Eindhoven University of Technology
MASTER
Supporting the Eindhoven University of Technology to reach thermal energy balance at the Campus 2020
Spruijt, J.G.
Award date:
2015
Link to publication
Disclaimer
This document contains a student thesis (bachelor's or master's), as authored by a student at Eindhoven University of Technology. Student theses are made available in the TU/e repository upon obtaining the required degree. The grade received is not published on the document as presented in the repository. The required complexity or quality of research of student theses may vary by program, and the required minimum study period may vary in duration.
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
Supporting the Eindhoven University of Technology to reach Thermal Energy Balance
at the Campus 2020
Student: J.G. Spruijt ID: 0788803
Graduation date: 24 February 2015
Graduation Supervision Committee:
Prof.dr.ir. J.L.M. Hensen dr. M.H. Hassan Mohamed
Mr. J.I. Torrens Galdiz
Unit Building Physics and Services, Department of the Built Environment University of Technology Eindhoven
J.G. (Tjeerd) Spruijt
Supervised by: J.I. Torrens, J.L.M. Hensen Unit BPS
Eindhoven University of Technology Eindhoven, the Netherlands TjeerdSpruijt@hotmail.com NOMENCLATURE
AER Annual Energy Report
ATES Aquifer Thermal Energy Storage BTES Borehole Thermal Energy Storage CTES Cavern Thermal Energy Storage
MBE Mean Biased Error
MJA Multiple Years Agreement
RE Real Estate department of the TU/e RMSE Root Mean Square Error
TES Thermal Energy Storage
TU/e Eindhoven University of Technology UNFCCC United Nations Framework Convention on
Climate Change
UTES Underground Thermal Energy Storage ABSTRACT
Every building requires a certain amount of heating or cooling, whenever a building requires cooling (heat removal) in summer and heating (heat supply) in winter, a mismatch in time exists. A solution to this problem could be to store the thermal energy on a seasonal period. In the Netherlands water containing ground layers (Aquifers) are often used to store the energy because of their common availability [1]. It is expected that in 2020 a quantity of 20.000 ATES systems be applied in the Netherlands, while there were around 2.000 applications in the year 2011 [2]. One of the largest ATES systems in the Netherlands is located on the campus of the Eindhoven University of Technology (TU/e). This system has been providing the TU/e with warm and cold water for the last 13 years. During this time two cooling towers have been emitting heat into the air to overcome the difference between the heat and cold extraction. The last few years the TU/e has made plans to renovate the campus and its real estate [3]. This renovation might be the opportunity to create a better balanced ATES system. To do this, computational simulations were used to research the current and future balance in the heating and cooling demand of the TU/e buildings real estate. The result of the research shows that the TU/e real estate in 2020 will have a larger cooling than heating load. Some solutions are mentioned which could help decrease the mismatch
Keywords—Aquifer, ATES, Dymola, Low Grade Energy, Modelica, TES, Thermal Energy Storage.
1. INTRODUCTION
Everybuilding requires a certain amount of energy, to keep the indoor climate at a comfortable level. This amount of energy depends on the outdoor climate and the buildings properties. This amount of energy (cooling or heating demand) is different for every building. When a building requires cooling in summer and heating in winter, a mismatch in time exists [4]. If said in another way, this might sound even more obvious:
When in summer thermal energy (heat) has to be removed from a building, and in winter thermal energy has to be supplied to that same building, a mismatch in time exists.
1.1. Thermal Energy Storage
To overcome this mismatch, the thermal energy can be stored and made available for recovery when the energy is required. This is possible with a Thermal Energy Storage (TES). Storing thermal energy for a larger time then a few hours would require a large amount of insulation material to prevent the thermal energy from escaping the storage. Instead of using insulation material, an environment with a constant temperature like the ground can also be used to keep the energy from escaping, a system like this is called an Underground Thermal Energy Storage (UTES) [5] [6]. There are different types of UTES’s possible; ATES (Aquifer), BTES (Borehole) and CTES (Cavern) [7] [8]. The type of storage to use in the Netherlands is not difficult to choose, as aquifers (water-containing layers between two impermeable layers) are commonly available in the whole country. This is the reason why the ATES systems are becoming more and more popular in the Netherlands. It is even expected that in the year 2020 around 20.000 of these systems will be operative [2].
The ATES system can be carried out in two ways, a closed system or an open system. The closed system is a U- shaped tube inserted in the ground. A liquid of a certain temperature is send through the tube. Depending on the ground water temperature in the aquifer and the temperature in the tube, the system will extract/provide thermal energy from/to the aquifer. The open system extracts groundwater of a certain temperature, with the countercurrent principle thermal energy is extracted/provided to the groundwater and the water is then reinjected in another well with another temperature.
Supporting the Eindhoven University of Technology to Reach Thermal Energy Balance
at the Campus 2020
The most common reason to choose for the application of a TES system is to reduce the energy use of the building.
Since a TES system stores the energy, it can be reused later with only a small amount of extra energy. The storage uses multiple wells, and is thus able to store both heat and cold, so in winter less gas is necessary to heat the building and in summer less electricity is needed to cool the building.
When the United Nations signed the United Nations Framework Convention on Climate Change (UNFCCC) [9]
in 1992, a promise was made to prevent the increase of greenhouse gas emissions. Even stricter promises were made in the Kyoto-protocol [10], this protocol set the goal to reduce the production of greenhouse gasses instead of reducing the growth of the emissions. In response to these worldwide goals, the Eindhoven University of Technology (TU/e) made its own plan. The TU/e aimed to improve the energy efficiency by 20% between 1996 and 2006, to reach this goal, a large scale ATES was applied in 2002 to provide in the cooling and heating demand for the TU/e buildings located on the campus [11].
1.2. Regulations on ATES
Since an ATES system makes use of a common source (the freshwater), it is bound to some regulations. These regulations are there to prevent someone from making a common source unavailable to others [12]. A number of regulations are applied to the use of an ATES system in the Netherlands. These regulations are contained in the Water act [13] , the Milieu act [14] and the official gazette of the Netherlands [15] which states that:
“An open ground energy storage system reaches a moment where there is no excess heat within 5 year after first use and repeats this within 5 year after the last time this situation occurred.”
Excess heat as used in this law occurs when the total amount of thermal energy which is infiltrated into the ground is larger than the total amount of thermal energy that is extracted.
