Article preview
- Abstract
- Introduction
- Section snippets
- References (32)
- Cited by (19)
- Recommended articles (6)
Applied Thermal Engineering
Volume 134,
April 2018
, Pages 298-309
Author links open overlay panel
Abstract
Thermal response tests (TRTs) are often performed on vertical borehole heat exchangers, which are used in ground-source heat pump systems. Conventional analysis methods for TRTs provide estimates of the ground thermal conductivity and the effective borehole resistance. Both parameters are required in the design of GSHP systems. A parameter estimation method is proposed to analyze the TRT data using the temperature derivative in addition to the usual transient temperature curve. The method provides an estimate for the ground volumetric heat capacity in addition to the ground thermal conductivity and effective borehole resistance. Without the derivative curve parameter estimation gives unrealistic values of ground volumetric heat capacity, if the three unknown parameters are to be determined. The proposed method is demonstrated on TRT data sets from three boreholes.
Introduction
Ground source heat pumps (GSHP) systems are an efficient method to heat and cool buildings with low maintenance costs [1]. These systems often use vertical borehole heat exchangers (BHE) where a circulating fluid transports heat between the heat pump and the ground. The most frequently used configuration in the United States is a single U-tube within the BHE where the entering fluid flows down through one pipe and flows up through the other pipe. The space between the outside walls of the U-tube and the borehole wall is usually filled with grout, which serves as a barrier for the vertical movement of water and contaminants. Exceptions occur, for example, in Sweden where the common practice is to allow groundwater to fill the borehole and surround the U-tube [2], [3]. Other pipe configurations include double U-tube and coaxial pipe-within-pipe arrangements [4].
The ground thermal conductivity is an important parameter in the design of borehole heat exchangers. For commercial applications and other larger installations, a thermal response test (TRT) is performed on a test borehole to estimate the ground thermal conductivity. In addition, the TRT provides an estimate of the borehole resistance, which represents the thermal resistance between the circulating fluid and the borehole wall. The TRT was first proposed by Mogensen [5] as a method to estimate ground thermal conductivity.
The equipment for a TRT typically consists of an electrical heater, pump, flow meter, temperature sensors and data acquisition system. The heated fluid (water or water/antifreeze mixture) is pumped through a closed loop, which includes the heater and the U-tube. At the start of a test the borehole and fluid-filled U-tube are initially at the undisturbed ground temperature. With the fluid circulating, the heater is switched on, and the heat input rate is held nearly constant with time. A 48-h period of heating is typical. While the BHE discharges heat to the surrounding ground, the fluid inlet and outlet temperatures to the U-tube and the flow rate are recorded along with the electric power to the heater. Eklöf and Gehlin [6] and Austin et al. [7] built early portable TRT units for field tests.
Early models of borehole heat transfer treated the borehole as a line source of heat as explained by Ingersoll and Plass [8] with a solution given by Carslaw and Jaeger [9]. The mean of the inlet and outlet fluid temperatures is used as an estimate for the average temperature along the length of the borehole. This approximation assumes the heat transfer rate is uniform along the length of the borehole length. More sophisticated models treat the finite diameter of the borehole and the thermal storage of the circulating fluid [10], [11]. Yang et al. [12] provide a review of various borehole models. Spitler and Gehlin [13] have written a historical review of the measurement and analysis techniques for TRTs.
Past studies have combined numerical models of borehole heat transfer with methods to automatically adjust model parameters to minimize sum of the square errors (SSE) between the model and the measured temperatures. Marcotte and Pasquier [14] chose to estimate the ground thermal conductivity, ks, and the effective borehole thermal resistance, Rb∗, as the two unknown parameters, while all other parameters were held constant at independently evaluated values. Shonder and Beck [10] and Austin et al. [7] estimated the ground thermal conductivity along with the grout thermal conductivity. With the estimated value of grout thermal conductivity, the effective borehole resistance can be calculated, if other parameters are known. Wagner and Clauser [15] chose to estimate the ground thermal conductivity, ks, and volumetric heat capacity, cs, as the two unknowns. Li and Lai [16] used an analytical borehole model with a parameter estimation method to evaluate the ground thermal conductivity and the effective borehole thermal resistance.
