A Novel Analytical-ANN Hybrid Model for Borehole Heat Exchanger

Round 1

Reviewer 1 Report

A Novel Analytical-ANN Hybrid Model Borehole Heat Exchanger

Abstract: Optimizing the operation of ground source heat pumps requires simulation of both short 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.

Major and Minor Points:

1. Why the geothermal gradient is not taken into account, for the depth of 200/250 m.
2. Why the uncertainties in the measuring instruments are not taken in to account, because the accuracy of the predicted test results will be based on the input.
3. The effect of reduction of heat exchanger performance over a year due to fouling will increases the heating / cooling load, But the relative RMSE % reduces with the increase in year. Why?
4. Why the relative RMSE % is increasing for validation and testing than training. Explain.(Figure 8)
5. Will this hybrid model can be used for the prediction of BHE of other zones.
6. How much operation time of ground source heat pump was optimized through hybrid model can be included.

 

Overall, the above set points need to be considered before acceptance.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

• Improve the quality of schematic figure 1.
• Figure 2 is the same as the one presented in Puttige et al.; improvement of borehole heat exchanger model performance by calibration using measured data. Journal of Building performance simulation 2020. This could lead to copyright issues.
• Section 2.2 "filtering data". The wording explains that the “outliers” are detected by the standard deviation between the data shown in figure 3 and the model presented in Puttige et al. The results are expected (figure 4), since the results of Puttige et al. they were adjusted based on a recapitulated database from the same place from which the data in figure 3 of this manuscript were extracted. In any case, the “outliers” that are described are evidence of the goodness of the fit of the analytical solution to the data, but they are not measurement errors. It is necessary to clarify this and quantify the errors by means of error propagation and design of experiments. measurement.
• Line 225, there is no Table 1 in the manuscript.
• Explain the meaning of the subscripts and nomenclature in figure 5.
• The results of Table 2 and Figure 9 are the same, it is not necessary to present them in Table and Figure, choose just one.
• Figure 10 is not described in the text of the manuscript, apparently the results of Table 4 are those shown again in Figure 10.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

The submission entitled ‘A Novel Analytical-ANN Hybrid Model Borehole Heat Exchanger’ is considered for publication in ENERGIES. The work described a hybrid model to simulate the borehole heat exchanger to be used for the analysis of ground source heat pumps.

The methodologies to develop the simulation as described in the submission are sound. The simulation can be potentially useful as a reference for the researchers to develop a similar model. It can be considered for publication in the submitted journal.

Author Response

We thank the reviewer for his/her comments.