Scholarly article on topic 'Assessing the Optimal Use of Electric Heating Systems for Integrating Renewable Energy Sources'

Assessing the Optimal Use of Electric Heating Systems for Integrating Renewable Energy Sources Academic research paper on "Civil engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Energy Procedia
OECD Field of science
Keywords
{"Bottom-up modelling" / "energy system analysis" / "heating energy demand" / "scenario analysis" / "demand side response"}

Abstract of research paper on Civil engineering, author of scientific article — Tobias Boßmann, Rainer Elsland, Anna-Lena Klingler, Giacomo Catenazzi, Martin Jakob

Abstract In order to decarbonise the energy system, alternative transition pathways were assessed in recent years that agree on the crucial role of the heating sector, due to its potential to increase the renewable-based share of energy demand and to improve energy efficiency. Simultaneously, the increasing share of fluctuating renewable energy sources (RES), such as wind and photovoltaic, raises the hourly volatility in the energy system and challenges energy utilities. To balance energy demand and RES in an optimal manner storage-based heating systems can be used as flexible loads. This study discusses to what extent smart residential heating systems can contribute to the integration of RES and quantifies the trade-off between electricity savings and flexibility provision, when replacing storage heaters by heat pumps. To answer these research questions a simulation-based scenario analysis is conducted until 2050 encompassing France, Germany and the United Kingdom. These countries are of specific interest due to their substantial share of electric heating as well as their exponentially rising wind and photovoltaic (PV) generation capacities. The study reveals that the long-term potential of heating technologies for the integration of RES is relatively limited in countries such as France or Germany, given the improved insulation of buildings and the seasonal offset between PV generation and heat demand. However, in the short- to medium-term, or in countries with low shares of PV generation but high shares of wind power (such as the UK), heating technologies may facilitate the integration of RES, especially in the absence of alternative flexibility options.

Academic research paper on topic "Assessing the Optimal Use of Electric Heating Systems for Integrating Renewable Energy Sources"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Energy Procedia 83 (2015) 130- 139

7th International Conference on Sustainability in Energy and Buildings

Assessing the optimal use of electric heating systems for integrating

renewable energy sources

Tobias BoBmanna*, Rainer Elslanda, Anna-Lena Klinglera, Giacomo Catenazzib, Martin

Jakobb

a Fraunhofer Institute for Systems and Innovation Research ISI, Breslauerstr. 48, 76139 Karlsruhe, Germany b TEPEnergy, Rothbuchstr. 68, 8037 Zurich, Switzerland

Abstract

In order to decarbonise the energy system, alternative transition pathways were assessed in recent years that agree on the crucial role of the heating sector, due to its potential to increase the renewable-based share of energy demand and to improve energy efficiency. Simultaneously, the increasing share of fluctuating renewable energy sources (RES), such as wind and photovoltaic, raises the hourly volatility in the energy system and challenges energy utilities. To balance energy demand and RES in an optimal manner storage-based heating systems can be used as flexible loads. This study discusses to what extent smart residential heating systems can contribute to the integration of RES and quantifies the trade-off between electricity savings and flexibility provision, when replacing storage heaters by heat pumps. To answer these research questions a simulation-based scenario analysis is conducted until 2050 encompassing France, Germany and the United Kingdom. These countries are of specific interest due to their substantial share of electric heating as well as their exponentially rising wind and photovoltaic (PV) generation capacities. The study reveals that the long-term potential of heating technologies for the integration of RES is relatively limited in countries such as France or Germany, given the improved insulation of buildings and the seasonal offset between PV generation and heat demand. However, in the short- to medium-term, or in countries with low shares of PV generation but high shares of wind power (such as the UK), heating technologies may facilitate the integration of RES, especially in the absence of alternative flexibility options.

