The Transition of the Residential Heat Market in Germany -
A System Dynamics A pproach
Susanne Schmidt, Tobias J ager, Ute Karl
European Institute for Energy Research (EIFER)
Emmy-Noether-Str. 11
D-76131 Karlsruhe, Germany
Tel/Fax +49 721 6105 1375 / +49 721 6105 1332
Email address: susanne.schmidt@ eifer.org
Abstract
This paper presents a System Dynamics model for the study of the residential heat market in
Germany with regard to the European and national energy targets for the year 2020. It de-
scribes the model properties and specifies the stock-and-flow structures of the demand side
based on housing units and of the supply side which is formed by heating systems. An initial
model validation indicates the appropriateness of the model assumptions. Five policy scenarios
are introduced which take into account different measures for the promotion of renewable and
innovative heat generation technologies and obligations for energy-efficient renovation of
buildings. The discussion of the scenarios shows that with the given set of policies, the EU tar-
gets for heat demand reduction and CO emission mitigation in the residential sector would not
be met, while the envisaged share of renewable and innovative technologies seems to be
achievable.
Keywords
Energy policy analysis, residential heat market, System Dynamics
1 Introduction
In September 2010, the German Federal Government adopted the "Energy Concept for an
Environmentally Sound, Reliable and Affordable Energy Supply" (BMWi, 2010) defining the
future German energy system until the year 2050. This concept addresses nine fields of ac-
tion, among those the fortified use of renewable energy sources, energy upgrading of existing
and high energy efficiency standards of new buildings and enhanced research towards inno-
vative and new technologies. Nuclear power had been regarded as a bridging technology
within the concept. After the Fukushima incident, the nuclear phase-out was decided upon
and the "Energy Concept" was supplemented by a document announcing the "transformation
of the energy system" (known in German as the 'Energiewende', BMU, 2011). This program
is supposed to speed up the process started by the "Energy Concept" in order to compensate
the nuclear phase-out. Besides the announcement of a rapid expansion of the use of renewa-
ble energy for both electricity and heat generation, a more effective deployment of the funds
for cogeneration and a tightening of efficiency standards and more funding for energy-related
modemization of buildings are proclaimed. The milestones for the transition in households
are the reduction of greenhouse gas emission of 40 % in the year 2020 compared to the year
1990, the decrease of primary energy consumption of 20 % compared to 2008, a share of
18% renewable energies on the total final energy consumption in the year 2020 and the
doubling of the rate of energy-efficient renovation from 1 % in 2008 to 2 % in the year 2020.
With these targets, the German energy policy is in line with or even exceeds the European
Union's climate and energy targets of 20 % CO» emission reduction, 20 % energy efficiency
increase and 20 % share for renewable energies (EC, 2010).
In order to provide an economic and regulatory framework for Germany's "Energiewende",
environmental policy instruments for renewable energies (EEG, 2000, EEWarmeG 2009 and
MAP 2000) and cogeneration (KWKG, 2002) have been set up or amended. The increase of
energy-efficient renovations is supported by low-interest loans (KfW, 2012), and energy effi-
ciency standards both for existing buildings and for new constructions have been introduced
(EnEV, 2009).
The aim of this paper is to present a System Dynamics model for the study of the residential
heat! market in Germany with regard to the European and national climate and energy targets
for the year 2020. The model allows for the analysis of economic frame conditions and envi-
ronmental policies related to heat demand, heat production technology mix, heat prices and
CO, emissions of private household heating systems. The model simulates the effects of the
different policy instruments for the promotion of energy efficiency and renewable and coge-
neration (combined heat and power, CHP) heat technologies in the heat market for residential
‘In this paper, the term "residential heat" is used for space heating; hot water supply is not considered.
buildings in Germany. Recommendations for action and decision support can be derived,
when considering the impacts of different economic framework conditions such as fuel prices
(oil, gas and biomass) or different environmental policy instruments.
In the first part of the paper, a System Dynamics model for the residential heat market will be
described. In particular the properties of the model will be discussed and there will be a short
description of the model parts. The model will be tested for structure and behavior validity. In
the second part, results of the model will be presented showing various policy scenarios and
their impacts for achieving the EU and national energy and climate change mitigation targets,
which are further broken down into figures for the residential heat market in Germany.
2 Literature review
There are several options for model-based energy systems analysis of heat markets. With
respect to a bottom-up modeling approach energy system models can be classified as simula-
tion models or optimization models. Examples for simulation models using the System Dy-
namics methodology in the energy context can be found in Dyner et al. (1995) and Gaidosch
(2008). Weidlich and Veit (2008) and Genoese (2010) present the application of agent-based
modeling to energy systems as further simulation model type. Optimization energy system
models are described by Ravn (2001) and Fichtner et al. (2004). More examples for energy
system models, especially with the focus on Germany, can be found in TUB (2003) and IER
(2005).
There are a significant number of examinations of energy systems applying the System Dy-
namics method. Sterman (1983), Bassi (2006) and Akbarpour and Vaziri (2007) investigate
national energy and resources markets from a macroeconomic perspective. Energy policy
analysis on national level can be found in Naill and Belanger (1992), Dyner (1996), Bunn et
al. (1997) and Musango et al. (2009). Electricity market liberalization and deregulation
processes have been the focus of many System Dynamics models. An overview on the fun-
damental work in this field can be found in Ford (1997). Later studies, especially for the Eu-
ropean context, have been presented by Vogstad (2004), Pruyt (2007), Gaidosch (2008) and
Ochoa and van Ackere (2009).
Regarding energy efficiency policies in the residential sector, Dyner et al. (1995) used Sys-
tem Dynamics for the study of the energy saving potential from household appliances.
Groesser et al. (2006) examined the “diffusion dynamics of energy-efficient innovations in
the residential building environment" with focus on the interactions of physical buildings,
owners and architects. Besides developers and consumers acting on the market, Li and Dai
(2008) also include the government-issued policies for the analysis of an "energy efficient
residence market". Miller and Ulli-Beer (2010) study different energy-efficient renovation
strategies and relate them to CO2 emissions as an indicator for the policy effectiveness. The
model presented by Blumberga et al. (2011) aims at evaluating the EU energy efficiency tar-
get achievement of the housing stock in Latvia. Y ticel and Pruyt (2011) discuss the impor-
tance of policies for the existing building stock in the context of energy efficiency targets. In
summary, the listed studies for energy efficiency in buildings all focus on policies for the
demand side, while policies and technology diffusion on the supply side are not considered or
exogenous to the applied models.
3 A System Dynamics model of the residential heat market in Germany
The following table summarizes specific model properties of the heat market model.
