Efficient g-function approximation with artificial neural networks for a varying number of boreholes on a regular or irregular layout

Dusseault, Bernard; Pasquier, Philippe

doi:10.1080/23744731.2019.1634932

Cited by 9 publications

(3 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Others have directly computed g-functions for the purpose of directly predicting the GHE exiting fluid temperature. Dusseault & Pasquier [31] briefly mention the method, but a full derivation is given by Pasquier et al [32]. The approach provides a way to utilize the Eskilson-type g-functions and combine them with the short-timestep g-functions for directly computing the fluid temperature.…”

Section: Response Factor Modelsmentioning

confidence: 99%

An Enhanced Vertical Ground Heat Exchanger Model for Whole-Building Energy Simulation

Mitchell

Spitler

2020

Energies

View full text Add to dashboard Cite

This paper presents an enhanced vertical ground heat exchanger (GHE) model for whole-building energy simulation (WBES). WBES programs generally have computational constraints that affect the development and implementation of component simulation sub-models. WBES programs require models that execute quickly and efficiently due to how the programs are utilized by design engineers. WBES programs also require models to be formulated so their performance can be determined from boundary conditions set by upstream components and environmental conditions. The GHE model developed during this work utilizes an existing response factor model and extends its capabilities to accurately and robustly simulate at timesteps that are shorter than the GHE transit time. This was accomplished by developing a simplified dynamic borehole model and then exercising that model to generate exiting fluid temperature response factors. This approach blends numerical and analytical modeling methods. The existing response factor models are then extended to incorporate the exiting fluid temperature response factor to provide a better estimate of the GHE exiting fluid temperature at short simulation timesteps.

show abstract

Section: Response Factor Modelsmentioning

confidence: 99%

An Enhanced Vertical Ground Heat Exchanger Model for Whole-Building Energy Simulation

Mitchell

Spitler

2020

Energies

View full text Add to dashboard Cite

show abstract

“…Due to available computing resources, the work in this paper is focused on borehole configurations with a maximum number of boreholes around 1000. Dusseault and Pasquier (2019) demonstrated use of artificial neural networks to compute g-functions. 500,000 g-functions computed for fields of up to 10 boreholes were used to train the neural network.…”

Section: Introductionmentioning

confidence: 99%

Faster computation of g-functions used for modeling of ground heat exchangers with reduced memory consumption

Spitler¹,

Cook²

2021

Building Simulation Conference Proceedings

View full text Add to dashboard Cite

show abstract

“…Esen and Inalli [45], Sun et al [46], and Fannou et al [47] used ANN to predict the overall performance of the GSHP. ANN has also been used for design [48,49], determination of ground properties [50], and g-function generation [23,51].…”

Section: Introductionmentioning

confidence: 99%

A Novel Analytical-ANN Hybrid Model for Borehole Heat Exchanger

et al. 2020

View full text Add to dashboard Cite

Optimizing the operation of ground source heat pumps requires simulation of both short-term and long-term response of the borehole heat exchanger. However, the current physical and neural network based models are not suited to handle the large range of time scales, especially for large borehole fields. In this study, we present a hybrid model for long-term simulation of BHE with high resolution in time. The model uses an analytical model with low time resolution to guide an artificial neural network model with high time resolution. We trained, tuned, and tested the hybrid model using measured data from a ground source heat pump in real operation. The performance of the hybrid model is compared with an analytical model, a calibrated analytical model, and three different types of neural network models. The hybrid model has a relative RMSE of 6% for the testing period compared to 22%, 14%, and 12% respectively for the analytical model, the calibrated analytical model, and the best of the three investigated neural network models. The hybrid model also has a reasonable computational time and was also found to be robust with regard to the model parameters used by the analytical model.

show abstract

Efficient g-function approximation with artificial neural networks for a varying number of boreholes on a regular or irregular layout

Cited by 9 publications

References 39 publications

An Enhanced Vertical Ground Heat Exchanger Model for Whole-Building Energy Simulation

An Enhanced Vertical Ground Heat Exchanger Model for Whole-Building Energy Simulation

Faster computation of g-functions used for modeling of ground heat exchangers with reduced memory consumption

A Novel Analytical-ANN Hybrid Model for Borehole Heat Exchanger

Contact Info

Product

Resources

About