Conference PaperPDF Available

A NEW DISTRICT HEATING SYSTEM IN THE CITY OF BOLZANO: DEVELOPING SMART SOLUTIONS WITHIN "SINFONIA" PROJECT

Authors:

Abstract and Figures

In the context of the EU project "Sinfonia", that aims to reduce Bolzano's primary energy consumption up to 40%, SEL AG, an Italian energy company is planning to extend the existing district heating and cooling (DHC) network and explore strategies to improve efficiency, environmental, and economic performance. This research aims to assess the potential energy saving of temperature and peak heating load reduction in the Bolzano's DHC network. Historical performance data from district heating (DH) users were collected and residential building were classified based on construction year and energy performance. The thermal properties of the identified representative buildings were used to develop an indoor climate and energy simulation model by means of IDA-ICE software. The simulation model was validated through real on-site temperature measurements. The potential reduction of temperature supply and peak heat load was evaluated by creating different temperature night setback scenarios. Results indicate that the temperature supply level could be decreased on the consumers' side up to 60 °C, without significantly affecting people comfort. Moreover night setback strategies produced a reasonable reduction of the peak up to 35% from the initial condition, increasing the energy consumption by 4 %.
Content may be subject to copyright.
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
A NEW DISTRICT HEATING SYSTEM IN THE CITY OF BOLZANO: DEVELOPING SMART
SOLUTIONS WITHIN “SINFONIA” PROJECT
A. Mazzocato1,2, H. Li2 and S. Marchiori1
1 SEL AG
2 TECHNICAL UNIVERSITY OF DENMARK, DTU
ABSTRACT
In the context of the EU project “Sinfonia”, that aims to
reduce Bolzano’s primary energy consumption up to
40%, SEL AG, an Italian energy company is planning
to extend the existing district heating and cooling
(DHC) network and explore strategies to improve
efficiency, environmental, and economic performance.
This research aims to assess the potential energy
saving of temperature and peak heating load reduction
in the Bolzano’s DHC network. Historical performance
data from district heating (DH) users were collected
and residential building were classified based on
construction year and energy performance. The
thermal properties of the identified representative
buildings were used to develop an indoor climate and
energy simulation model by means of IDA-ICE
software. The simulation model was validated through
real on-site temperature measurements. The potential
reduction of temperature supply and peak heat load
was evaluated by creating different temperature night
setback scenarios. Results indicate that the
temperature supply level could be decreased on the
consumers’ side up to 60 °C, without significantly
affecting people comfort. Moreover night setback
strategies produced a reasonable reduction of the peak
up to 35% from the initial condition, increasing the
energy consumption by 4 %.
INTRODUCTION
District heating (DH) represents a cost-effective and
sustainable way to provide space heating and domestic
hot water to consumers [1]. The DH network within the
Bolzano province, with 788 km and 15000 users
connections, provides more than 3000 TJ yearly, of
which 2700 are produced from renewable sources as
biomass and biogas [2]. Within this context, SEL AG
manages the DH system of the city of Bolzano. Given
the planned development of the network and the
involvement in a European project called “Sinfonia”,
SEL AG is looking into solutions to improve the
efficiency of the network and thereby contribute to the
overall objective of reducing the energy consumption of
the city by 40 % in the coming years.
Previous researches show the influence of reducing the
supply temperature [3] [4], and of demand
management methods [5]. However, within the Italian
context, findings from previous studies are mainly
based on simulation models that are not based on real
cases. Thus, this study aims to evaluate the effects of
supply and demand strategies on a real case,
specifically the Bolzano DH network.
This research investigates how the creation of a smart
DHC network in Bolzano can be developed to improve
energy, environmental, and economic performance.
The specific objectives of this project are to:
- Create a building classification to identify the
representative building to analyse;
- Develop and validate a building simulation model
based on real measurements;
- Evaluate the potential DH temperature supply
reduction taking into account people comfort;
- Assess methods to reduce the peak heat load
demand by means of temperature night setback.
On one hand, reduced temperature supply could result
in a decreased comfort for those inhabitants of
buildings presenting poor thermal properties. On the
other hand, peak load reduction strategies could
contribute to overcome this problem, obtaining
economic and environmental benefits.
The development of the mentioned points provides an
overview on the correlation between existing buildings
and DH network testing solutions to improve the overall
efficiency of the system.