© 2015 The Authors.PublishedbyElsevierLtd. Thisis an open access article under the CC BY-NC-ND license

(http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of KES International

Keywords: Bottom-up modelling; energy system analysis; heating energy demand; scenario analysis; demand side response

* Tobias BoBmann. Tel.: +49-721-6809-257; fax: +49-721-6809-272. E-mail address: tobias.bossmann@isi.fraunhofer.de

1876-6102 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of KES International

doi: 10.1016/j.egypro.2015.12.203

1. Introduction

In its roadmap for building a low-carbon economy, the European Commission has drafted potential pathways towards an greenhouse gas emission reduction of 80 to 95% compared to 1990 levels [1]. The achievement of this target builds upon two major strategies: a comprehensive diffusion of renewable energy sources (RES) and the extensive use of energy efficiency measures to trigger energy savings. These two cornerstones aim to make the European Union's energy system more secure and sustainable and its economy more competitive.

Various scenario analyses have recently been performed to assess potential transition pathways (see e.g. the comparative study of Förster et al. [2]). The majority of studies agree that the heating sector will play a crucial role in the transformation of the energy system: Its electrification increases the use of RES-based, low-carbon electricity and helps to cut carbon emissions. In addition, the dissemination of highly-efficient heating technologies, namely heat pumps, substantially reduces primary energy demand.

However, RES deployment is expected to be realised primarily in the electricity sector, particularly through the diffusion of fluctuating generation technologies such as wind and photo voltaics (PV). Consequently, the net load which is defined as the national system load minus the generation from RES must be assumed to feature a corresponding increase in hourly volatility in the long run. At present, utilities and grid operators use storage heaters to shave load peaks and ensure a continuous utilisation of the electricity generation and transmission infrastructure, by scheduling them in night time. In the future, utilities might benefit from the shiftable capacity of storage heaters and heat pumps to balance supply and demand and integrate renewable electricity generation.

The research questions presented by the proposed framework are: To what extent can smart residential heating systems contribute to the integration of renewable energy sources? And what is the trade-off between electricity savings and flexibility, when replacing storage heaters by heat pumps?

This study aims to address these research questions by applying two simulation models in the context of a scenario analysis until 2050. It quantifies the extent to which electric heating systems in the residential sector can facilitate the integration of RES and smooth the net load under different pricing schemes. Given that the shape of the net load is also subject to transformation due to changes on the demand side (as demonstrated, for instance by [3]), the impact of electric vehicle charging is explicitly taken into account. The regional scope of this study encompasses France, Germany and the United Kingdom (UK). These countries are of specific interest because France has experienced a substantial electrification of the heating sector over the past decades and the provision of sanitary hot water here relies substantially on storage heater technology [4]. In contrast, Germany is characterised by a clear decline in the installation of night storage heaters. Simultaneously, the installation of wind and PV generation capacities is rising exponentially here and the country has an ambitious strategy for the insulation of buildings. The UK, on the other hand, is much less focused on housing insulation and it propagates the installation of heat pumps instead, aiming to cover 30% of all heat provided using this technology [5].

The remainder of this study is structured as follows: Section 2 introduces the two simulation models applied in this study. Section 3 describes the scenario analysis including the major assumptions and presents the key findings of the two models. Finally, Section 4 contains the conclusions and a critical discussion of the results.

2. Methodological approach

Within this study the energy demand models FORECAST (FORecasting Energy Consumption Analysis and Simulation Tool) and eLOAD (energy LOad curve ADjustment tool) are applied that are frequently used in German and European studies for policy makers and industrial customers [6-9]. FORECAST analysis energy demand on an annual basis, which is subsequently broken down into hourly load curves by eLOAD (see Figure 1). In the following, the structural framework and modelling procedure of both models as well as their linkage are discussed.

FORECAST

The energy demand model FORECAST aims to develop long-term scenarios of the EU28+3 (3: Norway, Switzerland, Turkey) by country up to 2050. FORECAST comprises four individual modules: industry, tertiary and residential module each representing a demand side sector and the rest is captured by the module 'Others' containing the agriculture and transport sector. While all sector modules follow a similar bottom-up methodology,

they also consider the particularities of each sector like technology structure, heterogeneity of actors, level of end-consumer energy carrier prices and data availability.

