Performance Analysis of Lake Water Cooling Coupled with a Waste Heat Recovery System in the Data Center

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

After carefully reviewing the manuscript, I am pleased to note that the authors have diligently addressed the work on a system integrating lake water cooling with waste heat recovery for DCs. To evaluate the energy-saving performance of the suggested system, the influence of waste heat recovery locations and volumes has been investigated. The authors cover various aspects, including synthesis methods, modification techniques, characterization, and mechanisms of free cooling and waste heat recovery techniques are promising approaches to reduce DC energy consumption. The review is well-structured, thoroughly referenced, and presents a critical analysis of recent advances in the field.

The following comments have been made to improve the manuscript:

1.      Can the impact of scenarios a2 and a3 on overall system efficiency be quantified more precisely? In section 3.1.

2.      How does the cooling energy consumption compare between scenarios a3 and a4 specifically?

3.      What specific criteria were used to determine the heat recovery levels (b1 to b5) in relation to the minimum cooling load of the DC? In section 3.3.

4.      How do scenarios b1 to b5 impact the reliability and stability of the heating and cooling systems in practical terms?

5.      What are the operational challenges associated with implementing localized (c1) versus uniform (c2) heat extraction methods? In section 3.4.

6.      Can the impact of heat extraction methods on system maintenance and lifecycle costs be addressed?

7.      How do variations in chilled water supply temperature and air supply temperature affect the system’s ability to meet cooling demands?

 

8.        What are the practical implications of different water tank volumes on system scalability and long-term efficiency?

Suggestions:

1.      Clarify the methodology used to ensure independence of heating and cooling systems in scenarios a1 and a2.

2.      Provide more detailed insights into how waste heat recovery affects operational adjustments in the cooling system in scenarios a3 and a4.

3.      Include a discussion on the economic feasibility alongside energy savings for each heat recovery volume scenario.

4.      Explain the rationale behind choosing scenarios b1 to b5 and how these scenarios contribute to the overall system efficiency.

5.      Provide a comparative analysis of the efficiency and scalability of localized versus uniform heat extraction methods.

6.      Discuss potential trade-offs between energy savings and system complexity introduced by each method.

7.      Expand on the economic implications of optimizing chilled water supply temperatures and air supply temperatures.

8.      Include a discussion on the environmental benefits and potential trade-offs associated with adjusting system parameters.

9.      Ensure clarity in defining terms such as PUE (Power Usage Effectiveness) and explain their significance in the context of your study.

10.  Provide more context on the limitations and assumptions of your simulations, particularly regarding their applicability to real-world scenarios.

11.  Consider integrating graphical data (figures) more seamlessly with the text to enhance readability and understanding.

Author Response

Reviewer 1

After carefully reviewing the manuscript, I am pleased to note that the authors have diligently addressed the work on a system integrating lake water cooling with waste heat recovery for DCs. To evaluate the energy-saving performance of the suggested system, the influence of waste heat recovery locations and volumes has been investigated. The authors cover various aspects, including synthesis methods, modification techniques, characterization, and mechanisms of free cooling and waste heat recovery techniques are promising approaches to reduce DC energy consumption. The review is well-structured, thoroughly referenced, and presents a critical analysis of recent advances in the field.

The following comments have been made to improve the manuscript:

1. Can the impact of scenarios a2 and a3 on overall system efficiency be quantified more precisely? In section 3.1.

Response: Thank you for your professional question.

We have added the description of the energy-saving performance of Scenarios a2 and a3 in Ssection 3.1.

2. How does the cooling energy consumption compare between scenarios a3 and a4 specifically?

Response: Thank you for your good question.

Scenario a3 involves recovering waste heat from the chilled return water, while Scenario a4 involves recovering waste heat from the return air of IT equipment. The energy consumption of mechanical refrigeration includes the energy consumption of the chiller, cooling towers, and the cooling water pump. Due to both Scenario a3 and Scenario a4 recovering waste heat on the load side of the chiller, there is only a slight difference in their impact on reducing mechanical refrigeration energy consumption. Similarly, the energy consumption difference between the lake water pump and fans is also paltry. Compared with Scenario a3, Scenario a4 reduces the return air temperature of IT equipment, lowers the flow rate of chilled water, and thus reduces the energy consumption of the chilled pump. Therefore, the cooling system's energy consumption in Scenario a4 is lower than that in Scenario a3. We have added the comparative analysis of these two scenarios as follows:

