A Computer Vision-Based Occupancy and Equipment Usage Detection Approach for Reducing Building Energy Demand
Round 1
Reviewer 1 Report
This is a very good paper. The author tried to identify the real internal loads and used it calculate the heating and cooling load.
Here are some minor comments for imporvement:
- Figures in Table 5 should be thicker to be easy to see.
- The paper contains many figures already. However, if possible, the authors should also present hourly values for the heating and cooling loads. In general, heating and cooling loads in kW when knowing their distribution and maximum values are chritical for sizing the equipment. Thereby, the results from this paper may be even more valuable.
Author Response
Please find enclosed the response.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Thank you for submitting your paper “A computer vision-based occupancy and equipment usage detection approach for reducing building energy demand” to the Journal of Energies.
The study is very well written and organized. It reports clearly all the information.
In my opinion, it is necessary to focus more on explaining the case study, adding more technical information to determine the thermal profile of the office analyzed.
In literature, there are other studies on the offices, such as:
https://doi.org/10.1016/j.enbuild.2019.109707.
https://doi.org/10.1016/j.buildenv.2020.107364.
https://doi.org/10.1016/j.energy.2015.08.078.
All the figures are well made, even if figure 3 is not very readable, it should be improved.
In figure 4 "test room" is cut in half.
The orientation of the room is not clear, it could be inserted North indication.
Please add the table of nomenclature.
Author Response
Please find enclosed the response.
Author Response File: Author Response.docx
Reviewer 3 Report
Review
The paper entitled “A computer vision-based occupancy and equipment usage detection approach for reducing building energy demand” proposes a vision-based occupancy and equipment usage detection approach for demand driven control systems to minimise the unnecessary energy usage and enhance thermal comfort. In order to do this, a model has been proposed and validated and presents the capabilities of the model to detect the equipment usage and recognise the differences between the corresponding human poses for each specific activity and provides an efficient alternative to accurately detect and estimate the internal heat gains.
In my opinion, the article is interesting and provides new knowledge applicable to optimizing the energy use of buildings, however, there are some aspects that could be improved for a correct publication of the article:
Minor flows:
Introduction
In my opinion, the bibliographic review should be somewhat more extensive, especially in relation to deep learning, citing more research that has addressed similar topics and mentioning the results that the researchers reached. However, I very much appreciate the inclusion of the section “Literature gap and novelty” as it clearly shows the shortcomings of existing research on this specific field.
Methods
The methodology proposed in this research seems correct and well described, however, section 2.1 “Deep learning Method” I think should be described more extensively in the introduction, while in the method section, only the what and how it will be applied in the investigation of this article.
The same situation happens with section 2.1.2. Detection Model: CNN-Based Model Selection and Configuration. I think this section makes more sense in the introduction, especially the paragraph from line 172 to line 183, “According to previous works, many choose TensorFlow as the desired platform for the development of solutions for building-related applications. This includes [35] where TensorFlow has been used as a platform to train the desired deep learning model. Vázquez-Canteli et al. [36] fused TensorFlow technique with building energy simulation (BES) to develop an intelligent energy management system for smart cities and Jo and Yoon [37] indicated that TensorFlow was used to establish a smart home energy efficiency model. Additionally, the provision of pre-existing open source deep learning-based models by TensorFlow, such as the CNN TensorFlow object detection API [38] enabled researchers to use this framework as the base configuration for detection-based applications. This includes the applications by [29, 39, 40], which effectively fine-tuned the model to improve accuracy and to adapt for the research desired detection purposes. Therefore, the TensorFlow platform with the CNN object detection API was employed for the development of a suitable model for this study”. All this explanation should be described in the introduction, while the methods section should only indicate how we are going to apply it.
Results
The results are interesting and allow the optimization of energy use in buildings, but I think it would be interesting to add a “discussion” section in which it is specified why the differences found between scenario 1 and scenario 4 occur, and whether or not there are coincidences or disagreements of these results with previous investigations carried out and that have been cited in the introduction.
Other flows:
Line 19. R-CNN. Acronyms must be described the first time
Line 227. Error in the numbering of the section, instead of 2.3.1. it should be 2.2.1.
Line 279. Error in the numbering of the section, instead of 2.4.1. It should be 2.3.1.
Line 295. Error in the numbering of the section, instead of 2.4.2. It should be 2.3.2.
Line 315. Check the numbering of the section
Line 363. Table 4 is unconfigured. You have to give it a better format.
Comments for author File: 
Comments.pdf
Author Response
Please find enclosed the response.
Author Response File: Author Response.docx
