 0.7 Cement producer — Recover Unit 5 0.6 Waste CHP plant — Extra recovery unit 0.2 Heat pumps 0.2 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 55 Romain Sacchi & Yana Konstantinova Ramsheva authors would also like to thank Frede Hvelplund, Professor at the Department of Planning of Aalborg University, for the constructive discussions. Finally, the authors appreciate the help and time from the representatives at the Planning department of Aalborg Municipality, the utility company Aalborg Varme A/S and the waste treatment company I/S Reno-Nord. This study has been funded by Denmark’s InnovationsFonden and the Research, Quality and Technical centre of Cementir Holding S.p.A. References [1] European Commission. Energy 2020 A strategy for competitive, sustainable and secure energy. 2011. doi:10.2833/78930. https://ec.europa.eu/clima/change/ causes_en [2] Chertow MR. Industrial Symbiosis. Encyclopedia of Energy, Volume 3 2004;42:407–15. doi:10.1080/00343400701874123. http://www.worldcat.org/title/encyclopedia-of-energy/oclc/ 180851239 [3] Walls J, Paquin R. Organizational Perspectives of Industrial Symbiosis: A Review and Synthesis. Organization & Environment 2015;28:32–53. doi:doi: 10.1177/ 1086026615575333. http://journals.sagepub.com/doi/abs/ 10.1177/1086026615575333 [4] Harris S. The Potential Role of Industrial Symbiosis in Combating Global Warming 2007. http://www.hkccf.org/ download/iccc2007/30May/S3B/Steve%20HARRIS/The%20P otential%20Role%20of%20Industrial%20Symbiosis%20in%2 0Combating%20Global%20Warming.pdf [5] Lehtoranta S, Nissinen A, Mattila T, Melanen M. Industrial symbiosis and the policy instruments of sustainable consumption and production. Journal of Cleaner Production 2011;19:1865-75. doi:10.1016/j.jclepro.2011.04.002. http:// www.sciencedirect.com/science/article/pii/S0959652611001181 [6] Desrochers P. Cities and industrial symbiosis: Some historical perspectives and policy implications. Journal of Industrial Ecology 2001;5:29–44. doi:10.1162/10881980160084024. http://onlinelibrary.wiley.com/doi/10.1162/1088198016008402 4/full [7] Bojsen NI, Ulhøi JP. Industrial symbiosis in an extended perspective. Department of Organisation & Management The Aarhus School of Business 2000:24. https://www. researchgate.net/profile/John_Ulhoi/publication/5092802_Indu strial_Symbiosis_in_an_Extended_Perspective/links/0c960530 32468c49de000000.pdf [8] Yap NT, Devlin JF. Explaining Industrial Symbiosis Emergence, Development, and Disruption: A Multilevel Analytical Framework. Journal of Industrial Ecology 2017;21:6-15. doi:10.1111/jiec.12398. http://onlinelibrary. wiley.com/doi/10.1111/jiec.12398/full [9] Broberg Viklund S, Karlsson M. Industrial excess heat use: Systems analysis and CO2 emissions reduction. Applied Energy 2015;152:189–97. doi:10.1016/j.apenergy.2014.12.023. https://pdfs.semanticscholar.org/c342/8df4063f6bb76b6f4f9f2e 60111a89bf1d7c.pdf [10] Olsson L, Wetterlund E, Söderström M. Assessing the climate impact of district heating systems with combined heat and power production and industrial excess heat. Resources, Conservation and Recycling 2015;96:31–9. doi:10.1016 /j.resconrec.2015.01.006. https://www.diva-portal. org/smash/ get/diva2:789778/FULLTEXT01.pdf [11] Jönsson J, Svensson IL, Berntsson T, Moshfegh B. Excess heat from kraft pulp mills: Trade-offs between internal and external use in the case of Sweden-Part 2: Results for future energy market scenarios. Energy Policy 2008;36:4186–97. doi:10.1016/j.enpol.2008.07.027. http://www.sciencedirect. com/science/article/pii/S0301421508003613 [12] Chertow MR. Industrial Symbiosis. Encyclopedia of Energy 2004;3:407–15. doi:10.1111/j.1530-9290.2012.00455.x. http:// w w w . w o r l d c a t . o r g / t i t l e / e n c y c l o p e d i a - o f - e n e r g y / oclc/180851239 [13] Chertow MR. “Uncovering” Industrial Symbiosis. Journal of Industrial Ecology 2007;11:11-30. doi:10.1162/jiec.2007.1110. http://onlinelibrary.wiley.com/doi/10.1162/jiec.2007.1110/abstract [14] Chertow M, Ehrenfeld J. Organizing Self-Organizing Systems: Toward a Theory of Industrial Symbiosis. Journal of Industrial Ecology 2012;16:13–27. doi:10.1111/j.1530-9290. 