THE BOLZANO DH SYSTEM
In Italy the development of DH systems began in the
70’s. According to statistics provided by the EU
association Euroheat & Power [6], in 2011 only 5 % of
the Italian population was served by district heating
including a total transmission network of nearly 3.000
km.
The existing Bolzano’s DH system (Figure 1) is
composed by two generation units:
- Bolzano Sud thermal plant (BZ Sud) which
includes 2 combustion engines (natural gas)
working in cogeneration mode and 4 boilers
(natural gas and diesel). The maximum thermal
capacity is 35.7 MWt and the electric 3.5 MWe.
- Waste to Energy plant (WtE) with its maximum
capacity of 30 MWt and 10 MWe. The plant has
just replaced the old incinerator and the large
capacity is justified by the planned expansion
project for the DH network.
Today the network provides heat to 170 users
distributed over three areas: Ipes, Casanova, and the
1
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
Industrial zone while in the next future a larger share of
the city will be involved.
The current Bolzano’s DH network operates, as the
majority of the Italian systems, at a temperature level
ranging from 80 to 95 °C depending on the seasons,
and does not apply any strategy to reduce the peak
loads.
Figure 1: Bolzano DH network
METHODS
The method described below follows the specific
objectives previously introduced and is hereby
described according to the following structure:
1. Building classification;
2. Construction and validation of the building
model;
3. Temperature supply reduction analysis;
4. Peak heat load reduction analysis.
General information about DH connected buildings was
collected to perform the building classification. The
outcome was the definition of the most representative
building to further study. A second data collection
phase was then required including specific thermal
properties of the classification result, but also on site
measurements. A further step consisted of the
construction of a simulation model that was later
validated reflecting real conditions. Finally the model
was tested under different temperature supply and
operation strategies scenarios.
The briefly presented adopted method is hereby
described with more details.
Building classification
Building classification was carried out to identify,
among the DH connected buildings, the most
representative for the current research. In particular
such building should be characterized by poor thermal
properties and at the same time presents great
potential in terms of peak load reduction. In addition the
classification allows a simplification of the analysis and
at the same time makes it replicable.
The method adopted followed the guidelines proposed
by J. Portella including parameters as climate zone,
energy consumption, construction year, and type of
dwelling [7]. In addition to the mentioned features, type
of heating systems, its operation and, area’s dislocation
were taken into consideration.
Construction year and type of dwelling information was
collected by studying available documentations at the
city municipality. Specifically dwellings were classified
as residential and non-residential, while construction
years categorized within specific time span ranging
from earlier 1960 to date.
Climate zone could also be taken into account, but
considering the city of Bolzano this information become
less significant.
The missing documentation in terms of energy
consumption for a large part of the buildings, led to
follow a method that associates construction year with
energy consumption for buildings located in a middle
climate zone [8]. In this way buildings constructed
within defined time periods were related to specific
average energy consumption.
Based on these information the classification was
conducted for the zones named as Ipes and Casanova,
while the industrial area was excluded due to lack of
information.
Construction and validation of the building model
Energy and comfort simulations were required to
evaluate the potential results obtained from the
proposed temperature and peak load reduction
scenarios. With this purpose IDA ICE, a simulation tool
for accurate study of indoor climate and energy
consumption, was selected to assess the behaviour of
the building model under different operation conditions.
The adopted simulation model referred to a single
apartment belonging to the chosen building block.
The analysis in terms of temperature and peak load
reduction was based on a typical working winter week
from 21/01 to 27/01.
In order to study the effect of the suggested solutions
there was the need to develop an earlier stage of the
analysis which included the initial setting of the
simulation model, and its validation based on real on
site measurements.
BZ
Sud
WtE
Industrial
Casanova
Ipes
2
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
The first simulation model construction required the
acquisition of information related to the composition of
the selected building. From available construction
projects geometric features, and thermal properties of
building’s envelope were collected.
In the second part, the model was validated comparing
simulations results with real on site measurements.
Two simulation periods were identified for validating the
model: the first corresponding to 72 hours in March
and, the second to the initially established winter week
(21/01 to 27/01).
To this end, information was collected on both
consumer and network side. Particularly on the building
side personal indoor air temperature on-site
measurements by means of a temperature data logger
(Testo 174T), were recorded for 72 h in March 2014
with a time step of 15 minutes. The lack of real
measurements for the January period was overcome
with personal interviews which revealed the
temperature development throughout a normal winter
day.