Table 1: Summary of the properties of the heat market model
Model PROPERTIES Heat market MODEL
Model type Dynamic simulation (System Dynamics), myopic
One economic sector model Residential heating sector
Model approach Descriptive, bottom-up, supply side oriented
Multi-periodic/ time horizon In annual steps, 2005 - 2025
Geographic scale National, Germany
Description of technologies on the level of energy carriers and
transforming technologies in three size classes
Description of buildings by type, heated floor space and
energy consumption coefficients
Buildings:
Energy-efficient renovation quota,
required energy coefficient new construction /
energy-efficient renovation
r _— Technologies:
Environment related policy instruments renewable / innovation share obligation,
renewable energy feed-in tariff,
CHP bonus,
investment subsidies,
fuel tax
Investment decision: distinction of interest rates for two types
of owners - home owners and landlords
Supply side
Demand side
Behavior of market actors
The model type presented here can be categorized as a dynamic simulation (System Dynam-
ics) with a myopic perspective. This perspective allows for the study of the behavior of im-
perfect systems over time like energy markets without the assumption of perfect foresight or
the need for optimal results. Delays in technology implementation can be taken into account
and policy impacts can be analyzed in particular. Figure 1 provides an overview on the main
areas of study and their interactions.
Supply Side |
Environmental
Policy Instruments
Economic
Assessment
Heating Systems &
Capacity
Housing Units Heat Demand
CO, Emissions &
Primary Energy
Consumption
Figure 1: Scheme of the main model interactions
Environmental policy instruments influence the effects and the effectiveness of the evolution
of the heat demand side as well as the heat supply side. The long-term development of the
residential heat market can be evaluated using scenario analysis with comparison to the EU
three time 20 % targets. Table 2 summarizes the most important input and output parameters.
Table 2: Main input und output parameters
INPUT OUTPUT
Supply side
Number of installed heating systems
Initial average size of heating systems
Lifetime of heating systems
Techno-economic parameters of heating
systems
Interest rate by type of ownership
Initial number of housing units
Lifetime of buildings
Initial energy coefficients
Environmental policy instruments: Energy-
efficient renovation quota, required energy
coefficient new construction / energy-
efficient renovation per building type
Installed capacities for heat and CHP tech-
e Fuel price development nologies
¢ — Environmental policy instruments: renewa- ¢ Heat demand in residential buildings
ble / innovation production share obligation, e Energy coefficients
feed-in tariff, CHP bonus, investment sub- e Competitive heat prices per technology size
sidy, fuel tax e Primary energy consumption per fuel
Demand side e CO; emissions
e Heated floor space
3.1 Scope of the model
The model concentrates on the economic sector of residential heat generation and does not
consider the heat markets of other economic sectors (like industry or tertiary) or other energy
sectors like electricity or resource markets. The model can currently be used for the exclusive
analysis of the residential heat sector without analyzing macroeconomic impacts on other
sectors. However, there already exist links to the electricity market with the integration of
combined heat and power generating technologies. At this stage, a competitive heat price is
being derived from the residential heat market model. This output can be potentially fed into
an existing electricity market model (Jager et al., 2009) to study possible impacts of increas-
ing shares of CHP technologies in the residential sector on the electricity market in Germany.
The model shows a time horizon until the year 2025. The target values of energy policies
refer to the year 2020. Nevertheless, possible impacts of those policies can be observed
beyond this date. The residential heat market model presented here is able to display annual
changes of the output parameters over the entire period from 2005 to 2025. Anticipating a
future combination of the described heat market model and the electricity market model
(Jager et al., 2009) the time resolution for both models is set to daily values. Thus, it will be
possible to integrate intrayear heat demand profiles into the combined model.
The model described here considers the residential heat market in Germany. Building types
and their heated floor space as well as the heating systems are aggregated to the national level
as well as the decision parameters for new investments. The model also integrates an exogen-
ously defined percentage of district heating in the heat supply mix.
3.2 Heating technologies
The model can be characterized as a bottom-up techno-economic model. The parameters of
heating and CHP systems are described for the supply side of the heat market on an aggre-
gated level of technology classes distinguishing energy carriers and/or transformation tech-
nologies (Represented technologies are: Solid fuels, oil conventional, oil innovative, gas con-
ventional, gas innovative, gas CHP, pellets, wood logs, electric conventional, electric heat
pump, biomass CHP, hybrid 1: combination gas innovative and solar thermal, hybrid 2: com-
bination gas CHP and solar thermal, hybrid 3: combination pellets and solar thermal and hy-
brid 4: combination heat pump and photovoltaic). Further, the technologies can be divided
into three size classes: small, medium and big. Each technology differs in terms of invest-
ment, operating costs, CO2 emissions and its status of technological progress.
3.3 Heat demand from residential buildings
Although the presented model is focusing on the evolution of heating and CHP technologies
on the German heat market, the demand side is also modeled with a certain level of detail. It
takes into account different types of residential buildings: Single family houses (EFH), semi-
detached/terraced houses (RDH), multi-family houses (MFH), big multi-family houses
(GMH) and high-rise buildings (HH) (IWU/BEI, 2010). For each type, the heated space is
defined according to Destatis (2010). Depending on its energy efficiency level, each class of
buildings is assigned to a specific energy coefficient (Blesl et al., 2009), resulting in the de-
mand for space heating from the residential heat market. In order to derive the total heat de-
mand of the residential sector, a fixed amount of heat for hot water supply is added taking the
assumption that energy efficiency measures do not affect the individual's behavior concerning
hot water consumption.
3.4 Instruments for environmental policies
Environmental policy instruments in Germany have been analyzed and are integrated for both
the demand and the supply side. For the demand side, it is possible to study the impacts of
different energy-efficient renovation quotas and varying mandatory minimum requirements
conceming the specific energy coefficients of new or renovated buildings. Here, energy-
efficient renovation quota does not describe an existing policy instrument, but is derived from
the targets defined in the "Energy Concept" which includes the increase of the annual energy-
efficient renovation quota from 1 % in 2008 to 2 % in 2020. The policy instrument on specif-
ic energy coefficients of new or renovated building is based on the Energy Saving Ordinance
(EnEV 2009) which defines the technical minimum requirement of new and existing build-
ings and their components with regard to energy consumption.