2011.00450.x. http://onlinelibrary.wiley.com/doi/10.1111/ j.1530-9290.2011.00450.x/abstract [15] Aalborg Varme A/S. Teknologikatalog. 2017. https://www. aalborg.dk/media/6779880/teknologikatalog.pdf [16] Aalborg Kommune. Strategi for fossilfri varmeproduktion. Aalborg: 2017. https://www.aalborg.dk/media/6779884/ strategi- for-fossilfri-varmeproduktion.pdf [17] Danish Energy Agency. Regulation and planning of district heating in Denmark. 2016. https://ens.dk/sites/ens.dk/files/ Globalcooperation/regulation_and_planning_of_district_heatin g_in_denmark.pdf [18] Aalborg Energikoncern, Aalborg Varme A/S. Grønt regnskab 2015. Aalborg: 2016. https://aalborgforsyning.dk/media/ 536404/groent-regnskab-2015.pdf [19] Aaborg Varme. Omstilling af Varmeproduktion. 2017. https:// www.linkedin.com/pulse/omstilling-af-varmeproduktion- a a l b o r g - v a r m e - 2 0 1 7 - a n n a - b o b a c h ? a r t i c l e I d = 6281083727846219776 [20] The Public Utility Companies. Sustainable utility supply in Aalborg. 2004. https://www.aalborgforsyning.dk/media/ 1810/rapport-uk.pdf https://ec.europa.eu/clima/change/causes_en http://www.worldcat.org/title/encyclopedia-of-energy/oclc/180851239 http://journals.sagepub.com/doi/abs/10.1177/1086026615575333 http://www.hkccf.org/download/iccc2007/30May/S3B/Steve%20HARRIS/The%20Potential%20Role%20of%20Industrial%20Symbiosis%20in%20Combating%20Global%20Warming.pdf http://www.sciencedirect.com/science/article/pii/S0959652611001181 http://onlinelibrary.wiley.com/doi/10.1162/10881980160084024/full https://www.researchgate.net/profile/John_Ulhoi/publication/5092802_Industrial_Symbiosis_in_an_Extended_Perspective/links/0c96053032468c49de000000.pdf http://onlinelibrary. wiley.com/doi/10.1111/jiec.12398/full https://pdfs.semanticscholar.org/c342/8df4063f6bb76b6f4f9f2e60111a89bf1d7c.pdf https://www.diva-portal. org/smash/ get/diva2:789778/FULLTEXT01.pdf http://www.sciencedirect.com/science/article/pii/S0301421508003613 http://www.worldcat.org/title/encyclopedia-of-energy/oclc/180851239 http://onlinelibrary.wiley.com/doi/10.1162/jiec.2007.1110/abstract http://onlinelibrary.wiley.com/doi/10.1111/j.1530-9290.2011.00450.x/abstract https://www.aalborg.dk/media/6779880/teknologikatalog.pdf https://www.aalborg.dk/media/6779884/strategifor-fossilfri-varmeproduktion.pdf https://ens.dk/sites/ens.dk/files/Globalcooperation/regulation_and_planning_of_district_heating_in_denmark.pdf https://aalborgforsyning.dk/media/536404/groent-regnskab-2015.pdf https://www.linkedin.com/pulse/omstilling-af-varmeproduktionaalborg-varme-2017-anna-bobach?articeId=6281083727846219776 https://www.aalborgforsyning.dk/media/1810/rapport-uk.pdf 56 International Journal of Sustainable Energy Planning and Management Vol. 14 2017 The effect of price regulation on the performances of industrial symbiosis: a case study on district heating [21] Østergaard P, Mathiesen B, Möller B, Lund H. A renewable energy scenario for Aalborg Municipality based on low- temperature geothermal heat, wind power and biomass. Energy 2010;35:4892-4901. https://doi.org/10.1016/j.energy. 2010. 08.041 [22] Regeringen. Energistrategi 2050 – fra kul, olie og gas til grøn energi. 2011. http://www.stm.dk/multimedia/Energistrategi_ 2050.pdf [23] Ivner J, Broberg Viklund S. Effect of the use of industrial excess heat in district heating on greenhouse gas emissions: A systems perspective. Resources, Conservation and Recycling 2015;100:81–7. doi:10.1016/j.resconrec.2015.04.010. https:// pdfs.semanticscholar.org/5b22/9f0cc5d0804cecd08083cfb4992 c68c89328.pdf [24] Broberg S, Backlund S, Karlsson M, Thollander P. Industrial excess heat deliveries to Swedish district heating networks: Drop it like it’s hot. Energy Policy 2012;51:332–9. doi:10.1016/j.enpol.2012.08.031. http://www.diva-portal.org/ smash/get/diva2:578721/FULLTEXT01.pdf [25] Skat. E.A.4.4.10.2 Fordeling af brændsler mellem el- og varmeproduktion i kraftvarmeværker n.d. http://www. skat.dk/skat.aspx?oId=2061647&chk=212484. http://www. skat.dk/skat.aspx?oId=2061647&chk=212484 [26] Kragh J, Wittchen KB. Development of two Danish building typologies for residential buildings. Energy and Buildings 2014;68:79-86. doi:10.1016/j.enbuild. 