On top of this, the provincial water plan of Noord-Brabant [16] states that both heat and cold are used, only in exceptional cases discharge to the air or surface water is allowed to create a balanced situation underground.
1.3. TU/e Campus
The TU/e is one of the three technical universities located in the Netherlands and provides education for around 8.000 students and 1.200 PhD students. The university buildings are located on a campus of 121 hectares near the center of the city Eindhoven.
The ATES system located on the TU/e campus is a large open circuit system created out of 32 wells (16 cold and 16 warm) divided over 6 clusters (3 cold and 3 warm) [Figure 1]
[11]. The clusters are connected to each other making use of two ring tubes (cold and warm), the rings are kept at around 6°C and 18°C under a pressure of 2.5 bar. With this ring system buildings can extract or supply heat and cold at the same time. The most common heat and cold delivery system at the TU/e is direct cooling and heating via a heat pump, in most cases a peak load system with boilers is installed to deal with outdoor temperatures below -5°C. Figure 3 is a schematic representation of this type of system.
Figure 1: The six clusters of the TU/e ATES system [IF Technology]
Figure 2: Energy extraction from the ATES
2003 2004 2005 2006 2007 2008 2009 2010 2011 Cold use -8.384 -8.407 -8.882 -10.505 -9.400 -8.360 -9.180 -8.124 -8.922
Heat use 332 2.393 3.384 4.316 3.200 4.460 3.600 4.350 3.634
Heat extraction CT 0 5.323 4.326 4.858 3.900 4.800 5.110 5.690 5.080 Cum. Imb. according to AER 0 -691 -1.863 -3.194 -5.494 -4.594 -5.064 -3.148 -3.356
-12.000 -10.000 -8.000 -6.000 -4.000 -2.000 0 2.000 4.000 6.000 8.000
Energy [MWh/year]
Time [Year]
TU/e ATES Extraction
49%
21%
30%
When the TU/e ATES system was created in 2002, twelve buildings were connected to the system. Ten of these buildings are combined users, one is only using heat and one is only using cold. The buildings have different functions, sports, education, research and related business.
Since the European Union has set new goals [17], the TU/e is required to make some changes in its real estate collection.
A plan “Campus 2020” was made to renovate the campus before the year 2020. The plan includes a number of measures like a reduction of 60.000m2 in floor area and renovation of existing buildings to come to an energy neutral campus.
Another step to make is the increase in efficiency of the ATES system, which is possible because the twelve buildings that are connected are together not in a thermal energy balance. This problem is currently overcome using two cooling towers. Figure 2 shows the energy extraction from the ATES over the years 2003-2011.
Figure 3: Schematic representation of the Heating and cooling installation of Vertigo [18]
1.4. Research
As a response to the new European goals, the TU/e created its own plan to come to an energy neutral campus. Part of this plan is the “Campus 2020” which is a plan to increase the energy efficiency and decrease the carbon emissions. This large renovation of the TU/e campus is happening as we speak. The renovation of the campus includes multiple buildings that are currently energy inefficient, as well as the optimalisation of installations like the ATES system. This system is currently unbalanced and two cooling towers are used to balance out the ATES. This is a waste of thermal energy, as well as electricity and money. Since the TU/e is
currently renovating, this might be the moment to research if and how a thermal balance can be found in the TU/e real estate/ATES. To do this, the heating and cooling energy demand of the buildings has to be known. Real measurement data will be used as well as simulation based performance analysis of the different TU/e buildings to check what the heating and cooling demand of the buildings is or will be after renovation.
2. METHODS
The TU/e campus covers 121 hectares and tens of buildings are located on this campus. Some of these have been connected to the ATES since the ATES was created. To get an overview of the last 13 year that the ATES system was in production, the available data of all buildings will be analyzed, mainly focusing on the total ATES balance. The amount of heat or cold extracted for use and the heat extracted for the cooling towers will be analyzed to see what the actual imbalance is.
2.1. Classification of buildings
To create some overview in the wide range of buildings located on the TU/e campus four partitions were made, of which one can directly be excluded as it contains the buildings that are not and will not be connected to the ATES system.
The division is based on data from the Annual Energy Reports (AER) of 2003-2010 [19]–[26] and the Management review Multiple Year Agreement Energy Efficiency TU/e [27]. These documents contain the annual energy consumption per building in gas, water, ATES cold and ATES heat. For determination of the future ATES users, the pamphlet “TU/e science park” is used [28].
According to these documents, twelve buildings were connected to the ATES one year after it was built in 2002. Of these twelve buildings, only seven will remain the same between 2010 and 2020, the other five will be changed, renovated or demolished. Nine buildings will be newly connected to the ATES system before 2020. The division of the buildings over the groups is given in -3.
2.1.1. Group 1
The first group contains seven buildings that are still connected, and will not be getting a renovation before 2020.
For this group; a lot of measurement data is available, as well as architectural drawings and installation schemes. Since a lot of real data is known for these buildings, they are simulated with a computational model. After simulation they are calibrated using the simulation results and the measured data.
The calibrated model is used as a guide for the other buildings.
Figure 4: Group 1, 2 and 3 on campus
Table 1: TU/e buildings divided into groups Groups
1 2A 2B 3
Sportcentrum Ceres Hoofdgebouw Flux Auditorium Metaforum Potentiaal Differ
Vertigo Spectrum Gemini- Noord
Vestide toren Matrix Laplace Gemini-Zuid Studenten
dorp Helix Catalyst
Cyclotron Kennispoort
2.1.2. Group 2
Group 2 contains existing buildings as well, of these buildings some will receive just minor changes or have been connected to the ATES recently, others will receive a complete renovation before 2020. The group is thus split into two parts, A and B.
Group A contains five buildings of which two already have a connection to the ATES system, however some changes will be made in the buildings resulting in a changed energy
demand. The other three buildings are new connections to the ATES since 2011, this means that two or three years of measurement data of these buildings is available. For the first two buildings some calculations will be done to tell what the amount of cooling or heating energy will be, for the other three the measured data will be used as the energy demand for those buildings.
Group B of the second group is made out of buildings that will be fully renovated. Two of these buildings were already connected to the ATES (Gemini Noord and Gemini Zuid), so measurement data is available for those. The other two will be new connections to the ATES, gas and electricity data is available of these buildings, so this can be used. These four buildings will be simulated using the same model as group one, the real measured data of the un-renovated buildings will be used to make a comparison with the simulation results.