The bottom-up approach allows modelling the diffusion of technologies as the result of individual investment decisions taken over time. For all types of investment decisions, the model follows a simulation approach rather than optimisation in order to better capture the real-life behaviour of companies and households. The investment decision is modelled as a discrete choice process, where decision makers choose among rival technologies (considering different energy carriers) to satisfy a certain energy service. The discrete choice modelling is implemented as a logitapproach considering the total cost of ownership (TCO) of an investment plus further intangible costs and benefits. This approach ensures that even if one technology is more cost-effective than the others, it will not gain a 100% market share. This effect reflects heterogeneity in the market, niche markets and non-rational behaviour of companies and households, which is a central capability to model policies. Especially in the case of heating system replacement, path dependencies need to be considered in the context of decision-making as cost advantages vary depending on the system to be replaced. This means that lock-in effects have to be considered when modelling the costs of fuel switching in terms of investment cost mark-ups as they essentially influence the transition probability. FORECAST also captures the transition to and from non-electricity driven rival alternatives.

In this study the entire electricity demand of each of the three countries is considered, whereas the focus is set on the residential sector (FORECAST-Residential) which is composed of two modules: 'household appliances and lighting' (Module 1) and 'heating systems' (Module 2). The following discussion is narrowed to Module 2.

The framework of heating system analysis is provided by the useful energy demand for heating purposes derived from the country-specific building typology differentiated by construction period (<1960, 1961-1990, 1991-2008, 2009-2020, 2021-2050). These are in turn divided into building types (single-family-houses (SFH), and multi-family-houses (MFH)) and five thermal efficiency standards. The standards are represented by a set of building elements of roof, wall, floor and window together with their thermal characteristics. This allows to analyse the useful energy demand based on the buildings' physics distinguished by transmission and ventilation heat losses as well as solar and internal heat gains. Splitting the new building stock into those constructed before and after 2020 is related to the fact that major policy regulations regarding the energy performance of buildings are defined for the year 2020 (e.g. EPBD recast) [10]. Considering the building typology in this way, results in a total of 50 reference building segments per country.

The useful energy demand coverage by heating systems for space heating and sanitary hot water purposes is calculated in a subsequent step. The electricity-related heating systems analysed in the heating module are: heat pumps, direct electric heating and storage heater. To receive the electricity demand by heating system on an annual basis, the useful energy demand is transformed into final energy demand by system-specific performance factors. The latter incorporate the entire set of technological parameters that determine the utilization factor of a heating system, e.g. heating system efficiency, efficiency decrease due to part load operation and heat losses.

eLOAD aims to estimate the future shape of the national electricity system load curves and to quantify the contribution of demand side technologies to smooth the net load (also known as demand response, DR). It is available for all countries of the EU27 until the year 2050. eLOAD consists of two modules. The first module addresses the deformation of the load curve due to structural changes on the demand side and the introduction of new appliances (such as electric vehicles) by applying a partial decomposition approach (PDA). The technology specific annual demand projection from the FORECAST model (see Figure 1) serves for the identification of all "relevant appliances" that feature a significant increase or decrease in electricity consumption over the projection horizon. By using appliance specific load profiles, a load curve can be generated for all relevant appliances for the base year, according to the respective annual demand in the base year. These load curves are deduced from the system load curve of the base year. The resulting remaining load curve and the appliance specific load curves are then scaled for all projection years according to the demand evolution. Reassembling the scaled remaining load and the scaled load curves gives the load curve of the projection year.

The main advantage of the PDA consists of its ability to properly take into consideration structural changes in overall annual demand by explicitly modelling the main drivers for load curve deformation while preserving stochastic outliers and characteristic irregularities from historic load curves. Using the results from the FORECAST

model entails the benefit of integrating the assets of bottom-up simulation modelling. This ensures an appropriate representation of a large variety of appliances, of the diffusion of new technologies, of the evolution of techno-economic characteristics as well as the consideration of macro-economic drivers and political measures.

55 C J5

If (U ^

e 0 1 | S A 1/ H

FORECAST Macro

Gross value added, employment, households, etc.