On the DH side data collection was conducted through
the SCADA (supervisory control and data acquisition)
system that is used to control the network. The
gathered information refers to the substation installed
at the building block studied including:
- Nominal substation capacity [kW]
- Instant Power [kW]
- Temperature supply and return [°C]
- Mass flow [l/h]
Values were collected for two time periods: March, and
the working winter week with a time step of 15 minutes.
Weather data was also included in the data collection
for both periods to perform reliable simulations.
The flow chart in Figure 2 shows the adopted
procedure to go from the first setting initial simulation to
the validated final model, of which results in terms of
heat load and temperature profiles were comparable to
the real measured data. The most important
parameters to include when performing a IDA ICE
building simulation were location, weather file,
orientation, geometry, thermal properties of surfaces,
thermal bridges, infiltration and room temperature set
point. In addition the heating system needed to be
proper dimensioned to reflect original working
conditions. Such information was applied to the initial
simulation model which is represented by block 1 in the
flow chart.
Daily average space heating load (block 2) and
temperature (block 3) profiles were extrapolated from
simulation’s results and used for comparison with real
measured values. Specifically block 4 compares
simulated and real space heating demand, the latter
was estimated subtracting the domestic hot water
demand during no heating period from the total
measured load in the simulated period. The total space
heating demand was divided by the number of
apartments obtaining the single apartment space
heating load. This operation is made under the
assumption that hot water profile does not significantly
change its profile from summer to winter and the use of
space heating for different apartments normally occurs
simultaneously.
Figure 2: Flow chart, Validation Building Model
Temperature on site measurements collected during
the March period were used to evaluate the
temperature drop that was occurring in the house when
the heating system was stopped. Simple calculation of
thermal time constant τ [h] was conducted and
compared for real and simulated indoor air
temperatures according to relation 1:
  

  

=
(1)
Where:
Tfinal [°C] is the indoor temperature at the end
of the period t
Tinitial [°C] is the indoor temperature at the
beginning of the period t
Tout [°C] is the outdoor average temperature
during the period t
t [h] is the time period
τ [h] is the thermal time constant
In order to positively verify blocks 4 and 5, few
parameters’ adjustments were required (blocks 6 & 7),
especially in terms of building thermal properties,
infiltration rate and temperature set points. Only when
3
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
blocks 5 and 6 were verified the model was considered
validated for the March period simulation.
The same procedure was adopted to additionally
validate the model in correspondence of the winter
January week. In this case heat load profiles were
compared as done for the March period while
temperature analysis was performed based on
personal interviews conducted at the inhabitants. Of
interest for this research was the demonstration that
temperature drop during night hours occurring in the
real case was comparable with simulations’ results.
The final validated model (block 14) satisfied both
stages of validation and was adopted to further assess
temperature and peak load reduction strategies.
Temperature supply reduction analysis
The temperature reduction analysis involved the
creation of different scenarios of the same building
simulation model. The developed simulations differed
only on the temperature supply variable.
Table 1 describes the adopted temperature levels. The
initial case was representative for the actual applied
conditions to the building studied. The temperature of
70 °C is the one provided to the building from the
secondary side of the heat exchanger.
The variation of this parameter within the developed
IDA ICE simulation was conducted under boundary
conditions at the boiler unit. Within the simulation the
boiler replaced the secondary side of the DH heat
exchanger.
Table 1: Tested Scenarios, Temperature Reduction
Initial Case T supply 1 T supply 2 T supply 3
70 °C 65 °C 60 °C 55 °C
Simulations testing the mentioned temperature levels
were conducted for the January winter week. Results
are presented as averages on a daily basis.
Specifically heat load demand and temperature profiles
were analyzed.
Peak heat load reduction analysis
Several scenarios were also developed to analyse the
effects obtained applying peak load reduction
strategies.
The analysis was conducted maintaining building
thermal properties unchanged while testing different
peak load reduction strategies based on the night
setback concept at different temperature supply levels
as presented in Table 2.
Table 2: Tested Scenarios, Peak Load Reduction
Case
A
The temperature supply was the same as in
reality (70 °C). No peak load strategies were
applied, so as the heating system was working
according to the real program. It was used as
base for comparison for all the other solutions.
Case
B
The temperature supply was reduced according
to the value reported for the specific case. No
peak load strategies were applied, so as the
heating system was working according to the real
program.