This comparison will give an estimation of the trend in the building heating and cooling demand.
2.1.3. Group 3
Third is a group with buildings that are being built at the moment, are just finished or will be built in the near future.
Some estimations may be made of these buildings, however no data is available of these four buildings.
Table 2: Basic information Group 1 [29]
Table 3: Basic information Group 2A and 2B
GROUP 1 Name
Sport-
centrum Auditorium Vertigo Matrix Helix Cyclotron Kennis- poort
ImageYear
1967 1966 1965 1960 1996 1967 2002
Renovation
2001 1995 2002 N/A N/A 2001 N/A
Function
Retail and
leisure Academic Academic Academic Academic Academic Related Business
FunctionalFloor Area
7169 5864 16098 3660 16124 5026 6708
GROUP 2A GROUP 2B
Ceres Meta-
forum Spectrum Laplace Catalyst Hoofd-
gebouw Potentiaal Gemini Zuid
Gemini Noord
1959 1959 2002 1972 2012 1963 1963 1974 1974
2012 2013 N/A N/A N/A 2015 N/A N/A 1999
Related
Business Academic Academic Academic Related
Business Academic Academic Academic Academic
49 10531 4431 5534 24014 10995 11764 8825
2.1.4. Group 4
The fourth group exists out of buildings that are not and will not be connected to the ATES, as this group has no influence on the balance of the TU/e ATES the group will not further be mentioned. Further information about each building is available in appendix 2
2.2. Data analysis
The creation of the ATES system started in 2002, in this time a number of buildings were directly connected to the ATES, creating a load on the system. At the same time two cooling towers were used to counter the calculated imbalance. Data about the ATES can be found in the Annual Energy reports [19]–[26] and the Multiple Year Agreement Energy Efficiency TU/e [27]. The data measured of the ATES contains heat and cold extraction per building, heat extraction for the cooling towers, electricity and water.
Since group 1 and 2 [Table 1, Table 2, Table 3] contain only existing buildings, real measurement data is available for each one of these buildings. This data has been extracted from the real estate servers as well as MYA [27] documents and AER [19]–[26]. Only measurement data since 2002 are needed since that is when the ATES was applied. The year 2011 will be used as the simulation year because this is the last year that only the first connected buildings were applied to the ATES, after this year new buildings have been connected, and other buildings have been demolished which influences the ATES.
The measured data contains hourly measurements of the ATES as well as gas, electricity and water use. Most buildings are supplied with sensors for the HOK, HLK and
HRK which are the main electricity sensors for fluctuating, lighting and constant power. The fluctuating power net supplies energy to elevators and installations, the lighting power is for the light net, and the constant power is used for measurement equipment. For some buildings extra sensors are applied for specific parts, like a heat pump electricity sensor in the Vertigo building. If this is the case, the coefficient of performance (COP) can be used to calculate the production of the heat pump.
The gas use can be multiplied with the caloric energy of the gas and a boiler efficiency.
2.3. Model
De collected data will be used for the computational simulations using Dymola, which is a simulation environment in the open Modelica modelling language. The model used is based on the ISO 13790- which is a five resistances 1 capacitance (5R1C) model [Figure 5] [30]. Extra additions to this model are a user schedule, ventilation with heat recovery and weather influences, the computational code of the model can be found in appendix 1 [31].
A base load is calculated using computation simulation models of the first group, this group is calibrated with the real measured data. This base load represents the load that has always been and will be present, from this point the balance is checked. The second step is to add group 2A, this is possible by adding the measured data, which is available for 3 of the 5 buildings since they are already in use.
The other two buildings will receive minor changes in their system, resulting in a different energy demand towards the ATES system.
The third group (2B) is simulated using the same calibrated model of group 1, the current un-renovated building is the main input for this model. Then the insulation values of the most recent Dutch Building Code [Table 4] will be applied to the model, and the function change is kept in mind.
Figure 5: ISO-13790 model (5R1C) [30]
Table 4: Dutch building code requirements U-value
Wall [W/m2K]
U-value Roof [W/m2K]
U-value Floor [W/m2K]
Residential
0.4 0.4 0.4
Commercial
0.22 0.167 0.286
The simulated energy demand from all groups is combined to analyze what the balance without new buildings will be. This will give an indication if the new buildings will need to use more heat or cold energy from the ATES. The new buildings will be discussed and finally some possible measures will be named.3. RESULTS
The data analysis of the ATES, shows that there is a small imbalance in the yearly heat and cold extraction [
Figure 2]. In this data the heat extraction by the cooling towers is not seen as an imbalance but as extraction, and with it the annual imbalance is around 1 to 2GWh. If the cooling towers are seen as imbalance (it is available heat), the imbalance would be around 5 to 6GWh.
When the average annual ATES energy demand per building is calculated [Figure 6], it shows that three buildings stand out due to their large influence on the ATES system, these buildings are; Helix, Cyclotron and Spectrum. These are three laboratory facilities located on the TU/e campus.
Figure 6: Percentage of total use per building 3.1. Schedules
To be able to create a user schedule in the simulation model, a plot is made of the weekly and daily energy demands. The third week of January was chosen as this is a normal week.
These plots for each of the buildings of group 1 are added to appendix 2. For the Auditorium building a measurement error was found [Figure 7] in the data. This is the ATES which collected data when 1.000kWh was provided instead of the hourly value. Another week of the same year shows correct data [Figure 8].
3.2. Simulation results
The first building simulated is the Vertigo building, this building was chosen due to the large amount of sensors and measurement data available. A description and schematic representation of the installation system were retrieved from the FAGO report [18] [Figure 3]. The base load for heating is provided by the ATES via a heat pump, if this is insufficient (exterior temperature -5°C) then the two gas-fired boilers can step by. The cooling is done directly, the ATES provides cooling to the ventilation system. As an extra cooling option the heat pump can be used in reverse mode. This system is not in detail implemented in the model although it would be possible it would increase the simulation time.