FORECAST Pricing

Sector specific end-consumer prices

FORECAST Annual demand

Industry Tertiary Residential Others Agriculture Rail Transport Electric mobility Others

Branch „J""1 ■SPaper Branch Finance

Process Pspe* Saving ShOfl Option press service Lighting m ra Technology Screens E^ncy LCD

Results

Consumption, Potentials, Indicators, GHG Emissions

Consideration of structural change

Load curves, load profiles Temperature

Hourly demand

Demand response (DR)

DR Parameters

Renewable electricity generation

Results

Hourly load curves and DR potentials

Fig 1: Overview of the structural framework and modelling procedure of FORECAST and eLOAD [8]

The second module of eLOAD addresses the active adjustment of the load curve by means of DR. The net load is defined as the overall system load curve minus the generation of RES. Smoothing the net load enables an efficient operation of existing electricity generation assets and grid infrastructure as well as a diminished need for investment in new capacities. A major input of the DR module are hourly generation profiles from renewable energy sources.

Net load smoothing is carried out by determining the optimal load schedule for appliances/processes that are particularly suitable for DR. The optimisation is performed from the consumer perspective. A consumer group receives a day-ahead price or net load signal. Based on the electricity consumption pattern (i.e. the load profile) of the respective process as well as techno-economic parameters and restrictions (e.g. capacity, storage or organizational constraints), a mixed-integer optimisation is carried out to determine the least-cost scheduling of the load. The DR module delivers a quantitative load shifting potential, providing detailed information about the seasonal, weekly as well as hourly load shifting availability of the individual appliances. It generates a smoothed net load curve that may be used in an electricity market model to quantify the impacts of DR on the electricity system.

3. Case Study

3.1. Scenario definition

The aim of this case study is to determine how electric heating systems in the residential sector influence the level of electricity demand and the extent to which they may contribute to the provision of flexibility to integrate renewable energies in France, Germany and the UK until 2050. In terms of the technological scope, the focus is set on storage heating, direct electric heating, and heat pumps for space heating (SH) and sanitary hot water (SHW) purposes in the residential sector. Every heating system is represented by an average reference technology. To

analyse the impact on national annual electricity demand for these three countries, the other demand side sectors (industry, tertiary, agriculture and transport) are captured in this scenario analysis as well. However, the hourly DR modelling performed with the eLOAD model is limited to storage heaters and heat pumps and the year 2050. Resistance heaters without storage and electric vehicles are only considered for the load curve projection, in order to appropriately represent the change in the overall system load.

In the DR analysis we distinguish two price mechanisms: time-of-use (TOU) and real time pricing (RTP). The major difference between the two mechanisms consists in their temporal variability. The TOU mechanism provides a price pattern that is fixed for each typical day. The RTP, instead, varies on an hourly basis and directly reflects the volatility of the wholesale price. In the absence of an electricity market model, in this study, the net load is used as indicator for the evolution of the wholesale price. Both, the TOU and the RTP, equal on average end-user price that is assumed for the FORECAST calculations.

3.2. Parameter framework

The socio-economic parameters set the quantitative framework for the scenario analysis, which are taken from a study published by the Energy System Analysis Agency (ESA2) [11]. In France and the UK population and the number of dwelling rises continuously from 2010 to 2050 with an ongoing time horizon mainly due to immigration. In contrast, population is decreasing in Germany. Electricity prices rise drastically in the first two decades and subsequently decrease by 10-20 % until 2050. The GDP growth rate is assumed to be 1.7% p.a. on average for 20102050, which is the main driver of electricity demand in the industry and tertiary sector. To ensure comparability, the framework parameters are equal in both scenarios. The building-related and heating-related parameters are taken from current projects such as ENTRANZE [12].

The hourly load profiles of heating technologies, used for the projection of the system load curve, originate primarily from literature surveys (e.g. across German utilities), load profile data bases (e.g. REMODECE, [13]) or load records from field surveys (e.g. DEFRA household electricity study, [14]). Load profiles of space heating technologies feature a daily variation in level as function of the outdoor temperature. In contrast, load profiles of DHW technologies are solely distinguished across the different typical days. The load profile of electric vehicles is based upon an extensive assessment of German vehicle driving profiles and the assumption that vehicle batteries are charged after the last journey of the day [15]. The Harmonised European Time-Use Survey, HETUS [16] indicates that daily activities in France (namely working, shopping, leisure) are characterised by a one-hour delay compared to Germany, the charging load profile is correspondingly shifted. No adjustment is required for the UK.