Case
C
The temperature supply was reduced according
to the value reported for the specific case. The
peak load reduction strategy implied a linear
increment of indoor air temperature set point from
4 AM at 18 °C to 5.30 AM at 21 °C.
Case
D
The temperature supply was reduced according
to the value reported for the specific case. The
peak load reduction strategy implied a linear
increment of indoor air temperature set point from
3 AM at 18 °C to 5.30 AM at 21 °C.
Case
E
The temperature supply was reduced according
to the value reported for the specific case. The
peak load reduction strategy implied a linear
increment of indoor air temperature set point from
2 AM at 18 °C to 5.30 AM at 21 °C.
Case
F
The temperature supply was reduced according
to the value reported for the specific case. The
peak load reduction strategy implied a linear
increment of indoor air temperature set point from
12 AM at 18 °C to 5.30 AM at 21 °C.
Case
G
The temperature supply was reduced according
to the value reported for the specific case. The
peak load reduction strategy implied a continuous
heating operation during the night with indoor air
temperature set point equal to 20 °C.
Case
H
The temperature supply was reduced according
to the value reported for the specific case. The
peak load reduction strategy implied a continuous
heating operation during night and day with indoor
air temperature set point constant at 21 °C.
Results of this part of the research include a
presentation of the new peak loads when applying
specific strategies, energy consumption related to
different operation conditions, indoor temperature
profiles and a simple estimation of the potential
percentage of consumers that could be theoretically
added under studied circumstances. The latter was
calculated by dividing the total peak load reduction with
the space heating dimensioning peak for the single
apartment which was assumed equal to 10 kW for this
case.
RESULTS
Results obtained from the current research are
presented in three main sections, the first describing
the building classification that guided the choice of the
user to further analyse. The second presenting results
obtained through the validation procedure of the
building considered. Results extracted from the
4
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
analysis of reduced temperature supply and peak load
reduction strategies are presented in the last part.
Buildings classification
The Bolzano’s district heating network feeds two
residential zones named as Ipes and Casanova and,
the Industrial zone.
The Ipes zone includes 48 residential building blocks
located at the farthest place from the generation units,
approximately 3 km far from Bolzano Sud thermal
plant. The Ipes zone is characteristic for Bolzano since
is mainly constituted by social housing. The totality of
the users in this area adopted the same technology to
interface the DH network based on instantaneous heat
exchange. With regards to the heating system, the
apartments adopt steel water radiators with the
peculiarity that the control allows the operation only
from 5.30 AM to 10 PM. Most of the buildings were
constructed within 80’s and 90’s, therefore an energy
consumption ranging from 70 to 120 kWh/m2year was
assumed [8].
Figure 3: Ipes, Construction Year
Casanova represents a newly constructed district in the
suburb of Bolzano. As Ipes, it is composed by
residential users, specifically 9 building blocks and 2
single family houses. This area can be considered
geographically the closest to the generation units since
it is located less than 1 km far from Bolzano Sud
thermal plant. Differently to the Ipes zone, the use of
DH was coupled with the energy produced by
renewable sources consequently DH substations were
provided with storage tanks. In terms of heating
system, the use of low temperature radiant heating was
preferred to the conventional water radiators without
setting any specific control’s constraints. Casanova
buildings, being constructed from 2006 to 2012 with a
special focus on energy consumption, present an
average energy demand approximately equal to 35
kWh/m2year.
The chosen solution was a building block located in the
Ipes zone which includes 75 apartments constructed in
1983. The selection was supported by the following
consideration:
- On one hand the adoption of low temperature floor
heating as in the Casanova district should avoid
comfort problems in case of lower DH temperature
operation. In addition, DH substations provided
with storage tanks limit considerably the formation
of high peak loads.
- On the other hand the large share of residential
buildings constructed within 80’s and 90’s as in the
Ipes zone and the use of water radiators
dimensioned to work at certain temperature levels
could lower people comfort if DH temperature
operation would be reduced. In addition the
present Ipes heating control strategy showed its
direct influence on the creation of high peak loads.
The simulation was then conducted for a 90 m2
apartment belonging to the chosen building block.
Table 3 presents the most important thermal properties
used for building the simulation model while Figure 4
the apartment’s layout.
Table 3: Thermal Properties, Ipes
Surface Thickness [m] U-value [W/m
K]
External Wall 0.32 0.462
Internal Wall 0.12 1.7
Internal Floor 0.175 2.38
Roof 0.36 0.7
Glazing / 1.8
Infiltration rate 0.4 ACH
Figure 4: Layout and Orientation, Ipes
The DH substation connecting the building to the
network included a 500 kW and 300 kW heat
exchangers for covering space heating and domestic
hot water demand respectively.