3.2.1. Data modification
As an example the Vertigo building is used in this article. The real measured data of the Vertigo building contains gas, ATES warm, ATES cold and electricity for the heat pump. The total amount of heat is calculated with the following steps:
- Gas use multiplied by 9. This value is the result of the caloric value of gas which is approximately 32MJ/m3 [32] multiplied by the efficiency. 1kWh is equal to 3.6MJ so, 1m3 of gas can provide 8.88kWh with an optimal combustion efficiency of 100% the conversion value was estimated at 9.
- The heat energy provide by the ATES is already in kWh_thermal and can be added after a check if the measured values of the real estate department are the same as the values provided in the AER.
- The electricity use of the heat pump is a more difficult story this is split between the heating and cooling, if cooling exists without heating, then the electricity for the heat pump is multiplied by 5 (EER) and otherwise it is multiplied by 4 (COP). [33]
0 02%
00%
04%
04%05%
17%
03%
12%
07%
11%
06%
01%
03%
01%
0 500 1.000 1.500 2.000 2.500 3.000
0 500 1.000 1.500 2.000 2.500 3.000
Cold usage [MWh/year]
Heat usage [MWh/year]
Av. perc. of annual energy use per building (2003-2013)
Sportcentrum Auditorium
Ceres W-hal / Metaforum
Vertigo Matrix
Helix Catalyst
Cyclotron N-Laag / Flux Spectrum W Hoog / Gemini zuid W Laag / Gemini noord Laplace
Kennispoort
Figure 7: Auditorium weekly schedule data error
Figure 8: Auditorium weekly schedule correct data
3.3. Calibration of the model
The errors in the comparison between the measured and simulated data [ Figure 9] can be due to the input values, which were assumed to be correct. This doesn’t have to be the case, and thus the model has to be calibrated.
For calibration of the model, first the Root Mean Square Error (RMSE) is used [Equation 1], an important note here may be that since a low-resolution model is used, the RMSE will never get as low as 20. After solving the errors in the model, the first simulation delivered a RMSE value of 45 for cooling and 56 for heating. The unit of the RMSE is kWh, and since it is calculated using the daily average, the value is in kWh/day.
Since the ATES should be balanced once every 5 year [15], the focus is not on the hourly or daily error, the yearly error is detailed enough. Thus the MBE [Equation 2] might be useful as well. The MBE uses the daily error as well, the difference with this and the RMSE is that the negative and positive errors eliminate each other in the MBE, while the RMSE uses both as a positive error. The residual in these equations is the result of the simulated minus the measured hourly data.
= ∑ /2 Equation 1: RMSE-value [34]
= ∑ Equation 2: MBE-value [34]
To calibrate the model (get a better RMSE/MBE) the calibration signature [Equation 3] and characteristic signature [Equation 4] can be used [Figure 10 and Figure 11], these are ways to see what effect a changed parameter has on the simulation results. Since a low resolution model is used, the data is fluctuating a lot, and instead of the expected line, a cloud was found as the calibration signature. To see which parameters had to be adjusted, a trend line was plotted into the graphs. Calibration signatures can be found for every
simulated building in appendix 3.
= −
! !
! "
100%
Equation 3: Calibration signature [34]
ℎ ' ' =
ℎ
" ' !(
! !
" ' !(
100%
Equation 4: Characteristic signature [34]
0 500 1000 1500
15 Jan 2011 0:00
16 Jan 2011 0:00
17 Jan 2011 0:00
18 Jan 2011 0:00
19 Jan 2011 0:00
20 Jan 2011 0:00
21 Jan 2011 0:00 Energy demand [kWh/hour]
Time [day]
Auditorium Week
Total cold Total heat
0 200 400 600
19 Nov 2011 0:00
20 Nov 2011 0:00
21 Nov 2011 0:00
22 Nov 2011 0:00
23 Nov 2011 0:00
24 Nov 2011 0:00
25 Nov 2011 Energy demand [kWh/hour] 0:00
Time [day]
Auditorium Week
Total cold Total heat
Figure 9: Un-calibrated annual energy demand Vertigo
Figure 10: Cooling calibration signature for Vertigo
Figure 11: Heating calibration signature for Vertigo
These calibration signatures show that the cooling has to increase when outdoor temperature is high and the heating has to increase when the outdoor temperature is low.
Different parameters are then changed to see what would be of correct influence on the results. The calibration signature is calculated with the first simulation as a baseline.
The final result is a calibrated model with a certain RMSE/MBE value as close to zero as possible. Table 5 shows the RMSE and MBE values for the buildings of group 1.
The calibrated model is found when a combination of different measures is made. This building was eventually calibrated with a RMSE of under 50 for both heat and cold, and an MBE around -3. The annual cooling demand [Figure 12] and the annual heating demand [Figure 13] show a great similarity with the measured data.
All buildings of group one were calibrated in this way, the results of these calibrations can be found in Table 5.
0 100 200 300 400 500 600
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 Daily average energy demand [kWh/day]
Time [day]
Vertigo Uncalibrated Cold
Measured_cold Simulated_cold
0 100 200 300 400 500 600
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 Daily average energy demand [kWh/day]
Time [day]
Vertigo Uncalibrated Heat
Measured_heat Simulated_heat
-50 0 50 100
-20Cold water [%] -10 0 10 20 30 40 50 Temperature [°C]
Vertigo Cal_Sig_cold
-20 0 20 40 60
-20Hot water [%] -10 0 10 20 30 40 50 Temperature [°C]
Vertigo Cal_Sig_heat
Figure 12: Annual cooling demand calibrated model
Figure 13: Annual heating demand calibrated model
Table 5: Calibration data group 1
1 A base load was removed during the simulation
2 These buildings have not been calibrated 3.3.1. First group
The measured/simulated data exists out of 2 parts, the gas/electricity and the ATES. Out of the measured data, the percentage of energy that was delivered by the ATES can be calculated. Vertigo is the only building where the electricity towards the heat pump is measured, so this would be the most precise measured data. Out of this data a percentage of around 90% is coming from the ATES for cold, and around 50% for heat. For large users like Helix and Cyclotron the heat part is around 65% and for small users like Auditorium and Matrix a percentage of 16 is reached.
