Projects per year
Abstract
The purpose of the work which is presented in this thesis is to develop and investigate simulation models for building integrated heating and cooling systems. These models can be used to find the thermal properties of building integrated systems and their influence on the thermal indoor climate and energy consumption in the building. A number of simulation models are developed with different level of detail. The simple models can be used for early estimates in the design phase of a building of the influence of using building integrated heating and cooling systems in buildings, while the models with more details are especially suited for product development and research purposes.
In this thesis two types of building integrated heating and cooling systems are presented and investigated. The first is floor heating, which is the most often used type of heating system in Danish singlefamily houses. The second is thermo active components – which is a relatively new technology. Thermo active components represents a way of cooling offices by using embedded pipes in the thermal mass of the building – typically in the form of concrete – where cool water is circulated to remove the heat from the concrete and consequently cool the offices in the building. Finally, as an extension to the investigation on floor heating systems, a third type of building integrated system has been included, as an investigation of thermal energy storage of solar energy in a slabongrade floor of a singlefamily house.
Initially, after defining the scientific approach and purpose of the work, a literature evaluation and state of the art review has been completed to establish the basis for the work.
The main element in this project is the development and implementation of simulation models of building integrated heating and cooling systems. For this purpose, two simulation models have been developed; one for floor heating, called FHSim, and one for thermo active components, called TASim. The models are both dynamical simulation models of a room with building integrated heating and/or cooling. The models include heat transfer, heat storage and temperature distribution in the building elements, which include floors, walls, ceiling, windows and elements with building integrated heating and cooling system. The room model includes detailed calculation of the heat transfer, which is split into radiation and convection. Radiation is included both as short wave solar gains on the surfaces and long wave thermal radiation between surfaces based on the view factor between the surfaces. Further, there are models for ventilation, infiltration and venting, as well as models for control of the heating/cooling system and input of weather data. The simulation models FHSim and TASim are for the most parts identical except for the models of the building integrated systems and controls hereof.
The models of the building constructions are based on the numerical Finite Control Volume method. Only heat transfer with constant material properties is included in the models.
The investigation of floor heating systems is the first of the two main parts in the thesis.
Initially, a twodimensional simulation model of a slabongrade floor with floor heating is developed and validated against measurements from a house in Bromölla, Sweden. Here it is found that by using the characteristic dimension of the floor, which is defined as the area divided by half the perimeter, as the width of the model, the numerical model will give results that are in close agreement to threedimensional measurements. This is the case even for the very narrow building used for the validation, which is very influenced by threedimensional
conditions. The use of the characteristic dimension to simplify the threedimensional heat flow problem to a twodimensional one has previously been demonstrated in the literature only for buildings without floor heating.
Based on the measurements and the validated simulation model of the slabongrade floor, the importance of using correct implementation of the floor heating system is demonstrated using both the actually measured temperature and fixed temperatures in the pipe. Here it is shown, that there can easily be large differences in the results using a fixed temperature rather than correct temperatures. This means that in order to find the correct heat loss to the ground from the floor, it is necessary to use a dynamical simulation where the pipe temperature is included and based on the actual energy demand in the room. For these analyses the simulation models introduced in this work are ideal.
The twodimensional simulation model of the slabongrade floor is used for investigating the effect on the energy consumption of improved insulation in floor and foundation. Here it is found that compared to a model without floor heating, the fact that the floor is heated means that the heat loss to the ground will be penalized through higher relative heat loss to the ground in case of poor insulation. In the course of this investigation it is also found that the value of the linear thermal transmittance of the foundation is influenced by the presence of
floor heating.
At the same time it is demonstrated that both the insulation under the floor as well as in the foundation is important to minimize the heat loss to the ground.
While being able to accurately model the conditions in a slabongrade floor with floor heating, the twodimensional model is very timeconsuming both with respect to defining the geometry of the floor and simulation time. This means that the model is an expert tool which can be used for research purposes and product development of the design of the floor construction. However, it is not a tool which can be used in the design phase, where there are both many unknown factors and perhaps a need for several simulations. A series of simplified models are therefore tested, both one and twodimensional finite control volume models and thermal network models using lumped resistances and capacities. Of special interest is that the simplest RCthermal network model where the linear thermal transmittance of the foundation is included, yield results for both energy consumption and heat loss to the ground that are close to those found by the detailed twodimensional model. Further, it has been shown that using an electrical inclusion of the floor heating pipe is not sufficient to model hydronic floor heating. Among the reasons is that the electrical implementation is unable to include the temperature of the pipe in the model, which has been shown to impact the energy consumption through different thermal climate in the room.
In the chapter on thermo active components a different approach is used than for the chapter on floor heating. Here emphasis is placed on two measurement setups with the purpose of testing different ways of turning a prefabricated hollow core concrete deck into a thermo active component. Two decks are tested. The simplest of these decks is constructed simply by placing the pipe directly in the cavities of the deck, which is tested in a simple setup. This method is of course not as efficient as the second type, where the pipe is integrated in the concrete. However, the cooling capacity is still large enough to cool the office in a consciously designed building or as a supplement to a natural ventilation system.
