Time Series Forecast of Cooling Demand for Sustainable Chiller System in an Office Building in a Subtropical Climate

Round 1

Reviewer 1 Report

I found the topic interesting and the manuscript is easy to read. I would like to have some minor suggestions:

1) page 7: wind speed v is one of the variables, and a graph for it of TMY and under future climate conditions is neccessary.

2) page 3: indoor air temperature is SET to 24C; page 6: Fig. 5, in Jan. and Dec., the outdoor air is always below 20C, but in Fig. 6 and 8 the cooling demand is NOT zero. Explanations/Discussions are suggested.

Author Response

1) page 7: wind speed v is one of the variables, and a graph for it of TMY and under future climate conditions is necessary. Added.

2) page 3: indoor air temperature is SET to 24C; page 6: Fig. 5, in Jan. and Dec., the outdoor air is always below 20C, but in Fig. 6 and 8 the cooling demand is NOT zero. Explanations/Discussions are suggested. The cooling demand is non-zero in winter months with outdoor temperatures below 20 degC because there are solar heat gains from windows and internal loads from occupants and equipment.

Reviewer 2 Report

This paper develops a simplified cooling demand model based on the time series of climatic and architectural variables to analyze carbon reduction by a sustainable chiller system. There are a few things to be checked, added, and/or corrected. The details are shown as follows:

1. Line 18: The research gap addressed here should be focused. From my view, the “simplified cooling demand model based on the time series of climatic and architectural variables” is the most innovative point of this paper, while “using variable speed chiller instead of constant speed ones to achieve energy saving” is a common sense.

2. Line 41: There are numerous studies referring to data-driven models and physical models for cooling load forecasting. The author is suggested to expand on the shortcomings of existing models and how this paper addresses those limitations. However, it may be more helpful for the author to provide a brief overview of existing models and then explain how this paper builds upon them. This would help the reader understand the context and importance of the proposed method.

3. Table 1: I think that the comparison is somewhat unfair, as the full load efficiency of the constant-speed chiller is lower than that of the variable speed chiller.

4. Table 2: It would be helpful for the author to clarify whether the coefficients for the chiller models were fitted by the author or provided by EnergyPlus. Additionally, the author could provide more information on how the models were validated, such as by comparing predicted and actual chiller performance.

5. Line 291: The outdoor air enthalpy is a function of temperature and humidity, which is strongly correlated with temperature, so I think the input parameters here should be temperature and humidity.

6. Section 4.3: It would be helpful for the author to clarify the purpose of the “Time Series Cooling Demand Model” developed in Section 4.2, while the results in Section 4.3 can be calculated based on the cooling loads from EnergyPlus as well. The author could explain how the two methods compare and why the “Time Series Cooling Demand Model” is valuable.

Author Response

1. Line 18: The research gap addressed here should be focused. From my view, the “simplified cooling demand model based on the time series of climatic and architectural variables” is the most innovative point of this paper, while “using variable speed chiller instead of constant speed ones to achieve energy saving” is a common sense. The suggested phrase is used to address the innovative point in abstract.
2. Line 41: There arenumerous studies referring to data-driven models and physical models for cooling load forecasting. The author is suggested to expand on the shortcomings of existing models and how this paper addresses those limitations. However, it may be more helpful for the author to provide a brief overview of existing models and then explain how this paper builds upon them. This would help the reader understand the context and importance of the proposed method. Table 1 is added to summarize existing models with their optimization algorithms in citations.
3. Table 1:I think that the comparison is somewhat unfair, as the full load efficiency of the constant-speed chiller is lower than that of the variable speed chiller. Table 1 (Table 2 in the revised version) shows two systems to be investigated based on typical performance specifications in reference [21]. Based on the COP and IPLV difference, the total energy uses of the two system types are assessed and compared.
4. Table 2: It would be helpful for the author to clarify whether the coefficients for the chiller models were fitted by the author or provided by EnergyPlus. Additionally, the author could provide more information on how the models were validated, such as by comparing predicted and actual chiller performance. The coefficients are not from default models in EnergyPlus. The performance characteristics of the constant speed chiller belong to a commercial chiller fitted and validated by a study [25]. The coefficients in EIRPLR of the variable speed chiller were fitted based on performance data reported in a study [26] (newly added).
5. Line 291: The outdoor air enthalpy is a function of temperature and humidity, which is strongly correlated with temperature, so I think the input parameters here should be temperature and humidity.

The authors examined the correlation between cooling demand and temperature, enthalpy and humidity. Among the three, humidity has very low correlation. Using enthalpy is more preferable than temperature only to enhance model accuracy.

6. Section 4.3: It would be helpful for the author to clarify the purpose of the “Time Series Cooling Demand Model” developed in Section 4.2, while the results in Section 4.3 can be calculated based on the cooling loads from EnergyPlus as well. The author could explain how the two methods compare and why the “Time Series Cooling Demand Model” is valuable.

This simplified model can be used when a complete set of weather data in the epw format is not available in the EnergyPlus simulation. The sensitivity of the cooling demand by different WWRs and climatic variables can be assessed directly without changing the building modules in EnergyPlus.

Reviewer 3 Report

The study develops a simplified cooling demand model based on the time series of climatic and architectural variables to analyze carbon reduction by a sustainable chiller system.

The weather data in 2020, 2050 and 2080 was produced by the weather generator program CCWorldWeatherGen, and an hourly cooling demands of a reference high-rise office building in Hong Kong under a change of window-to-wall ratios from 0.4 to 0.8 and climate conditions in the TMY, 2020, 2050 and 2080 was similated byEnergyPlus.

The input variables are the window-to-wall ratio, outdoor air enthalpy, global solar radiation, wind speed and their two steps ahead.