Table 6: Total demands versus ATES demands Total
Cold
[MWh/year]
ATES Cold
[MWh/year]
COLD
[%]
Sportcentrum
0 0 N/A
Auditorium
200 200 100
Vertigo
597 530 89
Matrix
831 831 100
Helix
1018 1018 100
Cyclotron
1451* 1432 100
Kennispoort
397 397 100
TOTAL 4494 4408 98
Total Heat
[MWh/year]
ATES Heat
[MWh/year]
Heat
[%]
Sportcentrum
950* 216 18.5
Auditorium
910 147 16.15
Vertigo
856 461 54
Matrix
1682 263 15.6
Helix
5773 2701 46.8
Cyclotron
1524 1298 85
Kennispoort
683 0 N/A
TOTAL 12378 5086 40
0 100 200 300 400 500 600
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 Daily average energy demand [kWh/day]
Time [day]
Vertigo Calibrated (Cold)
Measured_cold Simulated_cold
0 100 200 300 400 500 600
0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 Daily average energy demand [kWh/day]
Time [day]
Vertigo Calibrated (Heat)
Measured_heat Simulated_heat
RMSE Cold
[kWh/day]
RMSE Heat
[kWh/day]
MBE Cold
[kWh/day]
MBE Heat
[kWh/day]
Sportcentrum1
0 31 0.00 -0.55
Auditorium22 45 -1.23 -3.36
Vertigo
45 45 -2.42 -3.32
Matrix2
118 49 -80.00 -17.00
Helix2
45 236 -1.09 -18.20
Cyclotron1
26 33 0.33 -0.32
Kennispoort2N/A N/A N/A N/A
3.4. Second group 3.4.1. Part A
The second group exists out of two groups, group A is a group where not a lot will change, and/or buildings which are already renovated. Group B is the group with buildings that will be renovated before 2020.
Group A contains the Metaforum (former W-hal), Ceres and Catalyst, these buildings are already built/renovated and have been measured over the last 3 year. The measurement data of 2013 of these buildings will be used to estimate the balance of the ATES system, this year was chosen as it has the least startup errors. The result when group 2A is added is given in Part B
Group A contains the Spectrum building and Laplace, both of these buildings will undergo some changes which will influence the energy demand of the buildings. The Spectrum building is currently cooled using cooling machines, these machines are working fine, but a more efficient option with the chillers working as heat pumps is possible. A new building named Flux (former N-laag) will be using a lot more cold than heat, as a compensation project for this new building, the cooling machines of Spectrum will be changed for heat pumps, making it possible to reduce the gas use by 70%. This leaves the question where the compensation project for the Spectrum building itself is, as it has a larger cooling demand than heating as well. The heat normally provided to the building through the boilers, will now be provided by the heat pumps. A simple calculation can be done to calculate the extra heat that will be extracted from the ATES.
Spectrum used 258.000m3 of gas in 2011, this is comparable to 258000 * 9 = 2.322.000 kWh_thermal. 70% of this is 1.625.400kWh_thermal energy that will be extracted extra from the ATES in the future.
Table 7: ATES energy demand groups 1 and 2A
Laplace is built as the server building, when a computer needed a whole building for itself. Nowadays only a small amount of servers is left in this building, RE confirmed that the servers will be removed and placed at the High Tech Campus (HTC) this will decrease the cooling demand of this building.
The total energy demand of the existing buildings is shown in Table 7.
3.4.2. Part B
Group 2B are four buildings that will be renovated between now and the year 2020, these buildings are: Potentiaal, Hoofdgebouw, Gemini South and Gemini North. Of these buildings the façade area’s and floor surfaces are calculated using the buildings architectural drawings [appendix 2]. And the insulation parameters of the Dutch Building Code are used (Table 4).
Potentiaal will have a function change, this building will house students and expats, so the function will be residential.
In a decree of 2007 from the RE department [35], a few decisions for new and to be renovated buildings were made. The buildings concept should use the following steps in this order:
- Reduction of glass percentage max 20-40%
- Reduction of solar transmittance using a sunscreen or G-value <0.4
- Make use of the heat accumulating properties of the building.
- Use summer night ventilation - Reduction of internal heat gains
- Heat producing machines are placed in separate rooms
The window to wall ratio for the simulation of Potentiaal has been set at 30%, and the glazing has a G-value of 0.35. The full floor height is calculated as part of the zone, which would make use of the heat accumulating properties of the building. A user schedule has been programmed as nigh/daytime schedule: 7 days per week the day schedule from 8:00 till 23:00, the ventilation is doubled, as the students will be active, during night time the ACH is 1, which is exactly the minimum amount ventilation according to the Dutch Building Code. The internal gains have been set on 30W/m2, as the TU/e buildings of group 1 have an internal heat gain of around 30W/m2.
For the other 3 buildings the same values are used, the schedule is changed into the academic schedule. The results of these simulations have been modified using the percentage of cold from the vertigo building (90%) and the average percentage of heat (40%). After this modification, they have been added to Table 8.
Table 8: Group 2B simulated according to Dutch Building Code and with heat recovery
COLD
[MWh/year]
HEAT
[MWh/year]
Sportcentrum
Group 1
0 216
Auditorium
200 147
Vertigo
530 461
Matrix
831 263
Helix
1.018 2.701
Cyclotron
1.432 1.298
Kennispoort
397 0
Subtotal
4.408 5.086
CERES
Group 2A
28 48
Metaforum
321 379
Spectrum
2.780 1.625
Laplace
191 13
Catalyst
116 553
Subtotal
3.436 2.618
TOTAL
7.844 7.704
COLD
[MWh/year]
HEAT
[MWh/year]
Subtotal Group 1
+2A 7.844 7.704 Hoofdgebouw
Group 2B
1.547 64
Gemini Zuid
471 11
Gemini Noord
351 12
Potentiaal
248 63
Subtotal
2.617 150
TOTAL 10.461 7.854
3.5. Third group
The third group contains the Vestide tower (Residential), the student village (Residential), Flux (Academic) and Differ (Related business). As Table 8 shows, the trend is to require more cooling than heating.
The Vestide tower will provide space for 300 residential units, which is the same as the future residential building Potentiaal. The building is almost the same height as the Potentiaal building, and will have the same shape. Thus it is assumed that the Vestide tower will have the same energy demand as the Potentiaal building after its renovation.
Flux is built at the moment, it will be an academic building housing the electrical engineering and the applied physics departments. Since the Flux is imbalanced, the Spectrum building will be adjusted as a compensation project. The amount of thermal energy that Spectrum is compensating for is 1.265.400kWh, so the cooling demand for flux will be at least this amount higher than its heating demand.