In the second test setup, two decks are used as floor and ceiling of a room. The purpose of this investigation is, under controlled conditions, to find the cooling capacity under stationary conditions and the temperature variations under dynamic conditions, where the heat load in the room varies during the day and the flow in the pipe is only turned on during the night. The results clearly demonstrates the possibilities of using prefabricated hollow core concrete decks as thermo active components to cool the office, even for heat loads as high as 6070W/m² during stationary conditions. This should be compared to around 2530W/m² for the deck with the pipe placed in the cavities.
The stationary cooling capacity is found using different combinations of room air set point temperature and supply temperature to the pipe of the thermo active deck. A linear correlation between the cooling capacity and the temperature difference between fluid temperature and room temperature is found, which means that the cooling capacity coefficient expressed as cooling capacity per area and temperature difference is constant. This was also found for the first simple setup. In the second test setup, the measurements could be used to calculate the cooling capacity of the ceiling and floor surfaces individually – and for this setup it was found that the ceiling surface had a cooling capacity five times larger than for the floor surface.
Besides from the cooling capacity, the thermal conditions in the room with respect to surface (radiant) temperatures and vertical air temperature distribution can be investigated to find the operational conditions in the room with the mainly radiant cooling system.
The dynamic conditions in the room has been demonstrated by using a heat load which is high during the day and low during the night and a flow in the pipes which is only on during the night. Using this setup, the maximum heat load in the room can be found if the room temperature should not surpass the comfort range.
The test setup has been designed in such a way that it can subsequently be used for testing the system in combination with suspended ceilings or even operable ceilings, which can be used to manually control the cooling from the deck, as well as different types of ventilation systems and control strategies of the flow in the pipes.
For both stationary and dynamic conditions, the simulation model TASim has been shown to satisfactorily reproduce the results from the measurements. This means that also the room air model in FHSim has been validated.
The third part of the investigation is the use of thermal energy storage of solar energy in a slabongrade floor with two concrete decks, using the lowest deck for energy storage. It is found that even for a house with an already low energy consumption, it is possible to lower this even more through the using the heat storage.
Based on the investigations in this thesis, it is concluded that it is possible to model building integrated heating and cooling systems to find both energy consumption and thermal indoor climate. The implementation of such models in building energy simulation programs will represent an advance towards a more realistic implementation of building integrated heating and cooling systems.
In this thesis two types of building integrated heating and cooling systems are presented and investigated. The first is floor heating, which is the most often used type of heating system in Danish singlefamily houses. The second is thermo active components – which is a relatively new technology. Thermo active components represents a way of cooling offices by using embedded pipes in the thermal mass of the building – typically in the form of concrete – where cool water is circulated to remove the heat from the concrete and consequently cool the offices in the building. Finally, as an extension to the investigation on floor heating systems, a third type of building integrated system has been included, as an investigation of thermal energy storage of solar energy in a slabongrade floor of a singlefamily house.
Initially, after defining the scientific approach and purpose of the work, a literature evaluation and state of the art review has been completed to establish the basis for the work.
The main element in this project is the development and implementation of simulation models of building integrated heating and cooling systems. For this purpose, two simulation models have been developed; one for floor heating, called FHSim, and one for thermo active components, called TASim. The models are both dynamical simulation models of a room with building integrated heating and/or cooling. The models include heat transfer, heat storage and temperature distribution in the building elements, which include floors, walls, ceiling, windows and elements with building integrated heating and cooling system. The room model includes detailed calculation of the heat transfer, which is split into radiation and convection. Radiation is included both as short wave solar gains on the surfaces and long wave thermal radiation between surfaces based on the view factor between the surfaces. Further, there are models for ventilation, infiltration and venting, as well as models for control of the heating/cooling system and input of weather data. The simulation models FHSim and TASim are for the most parts identical except for the models of the building integrated systems and controls hereof.
The models of the building constructions are based on the numerical Finite Control Volume method. Only heat transfer with constant material properties is included in the models.
The investigation of floor heating systems is the first of the two main parts in the thesis.
Initially, a twodimensional simulation model of a slabongrade floor with floor heating is developed and validated against measurements from a house in Bromölla, Sweden. Here it is found that by using the characteristic dimension of the floor, which is defined as the area divided by half the perimeter, as the width of the model, the numerical model will give results that are in close agreement to threedimensional measurements. This is the case even for the very narrow building used for the validation, which is very influenced by threedimensional
conditions. The use of the characteristic dimension to simplify the threedimensional heat flow problem to a twodimensional one has previously been demonstrated in the literature only for buildings without floor heating.
Based on the measurements and the validated simulation model of the slabongrade floor, the importance of using correct implementation of the floor heating system is demonstrated using both the actually measured temperature and fixed temperatures in the pipe. Here it is shown, that there can easily be large differences in the results using a fixed temperature rather than correct temperatures. This means that in order to find the correct heat loss to the ground from the floor, it is necessary to use a dynamical simulation where the pipe temperature is included and based on the actual energy demand in the room. For these analyses the simulation models introduced in this work are ideal.
