A causality inspired acceleration method for the fast temporal superposition of the finite line source solutions

TL;DR

This paper introduces a faster computational method for thermal interactions in solids, leveraging heat wave propagation properties to significantly reduce precomputation costs in large-scale, time-dependent simulations.

Contribution

It presents a causality-inspired acceleration algorithm that improves performance over previous schemes for finite line source solutions in thermal problems.

Findings

Precomputation cost reduced by several orders of magnitude.

Method suitable for large-scale simulations with many sources and time steps.

Enhanced robustness and efficiency over existing non-history and FFT-based methods.

Abstract

We present a novel, fast method to compute thermal interactions in solids, useful for time-dependent problems involving several sources and several time and space scales such as the ones encountered in the physics of fields of closed loop borehole heat exchangers. The new method is based on the non-history temporal superposition acceleration algorithm, but presents better performance compared to the originally proposed scheme. The main idea behind it is to leverage the propagation properties of the heat wave. Despite the basic physical solutions of heat transfer being non-causal, it is possible to establish an influence region by fixing an acceptable error tolerance. This allows to reduce the necessary integration regions in such a way that numerical integration is favored. The better behaviour of the integrand arising from this approach allows us to replace the use of Bakhalov-Vasil'eva method in favor of the asymptotic method for the computation of highly oscillatory integrals that has better properties from a computational perspective in the present application. Extensive testing is presented to evaluate the robustness of the new method and to compare its performance against the originally proposed non-history method and the convolution using the FFT algorithm for a range of error tolerances. The results show that the computational cost is highly reduced for the precomputation, which includes all the computations done before starting the time-stepping scheme. The reduction is of several orders of magnitude, depending on the specific case. This cost was the bottleneck of the original non-history implementation, and reducing it in this way makes the method suitable for simulations involving hundreds of sources and hundreds of thousands of time steps that can arise in simulations of borehole fields.