“3.1. Comparative analysis of energy consumption of waste heat recovery locations

…

…The main reason for energy conservation in these two scenarios is to reduce the energy consumption of mechanical refrigeration. The energy consumption of mechanical refrigeration is defined as the sum of the energy consumption of the cooling tower, the cooling water pump, and the chiller…., while the operational time of the free cooling mode in Scenarios a3 and a4 is 5144 and 5267 hours, reducing by 58.72% and 60.13%, respectively. The mechanical cooling energy consumption of Scenarios a1 and a2 reaches 654.78 MWh, while that of Scenarios a3 and a4 is 479.97 MWh and 480.47 MWh, respectively. Furthermore, compared to Scenario a3, the heating system in Scenario a4 recovers the waste heat from the return air of IT equipment,…Furthermore, the COP of the cooling system in Scenarios a1 and a2 is 9.14, while Scenarios a3 and a4 achieve COP of 9.85 and 10.17 respectively due to the recovery of waste heat from the cooling system.”

3. What specific criteria were used to determine the heat recovery levels (b1 to b5) in relation to the minimum cooling load of the DC? In section 3.3.

Response: Thank you for your professional question.

To ensure that the waste heat of the data center can fully cover the heating demands of the hotel when designing the 100% waste heat recovery scenario (b5), we separately found the peak cooling load of the data center and the valley value of the heating load of one hotel and then expanded the heating load to make them equal. '100%' means that the valley value of residual heat is equal to the peak value of the heating load. Subsequently, 10%, 30%, 50%, and 80% heat supply scenarios are derived based on 100% of the hotel heat supply, named Scenario b1 to b4. We have clarified the design principles for Scenarios b1 to b5, as follows:

“3.2. Comparative analysis of waste heat recovery volumes

…To maximize the utilization of waste heat and ensure that it fully covers the heat demands, the expansion is carried out based on the heating load of one hotel until the peak heat load equals the valley value of the DC cooling load. It is named 100% heat supply (Scenario b5)…. the Scenario b5…b4.”

4. How do scenarios b1 to b5 impact the reliability and stability of the heating and cooling systems in practical terms?

Response: Thank you for your good question.

For the cooling system, the greater the utilization of waste heat in the data center by the heating system, the lower the temperature of the chilled water return, especially in Scenario a3. This may increase the time for the return temperature of the chilled water to be lower than the temperature of the lake water. In other words, waste heat recovery will enhance the utilization of free cooling sources in the cooling system and reduce reliance on mechanical refrigeration. For the heating system, the more waste heat is recovered while ensuring that the data center's waste heat covers the heating load, the higher the energy efficiency of the heating system. The basic heating system used for comparison is an air-source heat pump system that utilizes outdoor air for heating. Therefore, an increase in the level of waste heat recovery can bring dual benefits of reducing the energy consumption of the data center cooling system and the hotel heating system. At the same time, improves the reliability and stability of the system. We have added improvements to the system in Scenarios b1 to b5, as follows:

“3.2. Comparative analysis of waste heat recovery volumes

…

…Specifically, compared to Scenario a1, when the heat supply escalates from 10% to 100%, the cooling energy consumption in Scenarios a3 and a4 is 2496 MWh to 2217 MWh and 2400 MWh to 1925 MWh, diminishing by 11.16% and 19.80%, respectively…. It means that the more heat is recovered from the return air of IT equipment and the chilled return water, the lower the energy consumption of the cooling system…. even though the temperature of the return air of IT equipment is higher than that of chilled return water.

Fig. 7 indicates the trends of PUE and COP in Scenarios a3 and a4 at different levels of waste heat recovery. Whether it is recovering waste heat from chilled water return or return air of IT equipment, with the enhancement of the degree of waste heat utilization, the implementer of the PUE and COP of the DC. Specifically, recovering waste heat from chilled water return can achieve an increase in PUE from 1.199 to 1.186 and an increase in COP from 9.54 to 10.73. The effect of recovering waste heat from IT equipment return air is more significant. It can increase PUE from 1.194 to 1.173 and COP from 9.92 to 12.36.”

5. What are the operational challenges associated with implementing localized (c1) versus uniform (c2) heat extraction methods? In section 3.4.

