| Abstract |
This work was carried out in order to apply a statistical comparison between calibrated lines corresponding to two Thermal Response Tests in situ (TRT), substantially different in design, place and time. Also, a statistical method for comparing straight slopes was used, based on the analysis of variance homogeneity. That is, for the validation of the results obtained from a TRT of a particular horizontal exchanger, with respect to a conventional vertical exchanger, a mathematical model especially designed was applied to the latter. In San Luis, Argentina, a horizontal particular exchanger was developed and tested by applying the Kelvin infinite line source model (ILS). From this model, the coefficient of effective subsurface conductivity λeff and thermal resistance of soil Rb were determined by the graphical method of the slope. This raised a discussion about the reliability of the results obtained from the test mentioned previously. Due to this, in this work a statistical comparison method of slope of calibration lines is applied to the slopes obtained in the TRT tests; this method was conceived by the author. The reference straight line corresponds to a TRT, made at the National University of the Northeast (UNNe) Argentina, which is a standard vertical exchanger, and is compared with a straight line obtained from a particular horizontal exchanger, made at the National University of San Luis (UNSL) Argentina. The latter was embedded in the middle of a cement block of 15 cm of thickness of an existing excavation. The lower part of the block is in the subsoil and the upper part contains water from a pool; these characteristics make the exchanger different. First an "F" test is carried out to determine if the variances are homogeneous or not, and then a "t" test determines the similarity of the slopes. The "F" test establishes that the variances of the slopes are not homogeneous. The "t" test for inhomogeneous variances, and with probabilities of 95%, 97.5%, and 99%, determines that the slopes of both straight lines are different. The same "t" test for 99.5% defines similar slopes. These results establish that the method used in the particular horizontal exchanger to obtain the effective coefficient of heat transfer of soil is valid. The particular horizontal exchanger was considered as a black box, where the faithful reflection of its behavior is given by the temperatures of entry and exit, that is to say that all the factors that can influence it are given by these temperatures. The method used here for the evaluation of non-conventional exchangers yields acceptable results for the design of a commercial venture. In other words, and in this sense, it is not the same to start from laboratory parameters than from parameters obtained from a TRT in situ. Key words: Statistical comparison, thermal response test. |