With respect to the assessment of the optimal operation of the heating technologies, assumptions are required in terms of generation from renewable energy sources and underlying pricing schemes that trigger load adjustment. The renewable electricity generation builds likewise upon the ESA2 report [11]. It is characterised by significant annual growth rates of up to 7%, e.g. in the UK (see Table 1).

Table 1: Overview of renewable electricity generation in 2050

France Germany UK

Wind Solar PV Wind Solar PV Wind Solar PV

Installed capacity [GW] 76.4 90.7 65.2 99.0 113 43.8

Generation [TWh] 168 100 159 112 286 43.8

The electricity generation profiles of the different types of RES builds upon empirical data from 2010 [17]. They are explicitly exhibited in the results Section 3.4. The TOU pricing scheme considered consists of an 8 hour low tariff price zone, a 4 hour peak price zone and a 12 hour default price zone. The location of the price zones is endogenously determined by the eLOAD model based on the hourly wholesale price. The RTP reflects the daily net load volatility by normalising the maximum to the value of 1.

3.3. Annual demand results

When comparing electricity demand by country 2010 vs. 2050 (see left side of Figure 2), the analysis reveals that the national electricity demand increases from 449.2 TWh to 479.6 TWh (+6.8 %) in France, decreases in Germany from 527.3 TWh to 482.2 TWh (-8.55 %) and increases in the UK from 328.8 TWh to 364.3 TWh (+10.8 %). The general trends by sector behind these developments are the following:

• Residential sector: increase of population in the short- to medium term due to immigration, person per dwelling ratio decreases continuously, white appliances are in a saturation status, strong increase of ICT-appliances, diffusion of high efficient lighting technologies, strong increase of buildings' thermal efficiency

• Tertiary sector: continuing of 'tertiarisation', slow down of the increase of employee in tertiary sector due to ageing of population, floor area per employee increases, occupancy control leads to decreasing electricity demand, increase of ventilation and air conditioning demand, demand for ICT-appliances increases.

• Industry sector: as many processes were optimised in the past the electricity saving potentials of energy-intensive processes are relatively low, long lifetime of industrial equipment slows down diffusion of innovative production technologies, price pressure on electricity and emission prices prevent the faster diffusion of more efficient production technology, continuous shift from primary to secondary production in many industrial processes (steel, paper, copper), significant savings potentials are available for cross-cutting technologies at low cost.

• Agriculture and transport sector: the electricity demand allocated to these sectors is mainly driven by the diffusion of electro mobility, the political framework has a positive effect on reducing battery costs and puts customer acceptance on a broader footing, the technical (e.g. safety, service life) and economic development goals (around 200 to 300 euro/kWh battery price until 2020) are reached in succession, the hybridisation of vehicles becomes widespread because of their economic efficiency.

Analysing the useful energy demand by purpose in each country shows that space heating demand decreases by 28.4 % in France, by 28.8 % in the UK, and by 31.2 % in Germany until 2050. In parallel, the useful energy demand for sanitary hot water purposes increases by 22.1 % in the UK, by 13.2 % in France and decreases by 11.7 % in Germany until 2050. In terms of heating systems, the analysis reveals similar technological trends in each country: an increase of heat pumps and a decrease of direct electric heating and storage heating (see right side of Figure 2). However, in this context it needs to be considered that thermal efficiency of buildings essentially determines the technological and monetary framework conditions for the heating systems to be installed. For instance, single-family-houses with a very low level of thermal efficiency like passive houses are mainly equipped with heat pumps. Another example is that single-family-houses with a low level of thermal efficiency improvement during 2010 and 2050 tend to stick to storage heating systems.