Construction and validation of the building model
As presented in the method section, the validation of
the described building model was carried out for two
specific time periods. Results obtained from the March
validation procedure are reported in Figure 5
comparing the real estimated space heating profile with
simulations’ results. It is also presented in Figure 5 the
indoor air temperature profile according to simulation.
The validation procedure performed for the March
period showed that similar results were reached in
terms of real and simulated heat load profile. Both
2% 9%
33%
56%
before 1960
1961-1975
1976-1990
1991-2005
2006-Today
5
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
profiles showed a significant peak at morning hours
corresponding to starting heating system operation.
The curves then decreased up to evening hours, at
which another peak, of a smaller size, occurred.
In addition to the validation in terms of heat load profile,
Table 4 presents thermal time constant of the building
calculated based on temperature measurements and
simulation results. The temperature used for such
estimation resulted in two time constant that could be
considered representative of the same building.
Figure 5: Validation Space Heating Profile, March
Table 4: Time Constant, March Validation
Time constant (simulation) [h] Time constant (real) [h]
80 90.7
As mentioned in the methodology, the space heating
profile for the single apartment was not available and
consequently was extracted from the available total
measured data of the building. Specifically Figure 6
presents the total and DHW profiles for the whole
building block, from which the SH profile for the single
apartment, which constitute the simulations’ base, was
obtained.
Figure 6 showed also that the deviation between space
heating and domestic hot water profiles for the building
studied was considerable. The total peak demand
reached almost 600 kW, of which only 100 kW comes
from domestic hot water use.
Figure 6: Total relation SH & DHW, January
Figure 7 compares the SH demand profile for the single
apartment obtained from Figure 6 with the simulations
results.
Also in this case, the heat load profile found from the
simulation matched the estimated space heating profile
for the single apartment studied. The morning peaks for
both profiles were around 7 kW while the evening
demands around 4 kW.
The lack of real on-site temperature measurements for
the specific winter week did not allow the calculation of
the time constant. Anyway, the night temperature drop
was validated conducting personal interviews to the
inhabitants who confirmed temperature levels around
18 °C at morning hours for an average winter day.
Figure 7: Validation Space Heating Profile, January
The simulation model was consider representative for
the actual building and consequently was used for
further temperature and peak load reduction analysis.
Temperature supply reduction analysis
The current chapter presents the obtained effects when
applying a reduction of the temperature supply. Figure
8 shows that the heat load profile is reduced
consequently to the reduced temperature. Especially
lower the temperature mainly influences the peak hours
while does not significantly affect the rest of the day.
The largest decrement is registered at 55 °C, which
lower the peak from 7 (initial case) to 4.6 kW.
Figure 8: Heat Load Profiles, Temperature Reduction
The investigated issue related with operation
temperature reduction is the effect that could result on
the consumer’s side. Figure 9 revealed on that aspect
that little deviations could be obtained when
implementing temperature levels equal to 65 °C or 60
°C. Differently it shows that 55 °C as supply
temperature for the heating system is likely to result in
reduced comfort for the inhabitants since the indoor
6
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
temperature deviation is larger than 1 °C at morning
hours.
Figure 9: Temperature Profiles, Temperature Reduction
Peak heat load reduction analysis
Obtained results from simulations conducted applying
peak load reduction strategies are presented. The
simulated cases described in Table 2 were tested at all
temperature levels considered in the previous results
section. In order to provide an overview about the
changes obtained in terms of heat load profile, Figure
10 shows the curves for the cases studied at the actual
temperature level (70 °C).
Figure 10: Heat Load Profiles, Peak Load Reduction
The anticipation of heating system operation lowered
the peak which is occurring at morning hours in the real
case. More specific analysis in relation to the peak
reduction could be obtained from Figure 11 which
presents the percentage reduction for the cases
studied at all temperature levels. Few things are worth
mentioning observing Figure 11:
- Lower temperature operation as proved by Figure
8 leads to lower peaks;
- Largest reduction is obtained for the extreme case
in which the heating system operates continuously
throughout the day, this solution is not influenced
by different temperature levels;
- Except for Case H, the strategy proposed in Case
F leads to largest reduction. For instance,
considering temperature supply equal to 70 °C, the
peak is reduced by 5 %.