 

However, the envelope structure of the reference high-rise office building will also have great changes from 2020 to 2080, including the thermal resistance of the envelope material, water vapor permeability coefficient, sealing property and so on. Furthermore, the carbon emissions results also changed. This study should fully consider these influencing factors in the simulation model.

Author Response

However, the envelope structure of the reference high-rise office building will also have great changes from 2020 to 2080, including the thermal resistance of the envelope material, water vapor permeability coefficient, sealing property and so on. Furthermore, the carbon emissions results also changed. This study should fully consider these influencing factors in the simulation model.

The authors admitted these changes are important but the present study cannot cover these aspects and they will be addressed as limitations in this study. Recommendations on the future work on this are stated.

Reviewer 4 Report

In this manuscript, the authors develop a simplified cooling demand model based on the time series of climatic and architectural variables to analyze carbon reduction by a sustainable chiller system. EnergyPlus is used to simulate hourly cooling demands of a reference high-rise office building in Hong Kong under a change of architectural parameters and future climate conditions. An hourly cooling demand model with R2 above 0.9 is developed with inputs of the window-to-wall ratio, outdoor air enthalpy, global solar radiation, wind speed and their two steps ahead.  The validated model is then used to analyze carbon reduction potentials by free cooling and a full variable speed chiller system. Thus, this study is conducive to the development of carbon neutrality.

The topic falls within the scope of the journal and the reported data are detailed. However, there are some concerns that should be taken into consideration.

 1.     The authors trained the model using the data of TMY and tested the model on the data generated by a program. Since the generation process is "shifting and stretching the TMY data", there seems to be an information leakage problem. In other words, the test set is not "unknown" to the model. Such problem may result in over-optimistic results and potentially threaten the conclusion drawn by the authors. Please provide more details about the data generation stage and discuss the risk of information leakage. If it is indeed a problem, the authors may need to name it as a limitation in the Discussion section.

2.     Why did the author choose the office building with 40 floors? What are the reasons and criteria for choosing?

Author Response

1. The authors trained the model using the data of TMY and tested the model on the data generated by a program. Since the generation process is "shifting and stretching the TMY data", there seems to be an information leakage problem. In other words, the test set is not "unknown" to the model. Such problem may result in over-optimistic results and potentially threaten the conclusion drawn by the authors. Please provide more details about the data generation stage and discuss the risk of information leakage. If it is indeed a problem, the authors may need to name it as a limitation in the Discussion section.

       The authors admitted the information leakage problem as the three future climate scenarios cannot be validated by data in the present decade. Assumptions about the future weather data are further elaborated with addition of ref. [30]. The authors address this problem as one of the limitations in discussion in the revised manuscript.

2. Why did the author choose the office building with 40 floors? What are the reasons and criteria for choosing?

       The building configuration is a hypothesized case used to evaluate the energy performance of typical office buildings in Hong Kong. It is applied in two studies added in references [20], [21] and [22].

Reviewer 5 Report

The topic is interesting and the manuscript is well prepared and is well written. However, the manuscript still have some weaknesses and requires more clarification. The comments are as follows.

- In introduction, the authors must add the details why this proposed method give advantage over the previous work. This is to provide more specific research gap and shows more detail of the research originality.

- from literature review, the authors should summarize the weaknesses of the previous works and shows the solutions to fix it based on the proposed method.

- from the discussion, the reviewer did not see how the simulated cooling load can provide more sustainable in air-conditioning process.

- the authors must consider to add more details discussion about the refrigeration machine which can reduce the electricity consumption while the cooling demand is still satisfied. This will really give more sustainable in cooling or air conditioning process.

- the authors must explain more clear on how to calculate the cooling load from meteorological data (the solar radiation, ambient temperature, etc.). This will provide deep insight into the impact of each parameters on the predicted cooling load. 

- the authors should consider to add a specific section to discuss the topic how the simulated results can provide more sustainable in air-conditioning process with the help of the proposed results. This will give more interesting.

- more specific conclusions is still needed.

Author Response

- In introduction, the authors must add the details why this proposed method give advantage over the previous work. This is to provide more specific research gap and shows more detail of the research originality.

The red paragraph in lines 70-75 is added in introduction to address originality.

- from literature review, the authors should summarize the weaknesses of the previous works and shows the solutions to fix it based on the proposed method.

Table 1 is added to summarize previous works and more comments on them are provided. Please see red graphs from lines 42 – 75.

- from the discussion, the reviewer did not see how the simulated cooling load can provide more sustainable in air-conditioning process.

The sustainable system refers to using full variable speed control to the chiller system and free cooling. The corresponding energy savings and carbon reduction are reported and discussed in lines 385 – 473.

• the authors must consider to add more details discussion about the refrigeration machine which can reduce the electricity consumption while the cooling demand is still satisfied. This will really give more sustainable in cooling or air conditioning process.

Section 4.3 explains electricity savings with the full variable speed control and potential reduction of cooling demand due to free cooling.

- the authors must explain more clear on how to calculate the cooling load from meteorological data (the solar radiation, ambient temperature, etc.). This will provide deep insight into the impact of each parameters on the predicted cooling load.

More elaborations on the generation of future weather data are provided in lines 193-213. 

- the authors should consider to add a specific section to discuss the topic how the simulated results can provide more sustainable in air-conditioning process with the help of the proposed results. This will give more interesting.

More discussion is provided on how to achieve the energy and carbon reduction targets. Please refer to red paragraphs in lines 448-472.

- more specific conclusions is still needed.

Conclusions include more specific data, limitations and recommendations for future work. Please refer to red paragraphs in lines 475-532.

Round 2

Reviewer 2 Report

The authors have addressed all the issues.

Reviewer 3 Report

The author has fully limited and responded to the questioning in “Conclusions and recommendations”