On the supply side, two policy instruments are linked to the electricity market regulations:
feed-in tariffs are paid for electricity produced in CHP plants which are fed by renewable
energy sources (EEG, 2000) and a CHP bonus is paid for electricity generated in small fossil
fired CHP systems (KWKG, 2002). Further, the model takes into account investment subsi-
dies for pellets heating technologies and heat pumps (BMU, 2005). In 2009, the renewable
energy heat law came into force (EEWarmeG, 2009). House owners installing a new heating
system in their building are obliged to either generate a certain percentage of consumed heat
with regenerative energies or to install innovative energy-efficient technologies like cogene-
ration units or heat pumps. This fact is modeled by means of a renewable / innovation share
obligation. Renewable or innovative technologies include gas-fired CHP units, pellet boilers,
electric heat pumps, biomass-fired CHP units and all hybrid technologies (defined as combi-
nations of before-mentioned technologies). Finally, it is possible to consider increased fuel
taxes on certain fossil fuels such as fuel oil. The German tax system currently does not make
such a differentiation, but this is still under discussion for future energy policies in the resi-
dential sector (BMWi, 2010, 23).
3.5 Investment decisions
In the presented model, investment decisions for new space heating capacities are taken from
an individual investor's point of view. This means that a house owner decides on the basis of
his limited information on future economic and political conditions. He is only choosing a
new heating system when replacement is necessary at the end of the lifetime of existing
equipment. In order to distinguish two different types of ownership (and decision making) of
a building or housing unit, the heat market model applies a higher interest rate for landlords
and a lower interest rate for house owners. The two interest rates illustrate the fact that lan-
dlords are more risk-averse in their investment decision compared to house owners. Thus, the
behavior of the market actors only depends on economic factors. This means a simplification
of the complex decision-making process that underlies investment decisions in the heat mar-
ket.
4 Model components
The main model interactions as well as the major model properties have been explained in the
preceding chapters. In this chapter insight into the structure of the heat market model will be
provided by the description of the model components. Table 3 gives an overview on the ten
fundamental elements of the residential heat market that are taken into account in the pre-
sented model.
Table 3: Main parts of the heat market model
Housing Units
Floor Space
Heat Demand
Heating Systems
Heating Capacities
Resource Efficiency
Economic Assessment
Competitive Heat Price
CO, Emissions
Primary Energy Consumption
The model parts subsume four major aspects of the heat market model: The demand side con-
sists of Housing Units, Floor Space and Heat Demand. Heating Systems, Heating Capacities,
Resource Efficiency Thermal and Resource Efficiency Electrical form the supply side. The
Economic Assessment is a module on its own and Competitive Heat Price, CO2 Emissions
and Primary Energy Consumption provide additional results beyond those that can be derived
by the calculations on the supply and demand side. In the following, the four main model
components will be described.
4.1 Main model interactions
On the demand side, two important feedbacks have been identified. The first loop is describ-
ing the transition of the stock of energy-intensive buildings into a stock of energy-efficient
buildings (see Figure 2). This process is driven by the aging of the energy-intensive housing
stock and the policy instrument "energy-efficient renovation quota". The second loop stands
for the reluctance of house owners to carry out energy-efficient renovation measures. Build-
ings can be renovated without energy-efficiency gains. In this case they enter the stock of
energy-intensive buildings after a certain period of time which is represented by a delay. For
the supply side, a similar structure can be applied. One causal loop contains the transition of
the stock of old fossil-fuelled heating systems into a stock of renewable (RES) or innovative
(inno) heating systems. This loop in tum is decelerated by another loop which describes the
fact that old heating systems can - depending on the profitability of RES/inno compared to
conventional technologies - also be replaced by fossil-fuelled heating systems.
ulation on energy
consumption
coefficients :
\e ‘C02 emission
PX energy
energy-intensive reduction
savings buildings
+ +
+
4 Rene BE renovation
renovation ) S
energy-efficient Ron-energy-efficient +
buildings renovation
ge aes
share of RES inno © heatngs
oct predoctin
+
energy-efficient
renovation
—
th) 'sS amt
OFS eatings pe
Cae
RES/CHP
RES/inno fossil-fuelled new poh
heatings investment
v 4) ‘Transition towards }
) Resin beating stock picnestany ie
+ RES/inno heating
iniaciyahan nesinno new Ay s
RES/inno investment “t
—
energy-efficient
renovation quota
+ investment
subsidies
+ size reduction.
heatings:
Figure 2: Causal loop diagram of the System Dynamics model
Figure 2 futher illustrates the influence of the demand side on the supply side. Here, energy-
efficient renovations increase the potential for replacement of heating systems based on the
assumption that during the energy-efficient renovation, the heating system will also be
replaced. Due to energy-efficient renovations, the specific size of a new heating system to be
installed in a building decreases. In tum, investment costs are lower and profitability
increases. This effect is stronger for RES/inno technologies which are characterised by higher
investment costs and lower operation costs compared to conventional fossil-fuelled
technologies. Finally, the causal loop diagram highlights the influence of the environmental
policy instruments on the model structure and the model variables representing the results.
4.2 Demand side structure
For the modeling of the demand side of the German heat market, the first three model com-
ponents presented in Table 3 have to be combined. One reason for this procedure is the fact
that the two policies included for the demand side affect different parts of the demand side:
energy-efficient renovation quotas determine flows in the system of housing units, while ob-
ligatory energy coefficients for new constructions and renovation can be directly linked to the
part of heat demand. Further, energy coefficients are defined specifically relating heat de-
mand to the heated floor space. Hence, the demand side can be implemented by means of a
typical ageing chain structure (Forrester, 1969) of the housing units and two co-flow struc-
tures (Sterman, 2000, 497pp.) for the heated floor space and for the heat demand in the re-
spective housing units (see Figure 3).
renovation rate retention period
housing unis none housing units poor
Eneray-Efficient renovatin rate {———_\
Renovation Quota nousing units EE senaton
ZZ — St )
Housing Units Housing Units Housing Units
construcionrate | New Condiion |—gecay fltenew ® | Good Condition | decay rate good *”] P00" Condlion | Gemciion a
shew Rousing units housing units nowsing units 7 poorhousing unis
\
new construcifon { / | \
ue
}
\ Naan / wit prio 7 \ \ \
\ vommmedatecy «| gctatengste
i} \ \
weet) le Vee
‘construction rate | New Condition |~“Gecay ratenew | S04 Condition |” decay rate good — =
sermooE ee Rg MR 7 floor space
| { Tenovaiin rae
\ floor spaceEe
|
\ renovation rate ———|—
| sor space none E === XY
} ney sceagy conten
| cosmclantn sve) cones,
Energy Coefficient } (- interm buildings, ergy coemicis
Eoray Coeticient C \ i aitaings
Heat Demand Heat Demand ig HeatDemand’ |_| A
2
~Tnerease heat Newey Tncreaseheat |__COodHU increase in heat Pace ‘decrease in heat
demand new HU demand good HU demand poor HU ToMemand old buildings
Eneray Coefficient TEE> ——~p increase in heat incretaeta tigate
EE-renovated demand from EE (demand from nond@ 2
uilding renovation renovation
ae et
___, reduction factor sent
*heatdemand
Figure 3: Structure of the demand side in the heat market model
The ageing chain of housing units in the residential sector is driven by three exogenous pa-
rameters. A new construction quota determines the number of housing units that are added to
the system each year, increasing the stock of housing units in new condition. Depending on
the period housing units are typically expected to stay in new and in good condition, respec-
tively, the ageing process is simulated. Housing units that are in poor condition can either be
demolished according to the exogenous demolition quota, or they can be renovated with or
without energy efficiency measures. The share of non-energy-efficient renovation is derived
from the policy driven share of energy-efficient renovation. In parallel, the ageing of the
heated floor space can be modeled, using the specific values resulting from the co-flow struc-
ture. In order to keep the model as flexible as possible for further policy studies, the exogen-
ous factor of specific floor space of newly constructed housing units is modeled explicitly.