Figure 11: Peak Load Reduction, Percentage
Table 5, according to the calculation presented in the
method section, presents what could be one of the
direct effects obtained from the reduced peak load, that
is the increment of potential consumers. The study
shows that strategies adopted could increase by 10 to
30 % the number of consumers, reaching its maximum
in case H with 40 %. Anyway additional consideration
must be specified in terms of energy consumption as
introduced by Figure 12.
Table 5: Potential Consumers Increment
Consumers
Increment
[%]
Case B
Case C
Case D
Case E
Case F
Case G
Case H
T sup 70 °C
/
1.9
2.75
3.5
4.9
3
40
T sup 65 °C
10
11.5
12.3
13
14
9.8
40.6
T sup 60 °C
19.8
21.2
21.8
22.5
23.4
22.9
40.7
T sup 55 °C
29.2
30.6
31.1
31.6
32.5
32.5
41.5
As mentioned another important factor to consider
when implementing night set back strategies was the
energy consumption that derives from them. Indeed
some of the controls tested showed a negative
response in terms of energy consumption resulting in a
too large increment of it.
Figure 12: Energy Consumption, Percentage
Figure 12 presents the deviation from real conditions in
terms of energy consumption obtained from the
simulations conducted. A fundamental outcome of this
part of the research was that all the strategies applied
out of Case H resulted in an increment of the energy
7
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
consumed lower than 4 %. Differently, simulation under
Case H conditions showed energy consumption
increased by 13 %.
In addition to the presented results, daily temperature
profile were evaluated for all the tested temperature
levels. It is worth presenting the outcome of the
simulations conducted at 55 °C. Figure 9 proved that
this temperature level coupled with the actual operation
strategy could result in lower comfort for the
inhabitants. Figure 13, instead provides an overview of
how people comfort could be influenced by operation
strategies. The result showed that the adoption of the
lowest tested temperature level could be taken into
consideration if coupled with proper operation
strategies.
Figure 13: Indoor Daily Temperature, T sup 55 °C
DISCUSSION
The conducted research opens different discussion
points which are hereby described.
1. The temperature reduction analysis shows that some
limits should be preserved when adopting lower
operation temperatures. 55°C for instance, seems to be
too low in the case studied to provide enough comfort
to the occupants even though it was proved that this
temperature level could be taken into consideration
applying appropriate operation strategies based on
night set back. For instance simulations shows that low
temperature operation mainly influenced the morning
peak, and consequently applying an operation strategy
which start the heating system during the night and
linearly increases its set point up to the morning hours,
significantly reduced comfort problems.
2. The extreme tested case H was performed to define
the lower limit that could be theoretically reached when
the goal is reducing the peak. Anyway the analysis in
terms of energy consumption shows for this solution
the largest energy requirement categorizing this case
as inefficient for the research’s purpose.
3. The reduced peak demand could lead to another
significant advantage for the Bolzano’s case, inasmuch
the actual situation sees the waste to energy plant
covering the base load while Bolzano Sud takes care of
peaks. Looking at the planned development of the
network the studied strategies could be adopted to
maximize the use of the heat coming from the waste to
energy plant and at the same time minimize gas
boiler’s use. This will provide great economic and
environmental benefits being the Bolzano Sud plant
more expensive in terms of operational costs and
presenting a larger rate of emissions than the waste to
energy plant.
4. Peak load reduction strategies could lead to another
significant benefit. From a theoretical point of view the
reduced capacity required could be employed by DH
utilities to connect more consumers. In this way the DH
market share would be enhanced obtaining also
advantages from a societal point of view (larger
substitution of single boilers presenting higher
emissions rate).
5. With consideration to the studied Ipes zone, further
improvement in relation to the energy consumption
could be obtained by varying night set back strategies
adopted. Larger energy consumption is expected
applying the same strategy (e.g. Case F) to all
buildings rather than diversify the strategies over the
whole zone. This choice, in order to be efficient, has to
be coupled with a deep analysis of every single
building and its thermal properties to define the proper
control.
6. Methodological considerations must also be
provided:
Strength of the research is the application of real
measurements to validate the simulation model. In this
way a more representative outcome was obtained.
Anyway some deviations could have influenced the
results with consideration to on site measurements and
estimation of the single space heating profile for the
apartment studied.
OUTLOOK
At first the Industrial zone of the city should be included
in further analysis since it constitutes a great part of the
existing network in which different technologies than
the residential areas are applied.
