Ground source heat pump optimization and development of ground heat exchangers

Ground is an energy resource which has more appropriate and stable temperatures than air.Ground source heat pumps (GSHP) were developed to use ground energy for residentialheating. Using ground as a heat source is more expensive than air and water. The mostimportant part of a GSHP is the ground heat exchanger (GHE) that consists of pipes buried inthe soil and is used for transferring heat between the soil and the heat exchanger of the GSHP.Soil composition, density, moisture and burial depth of pipes affect the size of a GHE.Since the thermal properties of soil change in time, it is difficult to calculate the size of GHEproperly. Design of GSHP systems in different regions of US and Europe is performed usingdata from an experimental model. However, there are many more techniques including somecomplex calculations for sizing GHEs.In literature, there are two kinds of analytical approaches. The first one is the Kelvin LinearSource Theory and the other one is the Cylindrical Source Theory. In addition, there are manytwo or three dimensional steady and time dependent numerical techniques. Somemodifications must be done to Cylindrical Source and Kelvin Linear Source theories toinclude effects of adjacent pipes. A simplification of boundary conditions to solve equationsanalytically causes some error on results especially shorter simulation times. Kelvin LinearSource and Cylindrical Source theories find only symmetrical soil temperature distributionsaround the pipe. Analytical models do not consider the temperature change of soil by depthand the surface effects such as radiation, convection and surface cover are not included. Inorder to simulate all the weather conditions, all these effects must be included in the model.To find three dimensional temperature distributions in the soil, a new model including all theweather effects in real life was suggested. Heat transfer in the soil is a time dependent, threedimensional heat conduction. Temperature gradient along the pipe axis is so small that it canbe neglected and the heat conduction equation can be solved using dynamical boundaryconditions in two-dimensional geometry. By means of conservation of energy, thetemperature distribution of the fluid along the pipe was calculated and used for linking twodimensional solution domains. Because of the complexity of the boundary conditions, the heatconduction equation has been solved numerically using Alternating Direction Implicit (ADI)Finite Difference formulation. ADI method is stable for every time step and grid size and theresulting matrix system is tri-diagonal. Tri-diagonal matrix systems can be solved easily usingthe Thomas algorithm. For this purpose, software was developed in MATLAB environmentand the effects of solution parameters on the results were investigated. The simulation resultswere acceptable when a mesh size of 0.05 m in x and y directions, 1 m in z direction and 1800s as time step were used. A simulation was carried out with simplified boundary conditions tocompare results with Mei?s work. Results show good agreement with Mei?s work andexperimental data. A parametrical study on burial depth and distance between pipes wasperformed to obtain thermally efficient GHE dimensions. For continuous operation of aGSHP, the burial depth and the distance between the pipes are suggested as 1.6 m and 3 mrespectively.An experimental study was carried out to show the validity of the model. A GSHP having 4kW heating and 2.7 kW cooling was used. A ground heat exchanger consists of three parallelpipes with 40 m length and ½? diameter buried in soil at 1.8 m depth. The distance betweenthe parallel pipes is 3 m. An experimental GSHP system was installed at Yıldız TechnicalUniversity Davupaşa Campus on 800 m2 area with no surface cover. Temperature data werecollected during 37 days using 32 T-type thermocouples buried in soil horizontally andvertically at various distances from the pipe center and 2 T-type thermocouples at the inletand outlet of the ground heat exchanger. Experimental and numerical simulation results usingexperimental water inlet temperatures were compared. All the thermocouples were connectedto a 64 channel PLC system capable of saving data of hourly temperature measurements for 8days. Collected fluid inlet temperatures were used in the numerical simulation and the fluidoutlet temperatures were calculated for 910 hours. The maximum difference between thenumerical results and the experimental data is 10.03%. The temperature distribution in the soilwas calculated and compared with experimental data also. Both horizontal and verticaltemperature profiles matched the experimental data well.In the last chapter of the thesis, a thermo-economical analysis was accomplished for parallelpipe horizontal and vertical U-tube ground heat exchangers using finite time thermodynamics.Installation and operating costs were taken into account as total cost and a reference functionwas described. The effects of soil thermal conductivity, number of pipes, thermal capacity,pipe diameter and burial depth on reference function were examined. Because of higherinstallation cost of U-tube GHEs, it is recommended to use parallel pipe GHE for the samethermal capacity. To make the installation cost same for parallel pipe horizontal and verticalU-tube GHEs, the borehole cost must be $23/m while it is $62/m today. Additionally, usingtwo or more pipes instead of a single long pipe is suggested because the reference functiongoes to zero for all diameters with increasing thermal capacity.Keywords: Ground source heat pump, parallel pipe horizontal ground heat exchanger, U-tubeground heat exchanger, numerical solution, thermo-economic optimization.JURY:1. Prof. Dr. Galip TEMİR (Supervisor) Date 16.02.20062. Prof. Dr. İsmail TEKE Page: 1443. Prof. Dr. Hasan A. HEPERKAN4. Prof. Dr. Cem PARMAKSIZOĞLU5. Prof. Dr. Seyhan ONBAŞIOĞLU