Fig 2: Total electricity demand by sector (left) and analysed technologies (right) by country for 2010 and 2050 For heat pumps the strongest increase can be seen for Germany from 2.7 TWh to 19.7 TWh (+723.3 %) in 2050,

whereas 20.9 % are attributed to sanitary hot water purposes. Electricity-driven storage heating systems for sanitary hot water purposes show the strongest decrease in France from 18.9 TWh to 10.4 TWh in 2050 and in terms of space heating in Germany from 12.8 TWh in 2010 to 2.3 TWh in 2050. To put the results into perspective, Figure 2 also contains the electricity demand of electric mobility, which is substantial for the following discussion about the evolution of the system load and the load shifting potential.

3.4. Hourly load results

In the following the hourly results are presented and analysed. The study focuses on the hourly system load curves (i.e. the annual demand translated into hourly demand patterns), the net load curves (i.e. the system load less the generation from fluctuating renewable energy sources) and the load of the individual processes.

Figure 3 depicts the net and system load curves for France, Germany and the UK without the application of demand response tariffs. Due to the high number of direct electric heaters for space heating, France features still a very high seasonal demand variation in 2050 in comparison with the other countries, although it has declined by 11% due to efficiency measures. Seasonal variation also reduces in the UK (by 4%), but not in Germany. Here efficiency in heating systems comes with their electrification by replacing old systems with electricity driven heat pumps. While differing in seasonal variation, the analysed countries show a similar evolution in the daily demand structure: On the one hand the demand reduction in storage heating systems leads to a lower load level in the early morning hours, but on the other hand the diffusion of electric vehicles induces an evening peak that is especially pronounced on weekdays.

Fig 3: Average net (left) and system load curves (right) in 2050 in France (blue), Germany (red) and the UK (green).

On an average day, electric vehicles increase the evening peak by 13.6 GW in France, 10.3 GW in Germany, and 12.6 GW in the UK. The structure is also similar in the three countries when it comes to the net load curves, but with quite substantial differences in load levels. The UK features a high wind power generation which results in very low net demand levels with negative demands in the early morning hours in winter and the midday hours in summer where PV generation reduces the net load additionally. Due to the PV induced midday load valley the EV peak is even more accentuated in the net load curve than in the system load. Germany and France are characterised by their high generation of PV which leads to significant seasonal differences due to higher solar radiation in summer where the net demand becomes negative during midday.

Like described above, TOU and RTP price mechanisms are introduced for the optimal operation of the heating systems. For a better understanding, the effects of both tariffs are analysed in detail for the UK and on a more qualitative level for the two other countries. The left part of Figure 4 shows the average net load curve for the UK in 2050 under TOU and RTP pricing mechanisms. The non-optimised load curve and the empiric system load of the year 2010 are included as a benchmark. The maximum load shifted from one hour is 6.30 GW in the RTP scenario (about 8% of the maximum demand of 76.9 GW) which is apparently not enough to smoothen the EV evening peak without directly applying load management on the electric vehicles, but both pricing schemes reduce the amount of negative net demand from -43.1 TWh to -42.0 TWh and -40.6 TWh for TOU and RTP, respectively. Here the RTP

tariff suits better for load smoothing and RES integration. Another issue with the TOU tariff is the consumers' reaction it provokes: By applying the TOU price, all consumers shift their load simultaneously to the beginning of the low price zone (see right part of Figure 4) and a load peak is induced at 9 am (cf. left part of Figure 4). This so-called avalanche effect can be observed in more detail in Figure 5.

Fig 4: Average net load in the UK in 2050 (blue) under TOU (green) and RTP (red) price mechanisms and in 2010 (black) (left figure) and corresponding pricing schemes (right figure).

Figure 5 illustrates that under the TOU tariff heating systems immediately start working as soon as the low tariff is active (at 9 am in summer) and the operate again just before the low tariff hours are over (at 5 pm in summer) to fill their storages. To fill the heating demand in winter it is not sufficient to run the heating systems only during low tariff hours. This applies especially for heat pumps, which operate best at relatively low temperature levels and have therefore limited storage capacities. It can be seen that, when using the TOU pricing mechanism, heat pumps for space heating only spare the high tariff hours in the cold season but operate also during the normal priced periods. Storage heaters in contrast, which can store energy at higher temperature levels, only work during low tariff hours. The operation of both heating systems is more continuous with the RTP tariff scheme.