Furthermore in the next future the research will focus
on how the concepts studied and presented in this
paper can influence DH systems at all levels from
design to operation. It will be studied how such
changes can be applied to the existing Bolzano DH
network in terms of dimensioning criteria, heat losses
and pumps’ use.
Once the system will be optimized under these terms
the next is the development of a smart network
improving its performance through intelligent control
and communication integrated with system operational
optimization.
8
The 14th International Symposium on District Heating and Cooling,
September 7th to September 9th, 2014, Stockholm, Sweden
CONCLUSIONS
Given the considerations and uncertainties stated in
this paper some conclusions could be drawn.
- As mentioned in the outlook the final goal for
Bolzano is the achievement of the development of
a smart DH network. Anyway, several aspects at a
lower level were highlighted to show their
contribution to improve system’s efficiency.
- The analysis shows that a possible temperature
reduction can be taken into consideration from the
consumers’ side. However further researches are
required to study the effect of this choice on the
network side.
- With consideration to the peak load reduction
studies, it was proved that by simply setting
differently the set point, peaks could be reduced.
Such strategy appears easy to apply, especially for
those areas in which the heating system is
controlled by a single institution (social housing).
The company also might incentive the application
of night set back control through economic benefits
for the consumers.
ACKNOWLEDGEMENT
This project was supported by SEL AG and is the result
of an ongoing collaboration between SEL AG and
Technical University of Denmark (DTU).
REFERENCES
[1] A. Dalla Rosa, “The Development of a New District
Heating Concept, Doctoral Thesis, Lyngby, 2012.
[2] Municipality of Bolzano
www.provincia.bz.it/agenzia-ambiente
[3] P. Johansson, Buildings and District Heating -
contributions to development and assessments of
efficient technology”, ISBN 978-91-7473-130-9,
Lund, 2011.
[4] O. Gudmundsson, A. Nielsen and J. Iversen, “The
effects of lowering the network temperatures in
existing networks”, in Proc. of the 13th International
Symposium on District Heating and Cooling,
Copenhagen, 2012.
[5] F. Wernstedt, P. Davidsson and C. Johansson,
“Demand Side Management in District Heating
Systems” in Proc. of the International Conference
on Autonomous Agents 2007, pp. 1383-1389
[6] EU association Euroheat & Power
http://www.euroheat.org/Italy-82.aspx
[7] J.M.R. Portella, “Bottom-up description of the
French building stock, including archetype
buildings and energy demand”, Technical report no
T2012-380, Sweden, 2012.
[8] TABULA Project Team, “Application of Building
Typologies for Modelling the Energy Balance of the
Residential Building Stock”, ISBN 978-3-941140-
23-3, Germany, 2012.
9
ResearchGate has not been able to resolve any citations for this publication.
Conference Paper
Full-text available
This paper describes a multiagent system that has made the voyage from research project to commercialised product. The purpose for the multiagent system is to dynamically control a system so that the load of the system is below certain threshold values without reduction of quality of service and by that, to avoid the usage of top load production sources and to reduce energy consumption. The fundamental idea behind the system is that a large number of small local decisions taken all in all have great impact on the overall system performance. A field-test as well as a return of investment analysis are presented.
Buildings and District Heatingcontributions to development and assessments of efficient technology
  • P Johansson
P. Johansson, "Buildings and District Heatingcontributions to development and assessments of efficient technology", ISBN 978-91-7473-130-9, Lund, 2011.
The effects of lowering the network temperatures in existing networks
  • O Gudmundsson
  • A Nielsen
  • J Iversen
O. Gudmundsson, A. Nielsen and J. Iversen, "The effects of lowering the network temperatures in existing networks", in Proc. of the 13th International Symposium on District Heating and Cooling, Copenhagen, 2012.
The Development of a New District Heating Concept
  • A Rosa
A. Dalla Rosa, "The Development of a New District Heating Concept", Doctoral Thesis, Lyngby, 2012.
Application of Building Typologies for Modelling the Energy Balance of the Residential Building Stock
  • Tabula Project Team
TABULA Project Team, "Application of Building Typologies for Modelling the Energy Balance of the Residential Building Stock", ISBN 978-3-941140-23-3, Germany, 2012.
Bottom-up description of the French building stock, including archetype buildings and energy demand
  • J M R Portella
J.M.R. Portella, "Bottom-up description of the French building stock, including archetype buildings and energy demand", Technical report no T2012-380, Sweden, 2012.