Response: Thank you for your good question.

For the uniform heat extraction method, all air source heat pumps need to operate, but with unified control. However, when extracting heat locally, it is necessary to control the air source heat pumps separately according to the heating load, and the control logic is complex.

We further clarified the operational challenges in the paper as follows:

“3.3. Comparative analysis of heat extraction methods

…

For the uniform heat extraction method, all ASHPs need to operate with unified control. However, when extracting heat locally, it is necessary to control the ASHPs separately according to the heating load, and the control logic is complex….”

6. Can the impact of heat extraction methods on system maintenance and lifecycle costs be addressed?

Response: Thank you for your good question.

When the heating system adopts the localized heat extraction method, the air source heat pump works alternately, resulting in lower equipment maintenance costs. Figure 8 indicates that the uniform heating method brings more significant energy-saving effects to the cooling and heating systems. Therefore, the maintenance cost of uniform heating is lower, and so is the lifecycle cost. We added this paragraph to the article and highlighted it in red, as follows:

“3.3. Comparative analysis of heat extraction methods

…

…For economic performance, when the heating system adopts the localized heat extraction method, ASHPs work alternately, resulting in lower equipment maintenance costs. Fig. 8 indicates that the uniform heating method brings more significant energy-saving effects to the cooling and heating systems. Therefore, the maintenance cost of uniform heating is lower, and so is the lifecycle cost.”

7. How do variations in chilled water supply temperature and air supply temperature affect the system’s ability to meet cooling demands? 

Response: Thank you for your professional question.

The research results show that within a certain range, as the supply temperature of chilled water increases, the energy consumption of the data center cooling system will first decrease and then increase. When the temperature of the chilled water supply is low, the energy consumption of the cooling system mainly depends on the cooling tower, chiller, and lake water pump, etc. When the temperature of the chilled water is high, the energy consumption of the chilled water pump and fans of CRAHs plays a key role. When the temperature in the server room is low, the energy consumption of the refrigeration equipment is high. When the supply air temperature in the server room is high, the energy consumption of the fans of CRAHs accounts for a larger proportion of the energy consumption of the cooling system. Therefore, the supply temperature of chilled water and the air supply temperature cannot be too low, as the capacity of the refrigeration equipment may not meet the cooling requirements. However, the temperature of both should not be too high, as this may cause non-refrigeration equipment in the cooling system, such as the chilled water pump and fans, to fail to meet operational requirements.

8. What are the practical implications of different water tank volumes on system scalability and long-term efficiency?

Response: Thank you for your thought-provoking question.

A water tank serves as a buffer and stores heat. The load characteristics of the heating system exhibit certain fluctuations, and the air source heat pump may often be in a low energy efficiency state. The presence of the water tank weakens the negative influence caused by this inherent characteristics, improves equipment efficiency, and reduces energy consumption. As the volume of the water tank increases, this effect becomes more significant, so the energy consumption of the system becomes more diminishes. We have added commentary in the paper and highlighted it in red:

“3.4.3. Analysis of the volume of the storage tank

…A water tank serves as a buffer and stores heat. The load characteristics of the heating system exhibit certain fluctuations, and the ASHP may often be in a low energy efficiency state. The presence of the water tank weakens the negative influence caused by this inherent characteristic, improves equipment efficiency, and reduces energy consumption. As the volume of the water tank increases, this effect becomes more significant, so the energy consumption of the system becomes more diminishes….”

Suggestions:

1. Clarify the methodology used to ensure independence of heating and cooling systems in scenarios a1 and a2.

Response: Thank you for your good comment.

Due to the lack of a centralized heating system in Chenzhou City, Hunan Province, the air source heat pump heating system is currently the main heating method. Therefore, in this study, the basic scenario (Scenario a1) we used for comparison is air source heat pumps utilizing external air for heating. Scenario a2 employs water source heat pumps to recover waste heat from the lake water return, and the lake water will return to the lake. Therefore, the cooling system and heating system of Scenarios a1 and a2 exist independently. We have also added this section in section 3.1 and highlighted it in red, as flowers:

“3.1. Comparative analysis of energy consumption of waste heat recovery locations

Since there is no centralized large-scale heating in Chenzhou, the air source heat pump (ASHP) as a clean and efficient heating system, is currently the main method for building heating and hotel hot water supply in this city. Hence, this study designed a heat pump heating system as a comparison case (Scenario a1)…. It is worth noting that the heating system recovers waste heat from the lake water return in Scenario a2, and the lake water is discharged back to the lake, so it will not affect the cooling system. The heating system in Scenario a3 utilizes the waste heat from the chilled water return, which will affect the performance of the cooling system….”