Fig 5: Load profiles for heating systems in the UK, empirical (black) and in 2050 under TOU (green) and RTP (red) tariffs

Even though the heating load profiles change quite heavily from 2010 to 2050, especially in summer season when the cheapest hours are no longer at night but during midday, the overall change in the net load curve is limited to a few GW (see Table 2). This is reasoned in the relative small amount of electricity demand of heating systems compared to renewable generation. Table 2 lists the amounts of shifted load in the RTP scenario (that showed the higher potential of the two tariffs) and generated renewable power. Even in the UK with a relative low PV generation, this is the critical quantity due to its particular profile, only 1.56 TWh of shiftable load (with a maximum of 2.54 GW load reduction) stand against 23.4 TWh of PV power in the summer season. The contrast is even higher in France and Germany.

Table 2: Overview of the load shifting potential with RTP pricing in 2050 in summer and winter season

France Germany UK

Summer Winter Summer Winter Summer Winter

Heat Pumps [TWh] 0.623 2.67 1.19 4.01 0.471 1.58

Storage heaters [TWh] 2.51 3.43 0.449 1.24 1.09 1.85

Pmax (Heat Pumps) [GW] 1.06 2.54 1.28 2.64 0.985 1.87

Pmax (Storage heaters) [GW] 1.62 2.44 1.16 2.68 1.59 2.94

PV power [TWh] 47.4 20.6 55.2 15.9 23.4 5.33

Wind power [TWh] 51.1 79.5 44.7 73.2 76.4 138

In Table 3 some indicators are given to describe the changes in the net load curve with application of the TOU and the RTP price mechanism. In all countries both tariffs are especially effective in reducing the minimum load and therefore the negative net demand. The maximum load is only reduced by a small amount, due to the fix EV charging profile that induces the evening peak. In general the hourly volatility of the load curve reduces with the RTP, which is already indicated by the changes in Pmin and Pmax, but is expressed more explicitly in the ramp rate factor. The change in load between two consecutive hours is referred to as ramp rate (rr). The ramp rate factor (rrf) then represents the mean absolute ramp rate for the entire year, set with respect to maximum load. A different performance can be observed for the TOU tariff: although the overall minimum demand reduces with its application, the rrf increases. This effect accounts for the induced peak, when heating systems shift their demand towards the beginning of a low price period.

Table 3: Characterisation of the net load curve without demand response and with TOU and RTP tariff in 2050

France Germany UK

No DR TOU RTP No DR TOU RTP No DR TOU RTP

Net demand [TWh] 211 211 211 211 211 211 348 348 348

Net demand P<0 [TWh] -18.1 -14.7 -13.3 -21.8 -19.5 -18.7 -43.1 -42.0 -40.6

Pmin [GW] -48.4 -45.7 -40.6 -58.2 -55.9 -52.5 -37.5 -36.6 -31.6

Pmax [GW] 81.6 79.5 79.5 81.8 79.6 76.6 62.0 61.0 61.0

rrf [%] 5.97 6.04 5.25 6.24 6.38 5.75 5.70 5.59 5.12

Conclusions

Four major conclusions can be drawn from the scenario-based assessment of the heating technologies in the three different countries. First, the electricity demand attributed to heating systems is expected to decline substantially for two main reasons: improved insulation of the building shell reduces useful heat demand and the technological shift to heat pumps reduces the specific final energy demand per unit of useful heat by approximately the factor of three. Second, in exchange for the declining electricity demand, the shift towards heat pumps also implies a substantially lower load shifting potential. In strongly simplified terms, it can be said that equipping the same number of households with heat pumps instead of storage heaters for space heating decreases the load shifting potential by approximately 80%. Third, with respect to seasonal variations, our year-long hourly assessment reveals that heating

systems have limited suitability for integrating RES. They offer the greatest flexibility during the winter when PV generation is at a minimum and their flexibility is much lower (less than 25%) during the summer when PV generation peaks and the net loads drops significantly below zero. Fourth, an analysis of the two different pricing schemes, TOU and RTP, clearly underlines the strength of the RTP scheme, which triggers a real-time load adjustment, enabling a significant reduction of the peak load and surplus renewable energy. This is particularly true for countries that have high volatility in the net load due to significant shares of PV-based electricity generation. At the same time, however, the RTP requires more sophisticated communication and control infrastructure whereas the TOU scheme is more easily implemented and yields sufficiently good results in the winter time, when heating devices are required to run at night, as they do nowadays.