This can have a significant impact on the total heat demand. Finally, the co-flow of the resi-
dential heat demand is influenced by the given policies of energy coefficients required for
new constructed buildings and energy-efficient renovated buildings. From the heat demand
and the floor space co-flows, specific energy coefficients are calculated and a reduction factor
for the specific heat demand can be derived. This percentage reduction serves as input for the
supply side as explained in the following section.
4.3 Supply side structure
The use of ageing structures and co-flows proved to be expedient for the implementation of
the demand side of the German residential heat market. For the supply side, a similar ap-
proach has been chosen. Based on the ageing chain of the heating systems, co-flows for in-
stalled heating capacity, thermal resource efficiency and electrical efficiency for cogeneration
are defined. Figure 4 shows the structure of the described model parts.
Drivers of the dynamics of the heating systems are on the one hand the installation of new
heating systems and the decommissioning of heating systems, which are directly linked to the
construction of new housing units and the demolition of housing units in poor condition. On
the other hand, new heating systems replace old heating systems reaching the end of their
lifetime. Depending on the economic assessment of the technological possibilities for new
installation (cf. Chapter 4.3), on the accessibility of a gas grid and on the obligatory share of
renewable or innovative technologies, the technology mix for new heating systems is derived.
It is assumed that this mix also reflects the heating systems installed in new constructions.
The heating capacity co-flow is calculated from the reduction of the specific heat demand
(cf. Figure 3). Based on this reduction the installed capacity of new heating units replacing
old ones can be derived. The newly installed heating units are allocated to the group with
reduced capacity according to the rate of housing units with energy-efficient renovation.
<percentage
distictheating> new -<demolition rate
oor housing units>
’
replacement
Psst me
ea system in 2 \ ss
new housing unis ~~ } =—% heating systems
a a _ oN erste
/- New Heating gmat) / | ouneatng A \
/ ‘canatrucbon new
Heating atine \
Systems [ageing rate new” | Systems | ageing rate interm |_9¥St replacementold \
heating systems
\ replacement of heating
heating system heating systems T heating systems
\ systems without
\ reduction in capacity
choice of best
RES technology
~Faverage size | replacement of heating
old | system with reduction
[in capacity
4
‘technology mix new
\ eehaereene
share of housing _\~
unit without gas grid
Renewable / Innovative |
Technology Share <> New Heating Old Heating
‘Obligation investment new| Capacity | ageingratenew | Capacity | ageing rate interm | Capacity
heating capac heating capacity ‘eating capacity
\ \ \ \
| resource \
je reoares ‘esoure | owe, |
efficiency new ni | omer
| ceptacement of he | | {
without reduction / V. \I | \
{ rea Resources Resources
tkup resource efficiency ¢——st 3 | £ ou
new construction increase Capacities increase ¥—. capscttes J increase Capacities decrease
ns ences panto inter cap resources old cap sources old cap
Figure 4: Structure of the supply side in the heat market model
The remaining heating units stay constant in their installed capacity. Finally, technological
progress is influencing the development of the resource efficiencies of the heating and CHP
technologies in the model. By means of an exogenous function, the heating system's resource
efficiencies are increasing over time. This causes the dynamics of the resource requirements
co-flow resulting in a lower specific consumption of primary energy sources.
4.4 Economic Assessment
In the previous chapters, the dynamics of the demand and the supply side of the heat market
model for Germany have been described. It has already been mentioned that the technological
possibilities for new heating systems depend on an economic assessment. Moreover, four of
the environmental policy instruments that will be studied in this paper - investment subsidy,
feed-in tariffs for renewable energies, CHP bonus and fuel tax - have an impact through de-
creasing costs of the technologies they are designed for.
Each time a heating system has to be replaced because of age or as a measure of energy-
efficient renovation, the investor has to decide which new technology should be installed. In
the presented model it is assumed that investment decision is based on an economic ap-
proach. The annuity A of each technology m can be calculated according to the following
equation:
Equation 1: Annuity
(4tiymei
G+p"=1
+ Chy,
Am = —Im
where J, is the specific investment costs for a new heating capacity including investment
subsidies, CF, is the annual net cash flow, i is the interest rate and n is the lifetime of the
heating system. CF, can further be described by:
Equation 2: Annual net cash flow
CF, = Cyixm + (Coarm + Cruetm — Reecm — Reupm) * FLHm,
where C;;, are fixed costs of operation and maintenance, C,,, are variable costs of operation
and maintenance, Regg are revenues resulting from feed-in of electricity from biomass CHP
technologies, Rcyp are revenues from sales of electricity according to the CHP law and FLH
are the typical annual full load hours of operation of each size class. In the model, it is as-
sumed that technologies of the small size class are run in one family and semi-
detached/terraced houses, technologies of the medium size are run in multi-family houses and
technologies of the big size are run in big multi-family houses and high-rise buildings. After
the calculation of the technology-specific annuities, it is possible to create a merit order of
heating technologies by increasing annuities. That means that an investor will choose the
technology with the lowest (negative) annuity in order to minimize the total heating costs.
Since the model also takes into account a share of housing units that is not connected to a gas
grid and the obligatory share of renewable or innovative technologies, three different merit
orders are constructed: one including all technologies, one including all technologies except
gas-fired ones and one including only renewable or innovative technologies (cf. Chapter 3.2).
The first technology of each merit order finally serves as input for the investment into new
heating capacity on the supply side.
4.5 Additional results
Beyond the model outputs concerning heat demand, energy coefficients and installed heating
capacity that can be derived from the model parts of demand and supply, three values are of
special interest for the residential heat market in Germany: competitive heat price, CO2 emis-
sions and primary energy consumption. The competitive heat price is defined by the heat
production costs of the most competitive heat technology (cp. Woldt et al., 2007) and can be
used for the economic assessment of CHP technologies on the electricity market. With the
perspective of the combination of the herein described heat market model and an electricity
market model, the competitive heat prices for each technology size are explicitly calculated.