2. Provide more detailed insights into how waste heat recovery affects operational adjustments in the cooling system in scenarios a3 and a4.

Response: Thank you for your good suggestion.

We have provided a detailed description of the impact of Scenarios a3 and a4 on the operational adjustments in the cooling system, as follows：

“3.1. Comparative analysis of energy consumption of waste heat recovery locations

…

…The main reason for energy conservation in these two scenarios is to reduce the energy consumption of mechanical refrigeration. The energy consumption of mechanical refrigeration is defined as the sum of the energy consumption of cooling towers, the cooling water pump, and the chiller…. while the operational time of the free cooling mode in Scenarios a3 and a4 is 5144 and 5267 hours, reducing by 58.72% and 60.13%, respectively. The mechanical cooling energy consumption of Scenarios a1 and a2 reaches 654.78 MWh, while that of Scenarios a3 and a4 is 479.97 MWh and 480.47 MWh, respectively. Furthermore, compared to Scenario a3, the heating system in Scenario a4 recovers the waste heat from the return air of IT equipment,…Similarly, the COP of the cooling system in Scenarios a1 and a2 is 9.14, while Scenarios a3 and a4 achieve COP of 9.85 and 10.17 respectively due to the recovery of waste heat from the cooling system.

Fig. 5 depicts the PUE and the energy consumption of the cooling and heating systems in the four scenarios….”

3. Include a discussion on the economic feasibility alongside energy savings for each heat recovery volume scenario.

Response: Thank you for your useful suggestion.

For energy saving, as the heating volume grows, the heating system recovers more waste heat, and the cooling system reduces energy consumption. The PUE and COP of the cooling system are also improved. Although the energy consumption of the heating system rises, the energy savings of the heating system also increase. For the economy, the higher the level of waste heat recovery, the lower the operating costs of the cooling system. However, this leads to a growth in the initial investment costs of the heating system. Therefore, increasing the degree of waste heat utilization can achieve the goal of improving the energy efficiency and economy of the cooling system. However, the economic impact of the heating system needs to be analyzed in depth. We have supplemented this discussion in the text and highlighted it in red, as follows:

“3.2. Comparative analysis of waste heat recovery volumes

…

…Specifically, compared to Scenario a1, when the heat supply escalates from 10% to 100%, the cooling energy consumption in Scenarios a3 and a4 is 2496 MWh to 2217 MWh and 2400 MWh to 1925 MWh, diminishing by 11.16% and 19.80%, respectively….

Fig. 7 indicates the trends of PUE and COP in Scenarios a3 and a4 at different levels of waste heat recovery. Whether it is recovering waste heat from chilled return water or return air of IT equipment, with the enhancement of the degree of waste heat utilization, the implementer of the PUE and COP of the DC. Specifically, recovering waste heat from chilled water return can achieve an increase in PUE from 1.199 to 1.186 and an increase in COP from 9.54 to 10.73. The effect of recovering waste heat from IT equipment return air is more significant. It can increase PUE from 1.194 to 1.173 and COP from 9.92 to 12.36.

Undoubtedly, the improvement of waste heat recovery level has enhanced the PUE and COP of the cooling system in the DC, and the economy of the cooling system has increased due to the reduction of operation and maintenance costs. However, the growth in heating volume leads to an increase in the initial investment cost of the heating system. Therefore, the impact of the game relationship between waste heat recovery to improve energy efficiency and higher investment costs on the economy of heating systems needs to be analyzed in depth in the future.”

4. Explain the rationale behind choosing scenarios b1 to b5 and how these scenarios contribute to the overall system efficiency.

Response: Thank you for your useful suggestion.

We have provided a detailed description of the rationale behind choosing scenarios b1 to b5 and how these scenarios contribute to the overall system efficiency, as follows：

“3.2. Comparative analysis of waste heat recovery volumes

…To maximize the utilization of waste heat and ensure that it fully covers the heat demands, the expansion is carried out based on the heating load of one hotel until the peak heat load equals the valley value of the DC cooling load. It is named 100% heat supply (Scenario b5)….