When treating the ground thermal conductivity and effective borehole thermal resistance as the two unknowns, Marcotte and Pasquier [14] find the two-dimensional surface of the root-mean-square-error (RMSE), or equivalently SSE, displays a long elliptically shaped trough or valley with a global minimum. Similarly shaped surfaces for RSME (or SSE) are reported by Spitler and Gehlin [13] and Li and Lai [16]. In addition, Wagner and Clauser [15] found comparable RSME surfaces with ks and cs as the two estimated parameters. When the number of unknown parameters is increased beyond two, Austin et al. [7] and Spitler and Gehlin [13] report that parameter estimation methods assign unrealistic values to some parameters such as the volumetric heat capacities of the ground and grout. Li and Lai [16] report that the three parameters of ground thermal conductivity, borehole thermal resistance and ground thermal diffusivity cannot be reliably found simultaneously through parameter estimation.
All of the above TRT studies have focused on minimizing errors between the model and measured temperatures. Petroleum engineers [17], [18] have found in analogous tests on oil wells that test interpretation methods can be improved by using the derivative of the measured variable. In a producing oil well the measured downhole pressure corresponds to the mean temperature in a TRT. The oil production rate corresponds to the input (or extracted) heat rate in a TRT. Well tests in the petroleum industry are often called pressure transient tests, because the downhole pressure is recorded with time. Bourdet et al. [17], [18] introduce the use of the derivative of pressure in the analysis of a pressure transient test. The derivative is taken with respect to the natural logarithm of time. Beier and Smith [19] applied the corresponding temperature derivative to estimate the required time to reach the steady heat-flux period (quasi-steady-state period) during a TRT.
The purpose of this paper is to demonstrate the usefulness of the temperature derivative in the analysis of TRT data sets. In particular, estimation of the ground volumetric heat capacity is examined in detail. This heat capacity is estimated in addition to the two usual parameters of ground thermal conductivity and effective borehole resistance. The ground heat capacity is of interest, because the estimate of effective borehole resistance depends on the value of the ground heat capacity. Also, the ground heat capacity is needed in the design of borehole fields, because its value affects the ground thermal diffusivity, which plays a role in the required spacing between boreholes.
Section snippets
Analysis of late-time period of TRT
Before focusing on the temperature derivative curve, it is useful to recall how the ground thermal conductivity, ks, and effective borehole thermal resistance, Rb∗, are evaluated in a conventional analysis where the vertical borehole is treated as a line source of heat. This information provides background for more sophisticated analyses of TRT data sets in subsequent sections. Following Carslaw and Jaeger [9], Ingersoll and Plass [8], Beier and Smith [20] and Spitler and Gehlin [13] the mean
Temperature derivative
To interpret the temperature and temperature derivative data throughout a TRT one must have a model that applies for the entire testing period. The model used in this paper is a 1D composite model [19] of the borehole where the U-tube is represented as a single pipe, which is concentric with the borehole wall. The single pipe has an equivalent radius, which can be related to the effective borehole resistance as written in Eq. (A.23) in Appendix A. A ring of grout fills the space between the
Components of borehole resistance
Some concepts concerning borehole resistance are reviewed here, which provide needed background information for analyzing specific TRT data sets later.
Javed and Spitler [24] distinguish between local and effective borehole resistances. The local borehole resistance, Rb, at a given depth is based on the difference between the local mean fluid temperature and borehole wall temperature. If the borehole length increases or the circulating flow rate decreases by a sufficient amount, the mean local
Applications to TRT data sets
Numerical techniques are used to minimize the sum of the squares of the errors (SSE) between calculated temperatures from the borehole heat transfer model (Appendix A) and the measured temperatures of the circulating fluid during a TRT. In the present study the chosen minimization algorithm is the Levenberg–Marquardt method [27]. In all previous cases of TRT analysis known to the author the temperature data curve has been used in the minimization technique, which is also called parameter
Conclusions
The proposed analysis method for TRT data sets uses the temperature derivative with respect to the natural logarithm of time to extract additional information from TRT data sets. The procedure uses parameter estimation techniques (non-linear least squares) to match the borehole heat transfer model to measured temperature and the associated derivative curves. The method evaluates the ground thermal conductivity, effective borehole resistance and ground volumetric heat capacity. Using the
Acknowledgement
The author thanks ClimateMaster, Inc. for the TRT data sets. The research study reported in this paper did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References (32)
- R.A. Beier
Vertical temperature profile in ground heat exchanger during in-situ test
Renew. Energy
(2011)
- S. Javed et al.