Differ is a new research facility for the research into cleaner energy for the future. The building is not TU/e property, and thuse not the same designing principles have been applied. The amount of glazing is definitely more than 30%, and although the triple glazing should be placed in a direction to avoid direct sunlight entering the building it looks like the evening sun will be entering anyway.
The residential student village will contain multiple low two story buildings. As cooling is not a requirement for Dutch residential buildings, these buildings will probably not be cooled using the ATES. These assumptions are taken into account in Table 9.
Table 9: Cold and Heat demand from the ATES group 3
1 Not enough data available for more detailed estimations 4. DISCUSSION
4.1. Balance
Currently the ATES is imbalanced and according to the regulations all heat and cold should be used and cannot be emitted into the air. As a result of this there is currently an imbalance of around 6MWh/year when the cooling towers are not taken into account.
The imbalance for the buildings of group 1 (base load) is a lot better (46% cold 54% heat) than the original balance of 70%
cold versus 30% heat. When the renovated and new buildings of group 2 are added, the imbalance increases again. This might be due to the trend to increase the insulation (reducing glazing is increasing insulation as well) of the buildings, resulting in a lower heating demand and a higher cooling demand.
4.2. Simulation
Figure 7 shows an error in the measured data of the ATES for the Auditorium building, this was in 2011 the case for the month’s January, February and December. It was tried to use the data of 2012 instead, however during that year the gas sensor was not working, and no gas data was available. The 2011 data was used and as the average of the daily consumption was used it is expected that no large errors occur as a result of this measurement error.
Table 5 shows the calibration results of group 1, as visible not all buildings have been calibrated. This is due to some data issues.
4.2.1. Combined heating system
The Helix and Matrix buildings show large errors in the calibration results. These errors may be due to the measured data for the heating system. The buildings share the same heating system (boilers), the thermal energy is then transported from one building to the other. The data provided the total amount of gas for both buildings together, and the annual gas load for each building. Out of the annual gas load, the percentage of gas for each building was calculated and used to transform one set of hourly gas data into hourly data for each building. The error here might be that one building might need a lot of gas during a certain period, where the other might need it in another period, this would result in an error as the
“measured” and the simulated data will not line up. It should be noted that although Helix shows large errors, it is also a large consumer, so the error might only be a small amount of the total.
4.2.2. Simplified model
Another thing is that the Sportcentrum only uses the ATES to heat up the swimming pool, as the swimming pool is not simulated in the simplified model, it was chosen to remove the energy provided by the ATES from the measured data, and add it later to the simulated data.
4.2.3. Single zone
The cyclotron building uses a lot of heat and cold, this cannot be simulated using the single zone model as it would need to simulate heat and cold at the same time. To overcome this the average hourly cooling demand of the first week was used, and removed from the measured data for each week of the whole year. This amount of cold energy was later added back to the measured data as well as the simulated data.
4.2.4. Group 2A
Two buildings of the second group will undergo some changes during the renovation of the campus, these are Spectrum and Laplace. The heating for spectrum has been estimated using a change in the heating provided by the ATES.
The change in the Laplace building will be the removal of the servers which are located in this building. This will reduce the cooling load of the building, this has not been taken into account.
4.2.5. Group 2B
The second part of group 2 contains buildings that have been simulated to estimate what the heating and cooling load will be.
The current buildings properties like floor area have been taken as the starting point, and some values have been changed to fulfil the current building standards. If something would change in the design of these buildings, like the total floor area, the heating and cooling demands will change.
COLD
[MWh/year]
HEAT
[MWh/year]
Vestide toren1
Group 3
248 63
Student village1 0
Flux1 1.625
Differ1
TOTAL 1.773 63
4.2.6. Group 3
Not enough data was available to be able to simulate group 3, thus estimations have been used to check what the imbalance would be.
4.3. Solutions
4.3.1. Building level
Optimally the balance could be created by using more heat or less cold energy from the ATES. This however would result in an increase in electricity for cooling of the buildings, or a reduction in the insulation of the buildings. Other options might be to reduce the heat recovery in the buildings when a building is heated, this is still inefficient use of the energy as heat is brought into the air, but it will also decrease the amount of cooling needed [Table 10]. Free cooling would also help, this however could not be entered in the simulation model and is thuse not simulated.
4.3.2. Campus level
On campus level it would be possible to increase the comfort of the students and residents in the winter by deicing of the foot/cycle paths. To keep the snow detached from the path surface, 0.12kWh/m2 is required, for melting the snow 0.3kW/m2 would be needed [TAUW]. With an imbalance of around 4GWh/y, around 33km2 could be ice-free for an hour, divided by 2400 (100 days of cold in the Netherlands), makes this around 14.000m2 This type of system would already have been used in the Metaforum hall, but was canceled due to cost savings.
With the compact campus the buildings are all reachable by making use of the skyways. These could be heated in winter as well, while in summer the skyways can be cooled using free cooling.
Another solution would be to improve the TU/e community garden, which is a 200m2 area used as a kitchen garden (agriculture) [TU/e]. With the future residents on the TU/e campus, there might be an increase in the demand for kitchen gardens and with the large group of people of mixed descent it might be interesting to create a large horticulture at the TU/e campus. This could be heated using the ATES system and at the same time function as a research facility for reduction in energy demand for horticulture in the Netherlands.
Table 10: Group 2B simulated according to the Dutch Building Code without Heat Recovery
4.4. Data
Enough data is a key point in a research with real buildings, although measurement data is available for all TU/e buildings
over the last 10 years, not all data is specific enough. To be able to calculate the exact amount of heat or cold provided to a building, electricity provided to the installation has to be known. The total electricity can be found under the name HOK (Main Operation Power) but this is a sum of all installations like elevators, ventilation and heat pumps. Better would be to have separate sensors for the heat pump, which is the case for the Vertigo building but not for the other buildings. Since the electricity could not be split, the heat pumps were not really taken into account in the calculation of the total measured heat and or cold.
Other issues with the data would be the missing data for energy inserted into the ATES, and data about insulation values, ventilation amounts and heat recovery, these have been estimated during this research and adjusted during the calibration process. It is recommended to increase the amount of sensors to measure the energy supplied to the ATES as well as the energy demand of the heat pump installations separately.