…Specifically, compared to Scenario a1, when the heat supply escalates from 10% to 100%, the cooling energy consumption in Scenarios a3 and a4 is 2496 MWh to 2217 MWh and 2400 MWh to 1925 MWh, diminishing by 11.16% and 19.80%, respectively…. It means that the more heat is recovered from both the return air of IT equipment and the chilled return water, the lower the energy consumption of the cooling system and PUE and the higher the energy efficiency…. even though the temperature of the return air of IT equipment is higher than the temperature of chilled return water.

Fig. 7 indicates the trends of PUE and COP changes in Scenarios a3 and a4 at different levels of waste heat recovery. Whether it is recovering waste heat from chilled water return or return air of IT equipment, with the enhancement of the degree of waste heat utilization, the implementer of the PUE and COP of the DC. Specifically, recovering waste heat from chilled water return can achieve an increase in PUE from 1.199 to 1.186 and an increase in COP from 9.54 to 10.73. The effect of recovering waste heat from IT equipment return air is more significant. It can increase PUE from 1.194 to 1.173 and COP from 9.92 to 12.36.”

5. Provide a comparative analysis of the efficiency and scalability of localized versus uniform heat extraction methods.

Response: Thank you for your professional suggestion.

We have supplemented a detailed description of the methodology of two heat extraction methods and analysis, as follows：

“3.3. Comparative analysis of heat extraction methods

…localized…Localized heat extraction (Scenario c1) involves the ASHPs evenly recovering waste heat from the air inlet of partial CRAHs, the number of ASHPs is determined based on the heating capacity of the building….ASHPs…(37 units)…

…under the two heat extraction methods, as the waste heat recovery level increases, the heating system’s energy consumption becomes higher due to the growth amount of heating. The energy consumption of the DC cooling system drops continuously. For the same heating demand, the cooling and heating energy consumption of Scenario c1 is lower than that of Scenario c2. Therefore, localized heat recovery of IT equipment waste heat will achieve higher COP….localized…Compared with localized heat extraction, uniform heat extraction is a multi-unit group control regulation, which subjects to the weaknesses such as small unit capacity, low overall performance coefficient, and low operating load….”

6. Discuss potential trade-offs between energy savings and system complexity introduced by each method.

Response: Thank you for your professional suggestion.

For the heating system, recovering waste heat from lake water, chilled water return, and IT equipment return air can all reduce the system's energy consumption. Due to the higher energy efficiency of WSHPs compared to ASHPs, the return air temperature of IT equipment is relatively high, making the heating system for these waste heat recovery scenarios simpler. For the cooling system, longer pipelines will increase the uncertainty of the water system. Especially, the recovery of waste heat from IT equipment requires the addition of air ducts, which may potentially affect its thermal environment. Therefore, the potential trade-off between the improvement of energy efficiency and increase system complexity needs to be considered. We summarized the game relationship between the benefits of waste heat recovery and increasing system complexity at the end of section 3.3, as follows:

“3.3. Comparative analysis of heat extraction methods

…

…It is worth noting that the recovery of waste heat from IT equipment requires the addition of air ducts, which may potentially affect its thermal environment. Therefore, the potential trade-off between the improvement of energy efficiency and increase system complexity needs to be considered.”

7. Expand on the economic implications of optimizing chilled water supply temperatures and air supply temperatures.

Response: Thank you for your good suggestion.

Figure 9 indicates that the total energy consumption of the system diminishes first and then increases with the increase of chilled water supply temperature, reaching its lowest value at 16 ℃. Therefore, appropriately raising the supply temperature of chilled water can improve the economy of the system, but excessively high chilled water has a negative impact on the economy. Conversely, Figure 10 shows that as the air supply temperature grows, the total energy consumption of the system continues to rise, which is not conducive to improving the economy. Based on the analysis of these two parameters, Figure 12 displays the energy-saving performance of the improved Scenario a3. Setting the temperature of the chilled water and the air supply to 16 ℃ and 20 ℃ reduced energy consumption by 26.05%, greatly improving the economic performance of the system. We added the discussion of the economic implications of optimizing these two parameters, as follows:

“3.4.1. Analysis of chilled water supply temperature

…Therefore, appropriately raising the supply temperature of chilled water reduces the operational costs of the system, but excessively high chilled water has a negative impact on the economy.