With regard to the energy targets to be studied by means of the presented model, the devel-
opment of the CO2 emissions are calculated as a further model output. The simulation of the
heat demand development for space heating has already been explained in section 4.1. The
total heat demand is derived by adding a fixed percentage of heat demand for hot water. In
order to calculate the technology specific heat production, a percentage technology mix is
derived from the ratio of the installed capacity of each technology related to the total installed
capacity. The resulting values are multiplied by the total heat demand - heat from district
heating is supposed to remain constant and is excluded from this calculation - and the tech-
nology-specific CO2 emission coefficients, resulting in the total CO2 emissions from the resi-
dential heat market. Finally, the primary energy consumption is calculated using the total heat
demand and the percentage technology mix as described above, multiplied with the thermal
efficiency of each system.
5 Initial model validation
According to Forrester and Senge (1980) and Sterman (2000), the validation process should
establish confidence in the appropriateness of a model for a certain purpose. For this process,
they propose several tests which can be divided in two main parts: structure testing and beha-
vior testing. In this paper, a structure assessment test and a parameter assessment test as
structural tests and a basic behavior reproduction test as behavior test are presented.
5.1 Structure tests
A structure verification test of a System Dynamics model should examine whether the "mod-
el structure is consistent with relevant descriptive knowledge of the system" (Sterman, 2000,
859). The main model structures both of the demand side and the supply side are derived
from classical ageing chains and co-flow structures which represent the fact that replacement
of technologies and renovation of housing units is mainly driven by ageing processes (cp.
Vogstad, 2004, 102 and Miiller and Ulli-Beer, 2010, 16). Thus, it can be assumed that the
model represents the existing knowledge of the system. By initializing the stocks with statis-
tical data, the parameter verification has been realized. The tested parameters and their
sources are summarized in Table 4.
Table 4: Parameter verification of the heat market model
Parameter Source
Initial housing units by building type and vintage IWU 2007 (2005 data)
Initial floor space per building type and vintage IWU 2007 (2005 data)
Initial energy coefficients per building type and vintage | Blesl et al., 2009, p. 33 (2004 data)
Initial heating systems per technology, size and vintage ZIV, 2010 (2010 data)
Initial average installed capacity per technology and size ZIV, 2010 (2010 data)
The source for heating systems refers to the situation in the year 2010. Since the model simu-
lation starts in the year 2005, the data had to be adapted taking into account the average life-
time of one system and the average age of all systems of each technology.
5.2 Behavior tests
In order to test a model for its appropriate behavior, statistical measures of correspondence
between model simulation results and observed data can be applied (Sterman, 2000, 860).
This procedure requires adequate availability of historical data. In the case of the residential
heat market for Germany, it tumed out to be difficult to find exhaustive historical data for
most of the model variables. The Federal Ministry of Economics and Technology regularly
publishes data on space heating demand in Germany per economic sector (BMWi, 2011). na
first step, the simulation results are plotted against the historical data for the years 2005 to
2010 as a visual comparison (Figure 5).
600
oe
im)
ie
500
DB
a
ou
400
TWh
w
S
6
t) r T T T T T 1
2004 2005 2006 2007 2008 2009 2010 2011
@ Observed data Simulation results
Figure 5: Comparison of observed and simulated values for space heating demand from private households in
Germany in the years 2005 to 2010
The simulation results and the historical data have further been used to calculate statistical
measures (Andres and Spiwoks, 2000) presented in Table 5.
Table 5: Statistical measures of space heating demand in Germany as behavior reproduction test
Statistical measure Value
Mean Absolute Error 34.27 TWh
Mean Percentage Error 2.86 %
Mean Absolute Percentage Error 6.87 %
Mean Square Error 1801.86 (TWh)?
Root Mean Square Error 42.44TWh
Percentage Root Mean Square Error 1.65 %
These tests give a first estimation of the model's appropriateness to reproduce developments
on the demand side of the heat market in Germany. In a next step, they will be extended to
the supply side.
6 Scenario definition and analyses
The main output parameters of the simulation model for the heat market are “Heat Demand”,
“Share of Renewable and Innovative Heat Production” as well as “CO, Emissions”. The resi-
dential energy demand for space heating and hot water in Germany accounted for 533 TWh
in 2005 and for 535 TWh in 2010 (BMWi, 2011). According to BEE (2010), the share of
renewable heat production amounted to 6.8 % in the German residential heat market in the
year 2005 and 11.4 % in the year 2010. The households emitted 111 Mt of CO2 for heating
purposes in 2005 and 112 Mt of CO, in 2010 (BMWi, 2011).
As shown in Table 6, estimations for the developments of the input parameters cover the time
period from the year 2008 to 2020. The economic frame conditions are reflected in the fuel
prices. Fuel prices are driven by global demand. This demand of commodities depends on
global economic growth. Thereby it is assumed that the German economy goes along with
the development of the world economy and causes respective effects for the fuel price devel-
opments in Germany. The same coherence with the growth of the German economy is as-
sumed with the development of the construction of new buildings in the residential sector.
Again it is assumed that for e.g. many new buildings will be erected in Germany if the coun-
try has large economic growth rates. The selected main policy instruments in Germany pro-
moting the heat supply technologies are “feed-in tariffs for electricity from renewable ener-
gies”, “cogeneration bonus” and “investment subsidies for some distributed generation tech-
nologies”: For the heat demand of the residential sector the policies "energy-efficient renova-
tion quota”, “energy coefficient for new buildings” and “energy coefficient for energy-
efficient renovated buildings” are chosen. All input parameters are given as total develop-
ment over the considered time frame, except the energy-efficient renovation quota. This is
given in the usual formulation per year.
For the evaluation of the policies, the three 20 % goals set by the EU are the benchmark.
These European goals need to be allocated to national respective sector-specific goals. Refer-
ring to the leading scenario from DLR for the German government (DLR, 2009), a target for
the share of renewable energies with regard to final energy demand of 18 % for the year 2020
is determined. The government decided based on the same study for energy efficiency in the
year 2020 a 20 % reduction of primary energy compared to 2008 and also a 20 % cut refer-
ring to 2008 until the year 2020 for the heat demand of buildings. Germany faces EU re-
quirements with a 20 % goal and has set a more ambitious own target of 40 % reduction of
CO2 emissions compared to 1990. Buildings should contribute with a share of the reduction
of CO emissions of 46 %. Regarding the time period from 2008 to 2020, CO2 emission sav-
ings from the residential sector should account for 34.5 % (BMWi, 2011).
