A New Explication of Minimum Variable Sets (MVS) for Building Energy Prediction Based on Building Performance Database
Round 1
Reviewer 1 Report
The work is interesting and well structured, I recommend it for publication.
Author Response
～Thank you for the accurate extraction and conclusion of our research perspective in this study.
～We’ve reviewed the WHOLE manuscript carefully to avoid formatting, spelling, or grammar errors, and checked the citation format issues. In addition, we asked several colleagues including native speakers and skilled authors to check it. We believe that the language is now acceptable.
Reviewer 2 Report
The manuscript is relevant with the Journal scope and is generally well-written.
The main research questions are important and relevant in design practice, regarding which building variables are the most indispensable for energy prediction, and how to achieve accurate building energy prediction with limited variables. To address the questions, the concept 124 of minimum variable sets (MVS) is proposed and has been successfully tested.
Few minor comments:
In Eq. (4), please replace the asterisk sign (*) with multiplication sign (×).
The term cov and var are usually written in non-italic.
Line 326: 'meduim' should be 'medium'.
The title of Section 3.2: 'Applicant discussion of the BPD' perhaps should be 'Application of BPD'.
In Fig. 3, is it possible to put round numbers for the errors shown on the x-axis?
Line 59, 195, 576: 'there have ...' should be 'there are ...'
Author Response
～Thank you for the accurate extraction and conclusion of our research perspective in this study. We will improve the manuscript referring to your useful comments and recommendations.
～We’ve reviewed the WHOLE manuscript carefully to avoid formatting, spelling, or grammar errors, and checked the citation format issues. In addition, we asked several colleagues including native speakers and skilled authors to check it. We believe that the language is now acceptable.
～In the revised manuscript, revisions in the text are shown using red highlight.
Few minor comments:
Point 1: In Eq. (4), please replace the asterisk sign (*) with multiplication sign (×).
Response 1: Thanks for your careful check on our paper. The * sign has been revised to the formal multiplication sign .
Point 2: The term cov and var are usually written in non-italic.
Response 2: Thanks for your patience. The term cov and var in Eq. (4) has been revised to be in non-italic, as well as the same content in the descriptive paragraph.
Point 3: Line 326: 'meduim' should be 'medium'.
Response 3: Sorry for the typo. It has been revised in the updated version.
Point 4: The title of Section 3.2: 'Applicant discussion of the BPD' perhaps should be 'Application of BPD'.
Response 4: We agree with the reviewer about this. It has been revised by 'Application of BPD' in the updated version.
Point 5: In Fig. 3, is it possible to put round numbers for the errors shown on the x-axis?
Response 5: Thanks for your careful check on our paper. The numbers on the x-axis have been revised to round numbers in Fig. 3.
Point 6: Line 59, 195, 576: 'there have ...' should be 'there are ...'
Response 6: Thanks for your patience of reading and feedback. The above lines have been revised by 'there are ...'. Also, we’ve checked the whole manuscript for the similar grammatical mistakes.
Reviewer 3 Report
Thank you for your paper. Please find a few comments below:
1. Please explicitly mention your contributions, point by point, in your introduction. They are there but not explicit.
2. Provide an overview of the paper structure towards the end of the introduction.
3. Lines 365 to 387 do not make sense to me. Before line 365 you need to provide a line or two to explain what they are.
Author Response
Point 1:. Please explicitly mention your contributions, point by point, in your introduction. They are there but not explicit.
Response 1: We agree with the reviewer that the main contribution should be explicitly mentioned in the introduction part.
As for this comment, we’ve added the related description of the main contribution and the research objectives in the updated version. Please check the Lines 125 to 137 with red highlight. We’ve emphasized the main contribution of this study, including the designed BPD and the hierarchical MVSs, as well as the relationship between the MVS with the model predictive accuracy if using the MVS as model inputs, in the application of the resulted MVSs to forward and data-driven models. Through the above research work, this paper attempts to answer the mentioned two questions Q1 and Q2, and also provide a more universal and valuable reference for relevant researchers on inputs/variables selection of building energy prediction models.
Point 2: Provide an overview of the paper structure towards the end of the introduction.
Response 2: Thanks for your reminding of this key point.
As for this comment, we’ve added the overview of the paper structure at the end of the introduction part. Please check the Lines 145 to 148 in the updated version.
Point 3: Lines 365 to 387 do not make sense to me. Before line 365 you need to provide a line or two to explain what they are.
Response 3: Thanks for your patience and feedback. The original Lines 365 to 387 have been changed to Lines 383 to 405 in the updated version.
As for this comment, we’ve revised the contents before this part to provide a more clear explanation of the M0, M1, M2, and M3. Please check the Lines 374 to 381 in the updated version. This study compares the predictive accuracy of different models (M1, M2, and M3) through the application of varying MVSs in the forward simulation models. With a well calibrated model for an actual office building, M0, as the baseline of comparative analysis, M1, M2, and M3 refer to the adjusted simulation models of the office building that separately apply the hierarchical MVSs (VS_EUI_Max, VS_EUI_Med, and VS_EUI_Min) as their model inputs. The comparative models are then detailed by the Lines 381 to 405 in the updated version.