Calculation of borehole thermal resistance
- M. Li et al.
Parameter estimation of in-situ thermal response tests for borehole ground heat exchangers
Int. J. Heat Mass Transfer
(2012)
- D. Marcotte et al.
On the estimation of thermal resistance in borehole thermal conductivity test
Renew. Energy
(2008)
- J.D. Spitler et al.
Thermal response testing for ground source heat pump systems—an historical review
Renew. Sustain. Energy Rev.
(2015)
- H. Yang et al.
Vertical-borehole ground-coupled heat pumps: a review of models and systems
Appl. Energy
(2010)
- J. Acuña et al.
Distributed thermal response tests on pipe-in-pipe borehole heat exchangers
Appl. Energy
(2013)
- S.J. Self et al.
Geothermal heat pump systems: status review and comparison with other heating options
Appl. Energy
(2013)
- J. Acuña, P. Mogensen, B. Palm, Distributed thermal response test on a U-pipe borehole heat exchanger, in: Proceedings...
- S. Javed, Thermal modelling and evaluation of borehole heat transfer. (Thesis (PhD)), Chalmers University of...
TED – A Mobile Equipment for Thermal Response Test, (Thesis (MSc))
(1996)
Development of an in-situ system for measuring ground thermal properties
ASHRAE Trans.
(2000)
Theory of ground pipe heat sources for the heat pump
ASHVE Trans.
(1948)
Conduction of Heat in Solids
(1959)
Determining effective soil formation thermal properties from field data using a parameter estimation technique
ASHRAE Trans.
(1999)
Cited by (19)
Tikhonov regularization stabilizes multi-parameter estimation of geothermal heat exchangers
2023, Energy
Parameter estimation from thermal response tests (TRTs) becomes unreliable when testing time reduces or the number of estimated parameters increases because of low identifiability and ill-posed mathematical feature. To overcome this challenge, this paper reports an inversion algorithm integrating a short-time temperature response model and the zero-order Tikhonov regularization strategy. We applied the algorithm to a reference sandbox dataset and examined four scenarios: simultaneous estimation of four, five, six, or seven parameters of U-shaped geothermal heat exchangers. The preliminary results indicate that the Tikhonov regularization can improve the accuracy and precision of the nonlinear multi-parameter estimation of ground heat exchangers for both long (>48h) and short (<48h) tests. The improved performance is contributed to the short-time model, which enables the short-time high-sensitivity data to be useable, and the regularization, which stabilizes the iterative optimization-solving procedure.
Effect of geological stratification on estimated accuracy of ground thermal parameters in thermal response test
2022, Renewable Energy
The ground thermal properties are the basic parameters for the design of borehole heat exchanger (BHE) in ground–coupled heat pump system (GCHPs), and their accuracy directly affects the economy and reliability of the heat pump system. In traditional design, ground is usually regarded as an isotropic hom*ogeneous medium, and its thermal properties are obtained by solving the inverse heat transfer problem in BHE through in-situ thermal response test (TRT). In fact, different geological layers can be observed along the depth of BHE, and thermal physical properties of each layer are different based on the ground geological conditions. In order to investigate the effect of geological stratification on estimated accuracy of ground thermal parameters in TRT, a validated numerical layered BHE model (NLBM) is presented to simulated TRT, and the fluid temperature response is used to estimate effective ground thermal conductivity and borehole thermal resistance based on line source model (LSM). Secondly, the distributions of fluid temperature in U-pipe and heat flux of BHE along the depth are compared and analyzed based on the NLBM and the numerical hom*ogeneous BHE model (NHBM). At last, borehole thermal resistance from the NLBM and estimated borehole thermal resistance from LSM are compared. The results show the maximum heat flux in the 3rd layer of the NLBM is 21.1% higher than that of NHBM, and the minimum heat flux in the 1st layer is reduced by 46.9% in the 100h duration. The estimated using the fluid temperature responses in the NLBM is 1.3% higher than the thickness-weighted thermal conductivity in the NLBM. The minimum relative error between and is still up to 10.45% in the duration of 100h even though extending the duration is helpful to improve the estimated accuracy of .