Five scenarios are distinguished. A “Reference" scenario is characterized by an extrapolation
of the current trend of the parameters, an “Ecology" scenario emphasizes an environment
driven policy on the supply and demand side of the economy. The Scenario 3 ("Demand")
and the Scenario 4 ("Technology") should represent effects on the heat market of exclusive
policies for the demand side or the heat supply technologies. Scenario 5 describes a “Growth"
scenario and analyses the effects of high economic growth and high fuel prices in addition to
strong environment orientated policies for the heat sector.
The “Reference” scenario (Scenario 1) expresses a development based on the continuation of
the current trend of the values. It assumes a moderate economic growth, which means an av-
erage growth of the fuel prices as well. The decline of the values for feed-in tariffs and coge-
neration bonus follow the trend until the year 2015 as indicated by the law and beyond. In-
vestment subsidies remain on the current level. Regarding energy-efficient renovation quotas,
in this scenario it is assumed that the government is not able to attract owners of all types of
buildings in the residential sector to invest in energy-efficient renovation beyond the current
level. Energy coefficients for new and renovated buildings decline as it is foreseen in the re-
spective laws and continue with the same trend until the year 2020. The second scenario
(Scenario 2) can be characterized as an "Ecology" scenario, where the environmental policy
will be more successfully implemented. Although the economy can only achieve a moderate
growth, the government decides to follow a “green” development path. Due to high demand
of subsidized distributed generation technologies and renewable energies, pellet and wood
log prices as well as electricity prices rise stronger than in the "Reference" scenario. With
respect to the moderate growth situation the quota for energy-efficient renovation are raised
only for single family buildings to avoid further burdens for investors and safe the growth
path.
The third (Scenario 3: "Demand") and fourth scenarios (Scenario 4: "Technology") highlight
the supply or demand side of the heat market. The "Demand" scenario focuses exclusively on
the bundle of energy efficiency measures lowering heat demand to demonstrate the effect of
this bunch with regard to the EU goals. Scenario 3 assumes a low economic growth and only
measures on the demand side are implemented. The “Technology” scenario (Scenario 4)
deals exclusively with heat supply technology emphasizing policy instruments and is faced
with the same weak economic situation. Therefore no measures on the demand side beyond
the currently existing will be implemented, but the few financial means are put into the pro-
motion of decentralized technologies and technologies based on renewable energies. Finally
Scenario 5 ("Growth") contains a development path, where strong economic growth has the
main focus and due to this success financial policies for heat demand and supply are afforda-
ble and can be seen as a further stimulus for investments. Although high price increases for
fossil fuels can be regarded as an economic burden, price increases of pellets, wood logs and
electricity should reflect strong investment activities in renewable energies for heat and elec-
tricity. In addition due to the well endowed financial situation policies for the promotion of
energy efficiency on the demand side of the heat market are implemented.
Table 6: Economic and environmental policy based scenarios in the period 2008 to 2020
: 5 5 Scenario
Scenario | Scenario | Scenario 4 Scena-
Item/ Scenario Unit 1 2 3 Techno rio
Reference | Ecology | Demand 1 Growth
logy
Economic growth moderate | moderate low low high
New Boling quote Ber year | %4 1 1 05 05 15
Price development (change
for the whole period) +30 430 430 +40 +60
Teall +70 +70 +50 +60 +150
Worl anne % +100 +150 +70 +100 +150
ike +50 +70 +30 +50 +70
Electricity +30 +50 4-5 H-5 +50
2008 — 2020
Feed-in tariffs for renewable
Seas furl % H-5 430 -30 430 430
2008 — 2020
Cogeneration bonus (change
for the whole period) % 4-5 +30 -30 +30 41-5
2008 - 2020
Investment subsidies for heat
pumps and pellets heating . .
(change for the whole period) % +5 +30 45 +30 #30
2008 - 2020
Obligatory share of renewa-
ble / innovative new heating % 10 30 10 30 50
systems (in 2020)
Energy-efficient renovation
quota per year
EFH /RDH % i ; ; i 5
MFH / GMH/ HH
2008 - 2020
Required Energy Coefficient
for new buildings % -20 -80 -60 -40 -80
2008 - 2020
Required energy coefficient
es te % 20 -80 -80 40 -60
2008 - 2020
7 Results
The scenario results from the model calculations for the residential heat market of Germany
are presented for the parameters “Heat Demand”, "Share of Renewable and Innovative Heat
Production” and “CO2 Emissions”. According to the research questions the simulations have
been run for a time frame between the years 2005 and 2025. The scenario analyses refer to
the time period between the years 2008 and 2020 which corresponds to the horizon of the
environmental policies.
600
575
550
525
500
2005 2010 2015 2020 2025
Heat Demand : Reference —-s———1—_3—_ Heat Demand : Technology —s+———-s—4—_4
Heat Demand : Ecology ———2———2—_2— Heat Demand : Growth
Heat Demand : Demand
Figure 6: Development of heat demand of the residential sector in Germany
Considering the heat demand of the residential sector of Germany in the period 2008 until
2020, the development of the "Reference" scenario (Scenario 1) remains nearly stable, with
an increase of 0.5%. A decline of heat demand of 2.8 % between the years 2008 and 2020
can be found in the "Technology" scenario (Scenario 4), where stronger promotion of innova-
tive heat supply technologies and those based on renewable energies are implemented along-
side moderate policies on the demand side. Stronger decrease in the development of the heat
demand in the period of 2008 to 2020 can be seen in the "Growth" (3.1 %, Scenario 5) and
the "Ecology" (3.8 %, Scenario 2) scenarios. In both of them, effects of ambitious energy
efficiency measures on the demand side are damped by moderate (Scenario 2) or even high
(Scenario 5) economic growth. The strongest decrease of 7.2 % in heat demand between 2008
and 2020 can be observed in the "Demand" scenario (Scenario 3), where fortified policies on
the demand side are accompanied by low economic growth.
05
04
0.3
0
2005 2010 2015 2020 2025
“Share RES / inno" : Reference. ———z———3—43- “Share RES / inno" : Technology. ——+———s——s—+—
“Share RES / inno" : Ecology. —2———2——22 "Share RES / inno" : Growth
“Share RES / inno" : Demand. ——3——3——3—3—.
Figure 7: Development of heat production share from renewable or innovative technologies in the residential
sector in Germany
The lowest share of heat production from renewable or innovative technologies on the Ger-
man residential heat market for the year 2020 (14.6 %) can be found in the "Reference" sce-
nario (Scenario 1) with limited incentives for the promotion of renewable and innovative heat
production and with moderate growth rates and fuel price rises. A slightly higher value
(15.3 %) is achieved in a situation where demand side policies are preferred to technology
policies (Scenario 3: "Demand"). In the "Ecology" scenario (Scenario 2) which is characte-
rized by moderate economic growth, moderate fuel prices and enhanced policies on both the
demand and the supply side, 20.8 % of the heat production in year 2020 is provided by re-
newable or innovative technologies. A similar ratio (21.0 %) can be observed in the "Tech-
nology" scenario (Scenario 4) where priority is given to policies for the supply side of the
heat market. The highest share in comparison to all scenarios is achieved in the "Growth"
scenario (Scenario 5) with a share of 29.1 % of renewable or innovative heat production in
the year 2020. In this scenario high economic growth, the highest fossil fuel price increases
and strong technology policies promote a high share of renewable and innovative heat pro-
duction.
