Markov Chain Monte Carlo Based Energy Use Behaviors Prediction of Office Occupants
Round 1
Reviewer 1 Report
Line 260-261: "Air-conditioner can effectively improve the indoor thermal environment, but also affect the thermal feeling and thermal comfort of office occupants".
Expressed like this it seems that it's a bad consequence of the use of the air conditioner to affect the thermal feeling and thermal comfort. The purpose of the air conditioner is to do it. In fact, it is to reach the desired thermal comfort
Lines 262-263: I do not understand the following sentence "According to the research of indoor environmental parameters (temperature and humidity), the rules of air-conditioner action can be reflected."
Please rewrite it
Lines 264-265: "MCMC algorithm is used to build the energy use behavior model of manual on and off air-conditioner, and predict the user’s air-conditioner use behavior based on temperature data."
Please clarify how MCMC is able to do so.
Lines 265-266: I guess the following sentence is out of context "Jinan is
located in a cold area."
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
It is an interesting paper regarding energy use behavior in office buildings. Since I am not an expert on MC analysis my only comment relates to the application of the method described, in other case studies (office or residential buildings).
Is something like that possible and under what circumstances (e.g. in a residential building) given that performance data like the ones generated in the context of this paper may not be available. I would expect a discussion on this issue in the final section of the paper to guide interested readers (especially those not familiar with MC analysis) accordingly.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
No new comments
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The paper presents an MCMC method for modeling energy-use patterns of electric devices in an office building. The topic is interesting and relevant to the community and the testing set-up includes novel sensing techniques. However, moderate improvement of content and language is required before the paper can be published:
In the introduction, the authors do not explicitly explain the necessity of creating probability distributions for power-on/off actions. Why are these models useful or advantageous to other approaches? Can they be used for better scheduling or operation of building services? Does it help in saving energy or increasing occupants comfort? Also, why do we need statistical models? How often do we need to re-calibrate the models? Can they be applied to other buildings under certain assumptions? The values in Table 1 are confusing, because the units for RMSE and MAE are not mentioned. It helps if authors explain one row of the table in details, so the reader can determine, for example, if RMSE of 3.87 is high or low for a parameter value estimated at 0.03. The same critic goes for Table 2. The authors must show the units and explain an example. Also, variable "m" was already used in equations (9) and (10), so perhaps you can use a different symbol. There numerous spelling and grammatical errors in the text that make the reading very difficult. I strongly recommend that the paper goes through a round of proof reading by a native English speaker. Some examples: Line 28: "bdayes" --> "Bayes" and "macheine" --> "Machine" Line 136: "including" --> "include" Line 149: "It need to" --> "It needs to" Line 164: "need to be" --> "needed to be" Line 183: "the user usually power-on" --> "users" or "powers on" Line 264: "the user is trend to turn on" --> "the user tends to turn on" In various locations: "can not" --> "cannot" etc etc Please kindly revise the paper and resubmit an improved version.Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
In my opinion, the article lacks a comparative study with methods based on the sensation of thermal comfort, as well as the estimation of energy consumption for the climatic conditioning of buildings. Both types of study are necessary to correlate with user behavior.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Thanks for improving the paper!
Reviewer 2 Report
According to authors reply:
1) Authors reply: "we mainly focused on the prediction of energy use
behaviors from the operation and state data of desktop computer, water dispenser and light in the present study."
In my opinion this is far from the total energy consumption in office buildings, and therefore, of very low interest.
2)Authors reply: "We will expand the types of data collected to non-electrical
parameters such as light intensity, temperature, humidity and the user's sensation of specific environment. And some comparative studies will be carried out when the more complete dataset is ready."
My suggestion is to wait for that study
