Application of deconvolution to interpretation of distributed thermal response test (DTRT) and to determination of thermal conductivity profiles

Abstract

Distributed thermal response tests offer the possibility to retrieve both the effective and depth-dependent thermal conductivity profile of a ground heat exchanger. The purposes of this paper are to apply a novel deconvolution method on distributed thermal response test data, which has not been performed before, and to showcase the advantages of using short-term transfer function to calculate thermal conductivity values at each geological layer from a conventional interpretation method, which is also a novelty. To achieve that, the ground is separated in layers from which a set of transfer functions and a thermal conductivity profiles are estimated. Results show that at the ground heat exchanger outlet, the effective thermal conductivity is estimated with comparable accuracy from temperature signals and transfer functions, but for the depth-dependent thermal conductivity profile, the maximum error is reduced from 51.45% to 3.83% for a low thermal conductivity layer using transfer functions. A comparison between the analysis with temperature signals or transfer functions also shows the robustness of the latter to interpret tests with low flow rates. This methodology has the potential to aid decision-making when designing a ground heat exchanger field by providing a clearer representation of the heat exchange capacity of the ground at a site under study.