120M
110M
» 100M
90M
80M
2005 2010 2015 2020 2025
CO2 Emissions : Reference. —1————4——3. CO2 Emissions : Technology +4
CO2 Emissions : Ecology ——2———2———_2——_ C02 Enmissions : Growth
CO2 Emissions : Demand. ———-3———3—_3-—
Figure 8: Development of CO, emissions of the residential sector in Germany
The lowest reductions of CO2 emissions from the residential heat market in Germany are
achieved in the "Reference" scenario (Scenario 1) with 0.5 % between the years 2008 and
2020. The economic growth rate and the fuel prices are assumed as moderate and policy in-
struments for the supply and demand side of the heat market have moderate lowering effect
on COQ? emissions as well. 5.9 % of CO2 emission reduction in the time period between 2008
and 2020 are reached in the "Ecology" scenario (Scenario 2) in a situation with moderate
economic growth and fuel price development and the provision of incentives and environ-
mental constraints on both sides of the heat market. A CO, emission decrease of 9.1 % in the
year 2020 can be obtained in the "Technology" scenario (Scenario 4), where strong financial
incentives for renewable and innovative heat production are provided, but energy efficiency
gains on the demand side are comparatively low. The scenario emphasizing energy saving
activities on the demand side in a situation with low economic growth and fuel price rises
(Scenario 3: "Demand") allows for a CO2 emission reduction of 11.8 % in the year 2020
compared to the year 2008. The highest CO emission reduction can be found in the
"Growth" scenario (Scenario 5) with a decline of 12.8% from the year 2008 to 2020. Al-
though a high economic growth leads to a higher growth rate of new buildings, high fossil
fuel prices and strong policies both on the supply and the demand side enable significant CO2
emission reductions from residential heating.
8 Conclusions and outlook
In this paper, a System Dynamics model of the German residential heat market has been in-
troduced and described by its main parts. Different scenarios for future developments have
been defined taking into account economic growth and environmental policy instruments for
the supply and the demand side of the market and the obtained results have been presented.
Regarding the EU three times 20 % targets for the period from 2008 to 2020, the simulation
results can be summarized:
Economic growth has a strong impact on the development of the heat demand. Low econom-
ic growth together with strong measures on the demand side of the heat market will result in
high gains in energy efficiency. However, none of the considered scenarios will result in the
envisaged reduction of residential heat demand by 20 %.
Intensified policies for the promotion of renewable and innovative heat production lead to an
overachievement of the foreseen share of 18 % independently from the assumed economic
growth and the considered fuel price developments.
High economic growth could allow for strong environmental policy measures both on the
supply and the demand side. A high share of renewable and innovative heat production to-
gether with medium gains in energy efficiency will result in the highest reductions in CO2
emissions among the simulated scenarios. Nevertheless, the ambitious goals of 34.5 % of
CO, emission reduction target from residential buildings would be missed.
Further work will be dedicated to the integration of the investment decision on the demand
side and the resulting competition of supply and demand side measures on the available
budget. Non-economic criteria on the investment decision making could be introduced be-
cause it can be assumed that word-of-mouth effects, green image, ideology and lifestyle have
an impact on the investment decision as well. An additional step will be the combination of
the presented model with a System Dynamics model of the German electricity market which
allows for exhaustive studies of the (combined) effects of environmental policy instruments
on the level of the national energy system.
Acknowledgement
The work has been carried out on behalf of and with financial support from EnBW Energie
Baden Wirttemberg AG.
References
Akbarpour, M., Vaziri, H., 2007. An Investigation into Electricity Subsidy Dynamics by a
System Dynamics Approach. Intemational Conference of the System Dynamics So-
ciety, Boston, MA, USA.
Andres, P., Spiwoks, M., 2000. Prognosegiitemafe - State of the Art der statistischen Ex-
Post-Beurteilung von Prognosen. Sofia-Studien zur Institutionenanalyse, Darmstadt.
Bassi, A. M., 2006. Modeling U.S. Energy with Threshold 21 (T21). International Confe-
rence of the System Dynamics Society, Nijmegen, The Netherlands.
BEE, 2010. Bundesverband Emeuerbare Energien: Wege in die modeme Energiewirtschaft -
Ausbauprognose der Emeuerbare-Energien-Branche, Berlin.
Blesl, M., Kempe, S., Ohl, M., Fahl, U., Konig, A., Jenssen, T., Eltrop, L., 2009. Warmeatlas
Baden-W urttemberg — Erstellung eines Leitfadens und Umsetzung fir Modellregio-
nen, Endbericht, Stuttgart.
Blumberga, A., Zogla, G., Davidsen, P., Moxnes, E., 2011. Residential Energy Efficiency
Policy in Latvia: A System Dynamics Approach. International Conference of the Sys-
tem Dynamics Society, Washington, DC, USA.
BMU, 2005. Federal Ministry for the Environment, Nature Conservation and Nature Conser-
vation. Richtlinien zur Férderung von Mafnahmen zur Nutzung emeuerbarer Ener-
gien, 17th June 2005, last change 11th March 2011, Berlin.
BMU, 2011. Federal Ministry for the Environment, Nature Conservation and Nature Conser-
vation: The path to the energy of the future - reliable, affordable and environmentally
sound, Berlin.
BMWi, 2010. Federal Ministry of Economics and Technology. Energiekonzept fir eine um-
weltschonende, zuverlassige und bezahlbare Energieversorgung, Berlin.
BMWiz, 2011. Federal Ministry of Economics and Technology. Energiedaten - Nationale und
Intermationale Entwicklung, Berlin.
Bohringer, C., 1996. Allgemeine Gleichgewichtsmodelle als Instrument der energie- und
umweltpolitischen Analyse: Theoretische Grundlagen und Empirische Anwendung.
Frankfurt.
Bunn, D. W., Dyner, I., Larsen, E. R., 1997. Modelling latent market power across gas and
electricity markets. System Dynamics Review 13(4): 271-288.
DLR, 2009. Langfristszenarien und Strategien fir den Ausbau ermeuerbarer Energien in
Deutschland. Leitszenario 2009. Federal Ministry for the Environment, Nature Con-
servation and Nature Conservation (Ed.), Berlin.