4.5. Recommendations 4.5.1. Model resolution
During this research a low-resolution simulation program was used, it showed that the buildings on the TU/e are too complex to be simulated as a one zone model. It is thuse recommended to make a high-resolution model of all buildings to provide better insight in the energy flows in the buildings.
4.5.2. ATES measurement
Currently only the energy extraction from the ATES is measured, and the balance has to be found in the extracted data. Measuring the supplied energy to the ATES as well might give a proper view into the underground balance. This might help to control the extraction from the ATES and with that the balance could be created.
5. CONCLUSION
Although some possible options have been given to come to a balanced ATES, the goal, to provide a solution for the imbalanced TU/e real estate for the year 2020, has not been met.
This research shows that more data is needed to make correct estimations on the future balance.
The future campus 2020 will probably have a larger cooling than heating demand according to the research results in this article. This is mostly due to the fact that there is a trend in increasing the insulation value of the buildings shell. A possible solution has been provided in the shape of removing the Heat Recovery of buildings that will be renovated [Table 10]. It should however be mentioned that an increase in heating demand would also mean an increase in electricity use. This does not work in the same way for cooling, as cooling is used directly instead of via a heat pump.
The estimation on the thermal energy balance of the TU/e real estate does provide a guideline on how the energy demand will develop.
Possible solutions will have to be further researched.
COLD
[MWh/year]
HEAT
[MWh/year]
Subtotal Group 1
+ 2A 7844 7.704
Hoofdgebouw
Group 2B
530 224
Gemini Zuid
146 83
Gemini
Noord
161 104
Potentiaal
118 197
Total 8.899 8.312
ACKNOWLEDGEMENTS
I would like to thank Thijs Meulen of the Real Estate department and for his help with gathering data and information about the future campus. Also I would like to thank all the members of the TU/e CBPS department for their feedback and guidance.
REFERENCES
[1] H. Bouwmeester, “Wko 3x beter,” 2013.
[2] BodemenergieNL, “Visiedocument bodemenergieNL.” 2014.
[3] Dienst Huisvesting, Campus 2020. .
[4] L.F. Cabeza, “3.07 Thermal Energy Storage,” Earth Syst. Environ.
Sci., vol. 3, no. Solar Thermal Systems, pp. 211–253, 2012.
[5] H. G. Diersch and D. Bauer, Analysis, modeling and simulation of underground thermal energy storage (UTES) systems. Woodhead Publishing Limited, 2015, pp. 149–183.
[6] K. S. Lee, “A Review on Concepts, Applications, and Models of Aquifer Thermal Energy Storage Systems,” Energies, vol. 3, no. 6, pp. 1320–1334, Jun. 2010.
[7] H. O. Paksoy, Thermal Energy Storage for Sustainable Energy Consumption (Fundamentals, Case Studies and Design). Dordrecht:
Springer, 2007.
[8] H. O. Paksoy, “Energetic, Exergetic, Environmental and Sustainability aspects of Thermal Energy Storage systems,” 2007.
[9] United Nations, “United Nations Framework Convention on Climate Change,” 1992. [Online]. Available:
https://unfccc.int/essential_background/items/6031.php.
[10] United Nations, “Kyoto protocol to the United Nations Framework Convention on Climate Change.” 1998.
[11] A. L. Snijders and M. M. van Aarssen, “Big is beautiful ?,” pp. 83–
88, 2003.
[12] G. Hardin, “The Tragedy of the Commons,” Science (80-. )., vol.
162, no. December, pp. 1243–1248, 1968.
[13] Infrastructuur en Milieu, “Waterwet,” 2014. [Online]. Available:
http://wetten.overheid.nl/BWBR0025458/Hoofdstuk1/1/Artikel11/g eldigheidsdatum_12-03-2014.
[14] Infrastructuur en Milieu, “Wet Bodembescherming,” 2014. [Online].
Available:
http://wetten.overheid.nl/BWBR0003994/geldigheidsdatum_12-03- 2014.
[15] “Staatsblad van het Koninkrijk der Nederlanden,” vol. 112, pp. 1–
119, 2013.
[16] Provincie Noord-Brabant, Provinciaal Waterplan Noord-Brabant 2010-2015. 2010.
[17] European Union, “Energy roadmap 2050,” 2011.
[18] J. Becvar, E. Djunaedy, J. Hensen, M. Radosevic, and A. Yahiaoui,
“FAGO report 04.89,” Eindhoven, 2005.
[19] Dienst Huisvesting, “Energiejaarverslag 2003,” pp. 1–19, 2004.
[20] Dienst Huisvesting, “Energiejaarverslag 2004,” 2005.
[21] Dienst Huisvesting, “Energiejaarverslag 2005,” pp. 1–25, 2006.
[22] Dienst Huisvesting, “Energiejaarverslag 2006,” pp. 1–26, 2007.
[23] Dienst Huisvesting, “Energiejaarverslag 2007,” 2008.
[24] Dienst Huisvesting, “Energiejaarverslag 2008,” 2009.
[25] Dienst Huisvesting, “Energiejaarverslag 2009,” 2010.
[26] Dienst Huisvesting, “Energiejaarverslag 2010,” pp. 1–50, 2011.
[27] D. Huisvesting, “Management review MeerJaren Afspraak Energie Efficiency TU / e Inhoudsopgave,” no. april, 2014.
[28] Eindhoven University of Technology, “Van universiteitscampus naar Science Park.”
[29] Dienst Huisvesting, “Gebouwinformatie Campus Tu/e.” [Online].
Available:
http://w3.tue.nl/nl/diensten/dh/campus_2020/gebouwinformatie_ca mpus_tue/athene/.
[30] Iso, “Thermal performance of buildings — Calculation of energy use for space heating and cooling - ISO 13790:2005,” 2005.
[31] F. F. M. Soons, “A computational model for evaluating a district heating system using a biomass heat source and thermal energy storage,” Eindhoven University of Technology, 2014.
[32] “Calorische waarde, bovenwaarde en onderwaarde Calorische waarde.” [Online]. Available:
http://www.energieleveranciers.nl/energie/begrippen/calorische- waarde.
[33] Carrier SCS, “Watergekoelde vloeistofkoelmachines met schroefcompressoren.”
