Parameter identification algorithm for ground source heat pump systems

Abstract

This paper presents a new parameter identification (PI) algorithm for estimating effective and detailed thermal parameters of ground source heat pump systems using data obtained from the well-known thermal response test. The PI comprises an iterative scheme coupling a semi-analytical forward model to an inverse model. The forward model is formulated based on the spectral element method to simulate transient 3D heat flow in ground source heat pump (GSHP) systems, and the inverse model is formulated based on the interior-point optimization method to minimize the system objective function. Compared to existing interpretation tools for the thermal response test, the proposed PI algorithm has several advanced features, including: it can handle fluctuating heat pump power and inlet temperatures; interpret data obtained from multiple heat injection or extraction signals; produce accurate backcalculation for short and long duration experiments; and handle multilayer systems. The PI algorithm is tested against synthesized data, using a wide range of random noise, and versus an available laboratory experiment. The computational results show that the PI algorithm is accurate, stable and exhibiting relatively high convergence rate.