Dyner, I., Smith, R. A., Pena, G. E., 1995. System Dynamics Modelling for Residential Ener-
gy Efficiency Analysis and Management. The Journal of the Operational Research
Society 46(10): 1163-1173.
Dyner, I., Bunn, D., 1996. Development of a Systems Simulation Platform to Analyse Market
Liberalisation and Integrated Energy Conservation Policies in Colombia. Intemational
Conference of the System Dynamics Society, Cambridge, MA, USA.
EEG, 2000. Gesetz fir den Vorrang Emeuerbarer Energien, 29th March 2000, BGBI. I p.
305, last amended 4th A ugust 2011 in BGBI. I, No. 42, p. 1634, Berlin.
EEWarmeG, 2009. Emeuerbare Energien Warme Gesetz, published in Bundesgesetzblatt
volume 2008 part I No. 36, 18th August 2008, p. 1658, last amended 24th February
2011 as from 1st May 2011, Berlin.
EnEV, 2009. Regulation amending the German Energy Saving Ordinance of 1st February
2002, Berlin.
EC, 2010. European Commission: Energy 2020 - A Strategy for Competitive, Sustainable
and Secure Energy (COM (2010) 639 final of 10 November 2010).
Fichtner, W.; Enzensberger,N.; Rentz, O.; PERSEUS-ICE, 2004, in: Energiemodelle zum
Klimaschutz in liberalisierten Energiemarkten, Forum fiir Energiemodelle und Ener-
giewirtschaftliche Systemanalysen in Deutschland (Ed.), Band 21, 2004.
Ford, A., 1997. System Dynamics and the Electric Power Industry. System Dynamics Review
13(1): 57-85.
Forrester, J. W., 1969. Urban Dynamics. Pegasus Communications, Waltham, MA.
Forrester, J. W., Senge, P. M., 1980. Tests for building confidence in System Dynamics
Models. TIMS Studies in the Management Sciences 14: 209-228.
Gaidosch, L., 2008. Zyklen bei Kraftwerksinvestitionen in liberalisierten Markten - Ein Mo-
dell des deutschen Stromerzeugungsmarktes. Thesis. Fakultat VII — Wirtschaft und
Management, Technische Universitat Berlin, Berlin.
Genoese, M., 2010. Energiewirtschaftliche Analysen des deutschen Strommarkts mit agen-
tenbasierter Simulation. Thesis. Institute for Industrial Production, Karlsruhe Institute
for Technology.
Groesser, S. N., Ulli-Beer, S., Mojtahedzadeh, M. T., 2006. Diffusion Dynamics of Energy-
Efficient Innovations in the Residential Building Environment. International Confe-
rence of the System Dynamics Society, Nijmegen, The Netherlands.
IER, 2005. Forum Energiemodelle und Energiewirtschaftliche Systemanalysen.
http://www. ier.uni-stuttgart.de/forschung/projektwebsites/forum/index/a_index.htm
IWU, 2007. Institut Wohnen und Umwelt GmbH. Deutsche Gebaudetypologie- Systematik
und Datensatze, Darmstadt.
Jager, T., Schmidt, S., Karl, U., 2009. A system dynamics model for the German electricity
market — model development and application. International Conference of the System
Dynamics Society, Albuquerque, NM, USA.
K£W, 2012. KfW Bankengruppe, promotional programmes on housing, home modemisation
and energy conservation,
http://www.kfw.de/kfw/en/Domestic_Promotion/Our_offers/Housing.jsp
KWKG, 2002. Law for preservation, the modemization and the development of the combined
heat and power production (Kraft-W arme-Kopplungsgesetz). BGBI. I p. 1092, last
amended by article 11 as from 28th July 2011, BGBI. 1S. 1634, 1677, Berlin.
Li, H.-B., Dai, H.-X., 2008. System Dynamics Based Research on Energy Efficient Resi-
dence market. International Conference on Wireless Communications, Networking
and Mobile Computing, Dalian, China.
Miller, M., Ulli-Beer, S., 2010. Policy Analysis for the Transformation of Switzerland’s
Stock of Buildings. A Small Model Approach. Intemational Conference of the System
Dynamics Society, Seoul, Korea.
Musango, J. K., Brent, A. C., Bassi, A., 2009. South African Energy Model: A System Dy-
namics Approach. International Conference of the System Dynamics Society, Albu-
querque, NM, USA.
Naill, R. F., Belanger, S. D., 1992. A System Dynamics Model for National Energy Policy
Planning. System Dynamics Review 8(1): 1-19.
Ochoa, P., Ackere, A. v., 2009. Policy changes and the dynamics of capacity expansion in the
Swiss electricity market. Energy Policy 37: 1983-1998.
Pruyt, E., 2007. The EU-25 Power Sector: a System Dynamics Model of Competing Electric-
ity Generation Technologies. Intemational Conference of the System Dynamics So-
ciety, Boston, MA, USA.
Ravn, H. 2001: BALMOREL: A Model for Analyses of the Electricity and CHP Markets in
the Baltic Sea Region. March 2001, download under www.balmorel.com.
Sterman, J. D., 1983. Economic Vulnerability and the Energy Transition. International Con-
ference of the System Dynamics Society, Rensselaerville, New Y ork.
Sterman, J. D., 2000. Business Dynamics - Systems Thinking and Modeling for a Complex
World. New Y ork, Jeffrey J. Shelstad, The McGraw-Hill Companies, Inc.
TUB, 2003. Technische Universitat Berlin. deeco - Related models (which embed infrastruc-
ture capacity limitations). http://www. iet.tu-berlin.de/deeco/related-models.html
Vogstad, K.-O., 2004. A system dynamics analysis of the Nordic electricity market: The tran-
sition from fossil fuelled toward a renewable supply within a liberalised electricity
market. Thesis. Faculty of Information science, Technology, Mathematics and Elec-
trical Engineering, Department of Electrical Power Engineering, Norwegian Universi-
ty of Science and Technology, Trondheim, Norway.
Weidlich, A., Veit, D., 2008: A critical survey of electricity wholesale market models, Ener-
gy Economics Volume: 30, Issue: 4, pp. 1728-1759.
Woldt, T., Pforte, R., Fichtner, W., 2007. Mikro-Kraft-W arme-K opplungsanlagen auf Erd-
gasbasis — Eine Zukunftsoption zur urbanen Endenergiebereitstellung? © Forum der
Forschung 20/2007: Seite 47-54. BTU Cottbus, Eigenverlag, ISSN- Nr.: 0947 -
6989, Cottbus.
ZIV, 2010. Bundesverband des Schornsteinfegerhandwerks — Zentralinnungsverband (Ed.).
Erhebungen des Schomsteinfegerhandwerks fiir 2010.