Deconvolution of experimental thermal response test data to recover short-term g-function
2022, Geothermics
The thermal parameters of the ground and the borehole, combined with a thermal model such as the infinite line source, enable fitting a short-term g-function by using a unit heating profile. This classical method is sensitive to model errors. The approach developed in this work extracts the short-term g-function and its derivative directly by deconvolution of thermal response test data, without having to specify a thermal model. The method minimizes a least squares-based multi-objective function on a set of nodes. The approach is robust to random power fluctuations affecting the thermal response test and allows using different heating and cooling scenarios. Results show that the deconvolved short-term g-functions display expected characteristics, and the temperature reconstructions have an error of less than 0.1°C across various numerical and field test cases, an accuracy hardly attainable by classical analytic approaches.
Effect of temperature measurement error on parameters estimation accuracy for thermal response tests
2022, Renewable Energy
Obtaining soil thermophysical parameters is the premise for design ground heat exchanger in ground source heat pump system, but it may not be accurately determined due to the limitations of the analytical models. In this paper, artificial neural network (ANN) is used to directly establish the mapping relationship between temperature response and soil thermophysical parameters, and the identification accuracy of traditional method and ANN under different measurement errors is compared. In addition, Kalman filter and fitting regression are used to remove the interference noise. The results show that the identification accuracy and stability of the traditional method are relatively weak affected by temperature measurement error, but the identification accuracy is limited. The maximum deviation errors of thermal conductivity and volumetric heat capacity are 10.68% and 18.42%, respectively, and no matter which kind of noise reduction method cannot improve the identification accuracy. The identification stability of ANN is relatively greatly affected by temperature measurement error, but the identification accuracy is high. The maximum deviation errors of the two parameters are 10.05% and 5.4%, respectively. Through the logarithmic function fitting of noise date can further improve the identification accuracy and stability, the maximum deviation errors are only 2.12% and 3.65%.
Stepwise algorithm and new analytical model for estimating multi-parameter of energy piles from thermal response tests
2022, Energy and Buildings
It is critical to determine the thermal parameters of the soil and pile before designing energy piles. A new parameter estimation method is proposed to interpret the thermal response test (TRT) data of energy piles. A cylindrical heat source model that considers the difference in the thermal properties between the soil and pile is adopted. This analytical model accurately describes the heat transfer process inside and outside the pile at the initial stage. First, a three-dimensional numerical model is established and verified by a large sandbox experiment in the literature. The numerical model is used to simulate a series of virtual TRTs. Second, the sensitivities and correlations of different thermal parameters are investigated. Based on the above analyses, a stepwise parameter estimation method combined with the pattern search algorithm is proposed to estimate the thermal resistance in the pile Rb, effective thermal conductivity ks and diffusivity as of the soil successively. The results show that the proposed method is feasible and effective for using the early-time data to estimate multiple parameters accurately. The relative errors of Rb, ks, and as are equal to 5.2%, 1.2%, and 13.3%, respectively. Finally, the influences of pile diameters and configurations on the accuracy of the estimation results are discussed in detail. Compared with the traditional slope method and parameter estimation methods based on other analytical models, this proposed method can save the time cost of TRTs and obtain more accurate thermal parameters.
Thermal response tests: A biased parameter estimation procedure?
2021, Geothermics
Thermal response tests are used to estimate the thermal properties of the ground and the borehole heat exchanger being tested. They are thus important for the design of borehole thermal energy storages and ground source heat pump systems. In this study, a theoretical framework is proposed in order to investigate if noise on the heat rate leads to a bias in the parameter estimation. Under the sole assumption of a linear time-invariant system and the use of the sum of squared errors as cost function, it is shown analytically that estimates are in fact biased when the heat rate is corrupted by noise. To understand how large this bias can be, a Monte-Carlo study is performed. It includes more than 126,000 simulations with different noises, thermal parameters and heat rate profiles.