The twodimensional simulation model of the slabongrade floor is used for investigating the effect on the energy consumption of improved insulation in floor and foundation. Here it is found that compared to a model without floor heating, the fact that the floor is heated means that the heat loss to the ground will be penalized through higher relative heat loss to the ground in case of poor insulation. In the course of this investigation it is also found that the value of the linear thermal transmittance of the foundation is influenced by the presence of
floor heating.
At the same time it is demonstrated that both the insulation under the floor as well as in the foundation is important to minimize the heat loss to the ground.
While being able to accurately model the conditions in a slabongrade floor with floor heating, the twodimensional model is very timeconsuming both with respect to defining the geometry of the floor and simulation time. This means that the model is an expert tool which can be used for research purposes and product development of the design of the floor construction. However, it is not a tool which can be used in the design phase, where there are both many unknown factors and perhaps a need for several simulations. A series of simplified models are therefore tested, both one and twodimensional finite control volume models and thermal network models using lumped resistances and capacities. Of special interest is that the simplest RCthermal network model where the linear thermal transmittance of the foundation is included, yield results for both energy consumption and heat loss to the ground that are close to those found by the detailed twodimensional model. Further, it has been shown that using an electrical inclusion of the floor heating pipe is not sufficient to model hydronic floor heating. Among the reasons is that the electrical implementation is unable to include the temperature of the pipe in the model, which has been shown to impact the energy consumption through different thermal climate in the room.
In the chapter on thermo active components a different approach is used than for the chapter on floor heating. Here emphasis is placed on two measurement setups with the purpose of testing different ways of turning a prefabricated hollow core concrete deck into a thermo active component. Two decks are tested. The simplest of these decks is constructed simply by placing the pipe directly in the cavities of the deck, which is tested in a simple setup. This method is of course not as efficient as the second type, where the pipe is integrated in the concrete. However, the cooling capacity is still large enough to cool the office in a consciously designed building or as a supplement to a natural ventilation system.
In the second test setup, two decks are used as floor and ceiling of a room. The purpose of this investigation is, under controlled conditions, to find the cooling capacity under stationary conditions and the temperature variations under dynamic conditions, where the heat load in the room varies during the day and the flow in the pipe is only turned on during the night. The results clearly demonstrates the possibilities of using prefabricated hollow core concrete decks as thermo active components to cool the office, even for heat loads as high as 6070W/m² during stationary conditions. This should be compared to around 2530W/m² for the deck with the pipe placed in the cavities.
The stationary cooling capacity is found using different combinations of room air set point temperature and supply temperature to the pipe of the thermo active deck. A linear correlation between the cooling capacity and the temperature difference between fluid temperature and room temperature is found, which means that the cooling capacity coefficient expressed as cooling capacity per area and temperature difference is constant. This was also found for the first simple setup. In the second test setup, the measurements could be used to calculate the cooling capacity of the ceiling and floor surfaces individually – and for this setup it was found that the ceiling surface had a cooling capacity five times larger than for the floor surface.
Besides from the cooling capacity, the thermal conditions in the room with respect to surface (radiant) temperatures and vertical air temperature distribution can be investigated to find the operational conditions in the room with the mainly radiant cooling system.
The dynamic conditions in the room has been demonstrated by using a heat load which is high during the day and low during the night and a flow in the pipes which is only on during the night. Using this setup, the maximum heat load in the room can be found if the room temperature should not surpass the comfort range.
The test setup has been designed in such a way that it can subsequently be used for testing the system in combination with suspended ceilings or even operable ceilings, which can be used to manually control the cooling from the deck, as well as different types of ventilation systems and control strategies of the flow in the pipes.
For both stationary and dynamic conditions, the simulation model TASim has been shown to satisfactorily reproduce the results from the measurements. This means that also the room air model in FHSim has been validated.
The third part of the investigation is the use of thermal energy storage of solar energy in a slabongrade floor with two concrete decks, using the lowest deck for energy storage. It is found that even for a house with an already low energy consumption, it is possible to lower this even more through the using the heat storage.
Based on the investigations in this thesis, it is concluded that it is possible to model building integrated heating and cooling systems to find both energy consumption and thermal indoor climate. The implementation of such models in building energy simulation programs will represent an advance towards a more realistic implementation of building integrated heating and cooling systems.
Original language  English 

Place of Publication  Kgs. Lyngby 

Publisher  Technical University of Denmark 
Number of pages  239 
ISBN (Print)  8778771522 
Publication status  Published  Nov 2004 
Series  BYGRapport 

Number  R091 
Fingerprint Dive into the research topics of 'Modelling building integrated heating and cooling systems'. Together they form a unique fingerprint.
Projects
 1 Finished

Simuleringsmodel af bygninger med bygningsintegrerede varme og kølesystemer
Weitzmann, P., Svendsen, S., Olesen, B. W., Jóhannesson, G. & Olsen, L.
01/02/2001 → 16/11/2004
Project: PhD