2013.04.028. http://www.sciencedirect.com/science/article/ pii/S037877881300604X [27] Danmarks Statistik BOL 105. No Title n.d. http://www. statistikbanken.dk/BOL105. http://www.statistikbanken.dk/ BOL105 [28] SBi – Danish Building Research Institute at Aalborg University. Residential Building Typology Building Type Matrix n.d. http://episcope.eu/building- typology/country/dk/. http://episcope.eu/building-typology/ country/dk/ [29] Face-to-face interview with Susanne Majlund Sørensen and Henrik Skov, Reno Nord A/S 2017. [30] Face-to-face interview with Anita Rosenkilde Lodberg, Aalborg Kommune 2017. [31] Face-to-face interview with Anna Bobach and Agne Vaicaityte, Aalborg Varme A/S 2017. [32] Maxwell JA. Methods: what will you actually do? Qualitative research design An ineractive approach. Second Edi, Sage Publications, Inc.; 2005, p. 79-105. https://www. barnesandnoble.com/w/qualitative-research-design-joseph-a- maxwell/1100615037 [33] Danish Energy Agency. Technology data for energy plants. 2012. https://ens.dk/sites/ens.dk/files/Analyser/technologydata_ for_energy_plants_-_may_2012._updated_2015.pdf [34] Molio – Byggeriets Videnscenter. Molio Prisdata n.d. https://molio.dk/molio-prisdata/prisdata-footer/brug-molio- prisdata/. [35] Trafik- BB. Bygningsreglement 2015 n.d. http:// bygningsreglementet.dk/forside/0/2. [36] Muñoz I. Marginal Electricity in Denmark 2015. https:// consequential-lca.org/clca/marginal-suppliers/the-special-case- of-electricity/example-marginal-electricity-in-denmark/ (accessed August 3, 2017) [37] Aalborg Varme A/S. Årsrapport 2015. Aalborg: 2 0 1 5 . h t t p s : / / a a l b o r g f o r s y n i n g . d k / m e d i a / 5 3 6 4 0 0 / underskrevet_aarsrapport_2015_varme_dirigent-aalborg- varme.pdf [38] Aalborg Varme A/S. Årsrapport 2016. Aalborg: Aalborg Varme A/S; 2016. http://www.aalborgforsyning.dk/media/769261/ aalborg-varme.pdf [39] Wiedmann T, Minx J. Ecological Economics Research Trends Chapter 1: A Definition of ‘ Carbon Footprint. vol. 1. 2007. doi:10.1088/978-0-750-31040-6. http:// citeseerx.ist.psu.edu/messages/downloadsexceeded.html [40] Guo M, Murphy RJ. LCA data quality: Sensitivity and uncertainty analysis. Science of the Total Environment 2012;435-436:230–43. doi:10.1016/j.scitotenv.2012.07.006. http://www.sciencedirect.com/science/article/pii/S0048969712 009503 [41] Energitilsynet. Varmeprisstatistik. Udvidet Prisstatistik Pr 1 December 2016 2016. http://energitilsynet.dk/varme/ statistik/prisstatistik/. [42] Pavel GL, Budu AR, Moraru DE. Optimization of Energy Mix – Nuclear Power and Renewable Energy for Low Emissions Energy Source a Benefit for Generations to Come. Energy Procedia 2017;112:412–7. doi:10.1016/j.egypro.2017.03.1092. http://www.sciencedirect.com/science/article/pii/S1876610217 312171 https://doi.org/10.1016/j.energy.2010.08.041 http://www.stm.dk/multimedia/Energistrategi_2050.pdf https://pdfs.semanticscholar.org/5b22/9f0cc5d0804cecd08083cfb4992c68c89328.pdf http://www.diva-portal.org/smash/get/diva2:578721/FULLTEXT01.pdf http://www.skat.dk/skat.aspx?oId=2061647&chk=212484 http://www.sciencedirect.com/science/article/pii/S037877881300604X http://www.statistikbanken.dk/BOL105 http://episcope.eu/building-typology/country/dk/ https://www.barnesandnoble.com/w/qualitative-research-design-joseph-amaxwell/1100615037 https://ens.dk/sites/ens.dk/files/Analyser/technologydata_for_energy_plants_-_may_2012._updated_2015.pdf https://molio.dk/molio-prisdata/prisdata-footer/brug-molioprisdata/. http://bygningsreglementet.dk/forside/0/2. https://consequential-lca.org/clca/marginal-suppliers/the-special-caseof-electricity/example-marginal-electricity-in-denmark/ https ://aalborgforsyning.dk/media/536400/underskrevet_aarsrapport_2015_varme_dirigent-aalborgvarme. http://www.aalborgforsyning.dk/media/769261/aalborg-varme. http://citeseerx.ist.psu.edu/messages/downloadsexceeded.html http://www.sciencedirect.com/science/article/pii/S0048969712009503 http://energitilsynet.dk/varme/statistik/prisstatistik/. http://www.sciencedirect.com/science/article/pii/S1876610217312171