To sum up, the long-term potential of heating technologies for RES integration is relatively limited in countries such as France or Germany, given the seasonal offset between PV generation and heat demand and continuously improved building insulation. However, in the short- to mid-term perspective and in countries with low shares of PV generation but high shares of wind power (such as the United Kingdom), heating technologies may contribute to balancing demand and supply, especially in the absence of alternative flexibility options (such as electric vehicles). Future research activities should extend the scope of analysis to additional sensitivities in terms of heating technology diffusion. Comparing the load management potentials of alternative technologies across the coming decades could also provide additional insights.

References

[1] European Commission. A Roadmap for moving to a competitive low carbon economy in 2050. COM (2011) 112. Brussels; 2011.

[2] Förster H., Healy S, Loreck C, Matthes F, Fischedick M, Samadi S, Venjakob J. Metastudy Analysis on 2050 Energy Scenarios. ÖkoInstitut, Wuppertal Institut, Berlin, Wuppertal; 2012.

[3] Boßmann T, Elsland R, Lickert F, Wietschel M. The German load curve in 2050: structural changes through energy efficiency measures and their impacts on the electricity supply side. In: Summer study on energy efficiency (ECEEE 2013); Hyeres; 2013.

[4] Réseau de transport d'électricité. Bilan provisionnel de l'équilibre offre-demande de l'électricité en France - Edition 2013. Paris; 2013.

[5] CCC. The Fourth Carbon Budget - Reducing emissions through the 2020s. Committee on Climate Change; London; 2010.

[6] Energy System Analysis Agency. Shaping our energy system - combining European modelling expertise. Karlsruhe; 2013.

[7] Öko-Institut, Fraunhofer ISI, Forschungszentrum Jülich, DIW. Politikszenarien für den Klimaschutz VI - Treibhausgasemissionen bis zum Jahr 2030., Karlsruhe; 2013.

[8] Fraunhofer ISI, IREES, TEP Energy. FORECAST/eLOAD website. http://www.forecast-model.eu/forecast-en/index.php; Karlsruhe. Last accessed 25-02-2015.

[9] Elsland R, Bradke H, Wietschel M. A European impact assessment of the Eco-Design requirements for heating systems - What kind of savings can we expect?. Energy Procedia; Karlsruhe; 2014.

[10] EUR-Lex. Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010 on the energy performance of buildings (recast). Brussels; 2010.

[11] Energy System Analysis Agency. ESA2 website. http://www.esa2.eu/web/guest/mesap-web-access; Karlsruhe. Last accessed 25-012015.

[12] Intelligent Energy Europe. ENTRANZE. http://www.entranze.enerdata.eu/; Grenoble. Last accessed 17-11-2014.

[13] de Almeida A.. REMODECE - European Database. http://remodece.isr.uc.pt/database/database.htm. Last accessed 21-12-2014.

[14] DEFRA (2014). Household Electricity Study - EV0702. http://randd.defra.gov.uk/Default.aspx?Menu=Menu& Module=More& Location= None&Completed=0&ProjectID=17359. Last accessed 10-12-2014.

[15] Dallinger D, Schubert G, Wietschel M. Integration of intermittent renewable power supply using grid-connected vehicles - A 2030 case study for California and Germany. Applied Energy; 104:666-682.

[16] Statistika Centralbyran. Harmonised European Time Use Survey. https://www.h2.scb.se/tus/tus/. Last accessed on 04-12-2014.

[17] Schubert G. Modelling hourly electricity generation from PV and wind plants in Europe. In: 9th International Conference on the European Energy Market (EEM12); Firenze; 2012.