Negative biases as high as 0.44 W/(mK) (11%) and 1110 mK/W (4.1%) are observed for the thermal conductivity and borehole thermal resistance estimates, respectively. In addition, the parameter estimation is stochastic due to randomness of measurement noises. This cannot be ignored since only one thermal response test is performed, in general. Population of estimates with 95% confidence intervals as large as 1.0 W/(mK) (25%) and 2410 mK/W (9.4%) appear in this study. Although the bias and confidence intervals are not significant in all simulated cases, they cannot be generally disregarded and one should therefore be mindful of this potential issue when analyzing thermal response tests. An observed trend is that the confidence intervals and bias are higher for higher parameter values, with a particular dependency on thermal conductivity. To reduce the bias and spread of the estimates, having larger heat rate per meter appears to be a good strategy. Having a higher sampling frequency and/or longer tests might also help, but only in reducing the spread of the estimates, not the bias.
Recommended articles (6)
Research article
Potential of dual spray injectors for optimising the noise emission of gasoline partially premixed combustion in a 2-stroke HSDI CI engine
Applied Thermal Engineering, Volume 134, 2018, pp. 369-378
This paper evaluates the potential advantages of introducing a new injector concept with a dual set of orifices in a 2-stroke engine operating with gasoline partially premixed combustion (PPC gasoline), specially in terms of combustion noise reduction. The injector is based on a configuration with two concentric needle valves and dual actuators, which allows to switch among two independent crowns of holes. The first set of orifices is controlled by the primary needle valve while an additional needle valve manages the secondary holes crown. Moreover, both valves are coupled with the two actuators separately, providing enhanced selective control over injection with an extra degree of freedom. In this investigation, a combination of Computational Fluid Dynamics (CFD) simulations and the Design of Experiments (DoE) technique is proposed with the specific purpose of optimizing this new injector configuration. In addition to determining the most convenient design, this method is extremely useful to establish cause/effect relationships between the injection design parameters and their impact on the engine performance and emissions. Results show how this solution could increase the operating range of the PPC gasoline concept by improving the combustion stability while acoustic emissions are reduced.
Research article
Numerical and experimental investigation of induction heating process of heavy cylinder
Applied Thermal Engineering, Volume 134, 2018, pp. 341-352
In this work, the induction heating of heavy cylinder is numerically and experimentally studied. The electromagnetic-temperature-stress multi-field coupling simulation are performed, using the ANSYS software in order to investigated the induction heating process of heavy cylinder, simulation results show that the end of heavy cylinder would be overheating and burning because of the acute angle effect and the high magnetic flux density when the rectangular coil is used. And then, the acute angle effect problem of heavy cylinder in the induction heating process is investigated, a solution of welding heat treatment ring at the end of the heavy cylinder and adding the magnetizer at the end of coil is proposed, and the acute angle effect can be eliminated. Furthermore, the heating effects of heavy cylinder by traditional heating method and induction heating method proposed are compared, simulation results show that the heating efficiency and the energy saving of induction heating are improved greatly. Finally, the heat treatment experiment of heavy cylinder material based on induction heating is investigated, experiment results show that the mechanical properties of the heavy cylinder can meet the application requirements.
Research article
Process calculation method and optimization of the spiral-wound heat exchanger with bilateral phase change
Applied Thermal Engineering, Volume 134, 2018, pp. 360-368
Aiming at the complexity of the process calculation of the spiral-wound heat exchangers (SWHEs) with bilateral phase change, a method of segmented calculation is presented. According to the general working conditions of oil refinery, the engineering calculation model is established. The optimal design of the spiral-wound heat exchanger is carried out, based on the genetic algorithm. When the total heat transfer capacity is constant, the total heat exchange area is used as the objective function, and the winding angle (10–30°), the thickness of the spacer a (0–10 mm) and the axial tube spacing of the same layer c (0–10 mm) are taken as the design variables. When the heredity is given to a certain algebra, the heredity terminates, the a value is 0.00237 m, the c value is 0.00814 m, the value is 11.7°, and the objective function is the total heat exchange area, which is 4900.6 m2. The optimization results show that the optimized heat transfer area is reduced by about 23.4%, The tube side and shell side pressure drop are within the allowable range, although increased.