[34] M. Liu, D. E. Claridge, N. Bensouda, M. Heinemeier, S. Uk Lee, and G. Wei, “Manual of Procedures for Calibrating Simulations of Building Systems,” no. October, p. 53, 2003.
[35] Y. de Weerd, Besluit MT Dienst Huisvesting. 2007, p. 11.
Appendix 1
Modelica Dymola computational model code
Appendix 2
Modelica inputdata per building acquired from:
Building envelope: Architectural drawings Indoor climate settings: Estimations
Schedules: Measured data
Number of fl oors:
Roof area:
Floor height:
Facade area N/E/S/W:
Wtw ratio N/E/S/W:
G-factor glazing:
U-value walls:
U-value windows:
U-value roof:
U-value fl oor:
Heat Recovery Name:
Number:
Year of construction:
Year of renovation:
ACH min/max:
Internal gains:
Day length:
Day start:
Heating setpoint min/max:
Cooling setpoint:
Useful fl oor area:
Sportcentrum 0700
1967-2001 0.5/2 [m
3/m
2] 15 [W/m
2] (15/24)*100 [%]
6 [h]
19.5/21 [°C]
50 (no cooling) [°C]
9272 [m
2]
4
8530 [m
2] 7 [m]
1400/870/1400/870 [m
2] 0.07/0.18/0.21/0.21 0.7
0.4 [W/m
2*K]
1.53 [W/m
2*K]
0.25 [W/m
2*K]
0.4 [W/m
2*K]
75 % fresh air
0 100 200 300 400 500 600
15 Jan 2011 0:00 16 Jan 2011 0:00 17 Jan 2011 0:00 18 Jan 2011 0:00 19 Jan 2011 0:00 20 Jan 2011 0:00 21 Jan 2011 0:00
Energy demand [kWh_thermal]
Time [day]
Sportcentrum Week
Gas_heat
0 50 100 150 200 250 300 350 400 450 500
Energy demand [kWh_thermal]
Sportcentrum Day (19-Jan-2011)
Gas_heat
Name:
Number:
Year of construction:
Year of renovation:
ACH min/max:
Internal gains:
Day length:
Day start:
Heating setpoint min/max:
Cooling setpoint:
Useful fl oor area:
Number of fl oors:
Roof area:
Floor height:
Facade area N/E/S/W:
Wtw ratio N/E/S/W:
G-factor glazing:
U-value walls:
U-value windows:
U-value roof:
U-value fl oor:
Heat Recovery Auditorium
1300 1966 1995 3/5 [m
3/m
2] 32 [W/m
2] (13/24)*100 [%]
7 [h]
19.5/21 [°C]
22.5 (no cooling) [°C]
5859 [m
2]
7
4406 [m
2] 5.2 [m]
1479/853/1479/853 [m
2] 0.35/0.54/0.35/0.54 0.7
0.4 [W/m
2*K]
1.53 [W/m
2*K]
0.4 [W/m
2*K]
0.4 [W/m
2*K]
55% fresh air
0 50 100 150 200 250 300 350 400 450
19 Nov 2011 0:0020 Nov 2011 0:0021 Nov 2011 0:0022 Nov 2011 0:0023 Nov 2011 0:0024 Nov 2011 0:0025 Nov 2011 0:00
Energy demand [kWh_thermal]
Time [day]
Auditorium Week
Total cold Total heat
0 200 400 600 800 1000 1200 1400
Energy demand [kWh_thermal]
Auditorium Day (19-Jan-2011)
Total heat Total cold
Number of fl oors:
Roof area:
Floor height:
Facade area N/E/S/W:
Wtw ratio N/E/S/W:
G-factor glazing:
U-value walls:
U-value windows:
U-value roof:
U-value fl oor:
Heat Recovery Name:
Number:
Year of construction:
Year of renovation:
ACH min/max:
Internal gains:
Day length:
Day start:
Heating setpoint min/max:
Cooling setpoint:
Useful fl oor area:
Vertigo 5100 1965 2002
0.5/2.5 [m
3/m
2] 30 [W/m
2] (13/24)*100 [%]
6 [h]
19.5/21 [°C]
22.5 [°C]
16.098 [m
2]
11
1454 [m
2] 5.2 [m]
1790/3085/1830/3150 [m
2] 0.73/0.59/0.71/0.61 0.35
0.4 [W/m
2*K]
1.53 [W/m
2*K]
0.25 [W/m
2*K]
0.4 [W/m
2*K]
70% fresh air
0 100 200 300 400 500 600 700 800
15 Jan 2011 0:0016 Jan 2011 0:0017 Jan 2011 0:0018 Jan 2011 0:0019 Jan 2011 0:0020 Jan 2011 0:0021 Jan 2011 0:00 Energy demand [kWh_thermal]
Time [days]
Vertigo Week
Total_Heat Total_Cold
0 100 200 300 400 500 600 700
Energy demand [kWh_thermal]
Vertigo Day (19 januari 2011)
Total_Heat Total_Cold
Name:
Number:
Year of construction:
Year of renovation:
ACH min/max:
Internal gains:
Day length:
Day start:
Heating setpoint min/max:
Cooling setpoint:
Useful fl oor area:
Number of fl oors:
Roof area:
Floor height:
Facade area N/E/S/W:
Wtw ratio N/E/S/W:
G-factor glazing:
U-value walls:
U-value windows:
U-value roof:
U-value fl oor:
Heat Recovery Matrix
5300 1960 2/8[m
3/m
2] 10[W/m
2] (14/24)*100 [%]
4[h]
19/21[°C]
23[°C]
7000[m
2]
4
3059[m
2] 2.05[m]
805/147/805/147[m
2] 0.8/0.3/0.3/0.3 0.7
0.4[W/m
2*K]
1.53[W/m
2*K]
0.4[W/m
2*K]
0.4[W/m
2*K]
50% fresh air
0 100 200 300 400 500 600 700 800 900 1000
19 Nov 2011 0:00 20 Nov 2011 0:00 21 Nov 2011 0:00 22 Nov 2011 0:00 23 Nov 2011 0:00 24 Nov 2011 0:00 25 Nov 2011 0:00 Energy demand [kWh_thermal]
Time [day]
Matrix Week
Total cold Total heat
0 100 200 300 400 500 600 700
Energy demand [kWh_thermal]
Matrix Day (19-Jan-2011)
Total cold Total heat