Research article
Performance of absorption chillers in field tests
Applied Thermal Engineering, Volume 134, 2018, pp. 353-359
Absorption chillers can use waste heat, solar heat, or excess heat by cogeneration facilities to supply chilled water. Therefore, absorption chillers are important components for poly-generation, which can improve plant utilization and efficiency of the overall energy supply system.
Currently, 27 new small and medium-scale (30–160 kW cooling capacity) absorption chillers with high efficiency in full- and part-load as well as low specific size and weight are the focus of two research projects sponsored by German Federal Ministries. The chillers are located in 20 different sites, four of them in Jordan and the rest in Germany. The main objective of these projects is to enhance trigeneration and solar cooling system efficiency.
The Jordan projects and more than 70% of the German installations use dry cooling towers. Combined heat and power (CHP) plants or district heating grids provide the driving heat for regeneration at the German installations. Solar thermal collectors drive one German and all Jordan installations.
Within the projects, the chillers are provided with an intelligent control algorithm that allows achieving several objectives at the same time. One of the objectives, of course, is to match the desired chilled water temperature as well as the cooling capacity. Another objective is, for instance, to fit the hot water outlet temperature. This is commonly essential in cogeneration to avoid emergency shutdowns of the CHP engine. The field operation shows that the deviation of said temperatures from the set point is usually less than ±0.5 K in operation. In those operational hours in which the temperatures of the heat source or heat sink do not allow to reach all objectives, a set of hierarchical aims can be established within the algorithm so that the chiller can be controlled to match its prioritized objectives.
Furthermore, the designed absorption chiller performs dynamically and autonomously in a very large range of temperatures and flow rates. These can vary from 20% up to 150% of nominal conditions. This flexibility can be used to achieve up to 80% savings of electrical power consumption using speed-controlled pumps in part load conditions.
In contrast to the common opinion that small and medium-scale absorption chillers are not competitive as compared to compression chillers, this paper shows applications in which absorption chillers are advantageous because of synergies with and multiple benefits within the rest of the energy supply system.
Research article
Simulation of two-phase refrigerant separation in horizontal T-junction
Applied Thermal Engineering, Volume 134, 2018, pp. 333-340
This paper describes a 3D numerical simulation to model the two-phase separation of refrigerants in a horizontal run-type T-junction. The computations are based on the Eulerian method with the k-ε turbulence model. Compared with data from existing experiments and phenomenological models, the average deviation of the established model is less than 5% when the liquid mass flow rate ratio is in the range 0.2–0.6. The effect of multiple parameters on phase separation is investigated using the validated model. The results show that the inlet qualities (0.3–0.7), saturation temperatures (279.15–284.15 K), and working fluids of R22 and R134a have little influence on the phase separation at the T-junction, whereas the gas mass flow rate ratio increases with the increase in inlet mass flux (100–500 kg m−2 s−1). Furthermore, it is found that two symmetric recirculation regions occur directly after the junction at the entrance to the branch.
Research article
Numerical investigation on flow condensation of zeotropic hydrocarbon mixtures in a helically coiled tube
Applied Thermal Engineering, Volume 134, 2018, pp. 322-332
A numerical investigation was carried out to study the condensation flow and heat transfer coefficient of methane/propane and ethane/propane mixtures in a helically coiled tube. Two-phase heat transfer simulations had been performed at saturation pressures of 0.3–3.7 MPa with mass flux of 200, 300, 400 kg/(m2 s) and heat flux of 5 kW/m2. The effects of mass flux, saturation pressure, vapor quality and component ratios on the heat transfer coefficient were examined. In addition, annular/non-annular flow patterns were obtained by numerical simulations and experiments, and a new flow pattern transitional line for annular/non-annular flows was proposed. Meanwhile, the results were compared with the existing condensation heat transfer correlations, and the improved heat transfer correlation was proposed and well coincided with the data with a mean absolute relative deviation (MARD) of 9.2%.
© 2018 Elsevier Ltd. All rights reserved.