3.4.2. Analysis of the air supply temperature in the server rooms

…This has led to a significant increase in total energy consumption, which hurt the economy and the environment.

…

…It greatly drops the operating costs of the system and improves environmental benefits….”

8. Include a discussion on the environmental benefits and potential trade-offs associated with adjusting system parameters.

Response: Thank you for your good suggestion.

Due to the close relationship between environmental benefits and system energy consumption, properly raising the supply temperature of chilled water can reduce the carbon emissions caused by the energy consumption of the system. Raising the air supply temperature increases carbon emissions. Increasing the volume of the water tank achieves the goal of improving environmental benefits, but it also entails increased initial investment costs for the system. Add the following content:

“3.4.1. Analysis of chilled water supply temperature

…Properly raising the supply temperature of chilled water can reduce the carbon emissions caused by the energy consumption of the system….

3.4.2. Analysis of the air supply temperature in the server rooms

…This has led to a significant increase in total energy consumption, which hurt the economy and the environment.

…

3.4.3. Analysis of the volume of the storage tank

…The presence of the water tank weakens the inherent characteristics of the heating system, improves equipment efficiency, and reduces energy consumption and carbon emissions….”

9. Ensure clarity in defining terms such as PUE (Power Usage Effectiveness) and explain their significance in the context of your study.

Response: Thank you for your good suggestion.

We have added the importance of PUE for evaluating data center cooling systems and its definition, as follows：

“2.4. PUE analysis

An important metric, that is used to determine the productivity level of the DC, is the PUE index. It is the most common benchmark for measuring the power effectiveness of the DC. The PUE index indicates the ratio of total power consumed by the data center to the power consumption of IT equipment….”

10. Provide more context on the limitations and assumptions of your simulations, particularly regarding their applicability to real-world scenarios.

Response: Thank you for your good suggestion.

Table 4 lists the simulation condition settings and we have added the limitations assumptions and the applicability, as follows：

“3. Results and discussions

…

…Due to the location of the DC in a 5A-level scenic area, hotel construction is planned and it requires year-round heating. The hotel has 130 rooms with an area of 5840 m², and when calculating its heating load, it is assumed that its occupancy rate is 100%. Simulation condition parameters are listed in Table 4. In this paper, METEORORM software is used to export the dry bulb temperature, wet bulb temperature, and other data of typical meteorological years in Chenzhou, and the meteorological component in TRNSYS is employed to read them. The heat loss, pipeline resistance loss, and equipment performance degradation are ignored….”

11. Consider integrating graphical data (figures) more seamlessly with the text to enhance readability and understanding.

Response: Thank you for your useful suggestion. We have checked the figures and the text description to match them.

Reviewer 2 Report

Comments and Suggestions for Authors

Some suggestions can be considered and addressed by the authors: 

1. Please improve Fig. 1, in order to highlight the difference amond three configurations. i think you can consider to merge Fig 1(a) and (b), since they are all for the chilled water temperature lower than the lake temperature, and only difference is the cooling capacity.  Several pump directions in these figures are wrong, please rectify.  Also, plese make the description clearer. 

2. The assumptions for the simulation model have been missed.

3. Whether the simualtion results have been validated against any experimental data ? or partially validated against the published experimental data ? 

4. It is suggested to add and compare the COPs with various scenarios in the result discussion. 

5. There is no description of Fig.3 in the paragraph. 

6. Please quantify all your key discussion results in the conclusion, including the energy consumption and COP. 

Author Response

Reviewer 2

1. Please improve Fig. 1, in order to highlight the difference amond three configurations. i think you can consider to merge Fig 1(a) and (b), since they are all for the chilled water temperature lower than the lake temperature, and only difference is the cooling capacity.  Several pump directions in these figures are wrong, please rectify.  Also, please make the description clearer. 

Response: Thank you for your careful review.

Fig. 1 illustrates the specific arrangements of three heat extraction locations in this study for comparison with the scenario where waste heat is not recovered. Fig. 1(a) displays heat extraction from lake water after heat exchange with chilled water return (free cooling mode and hybrid cooling mode); Fig. 1(b) shows the extraction of heat from the chilled return water after heat exchange with the return air of IT equipment; Fig. 1(c) indicates heat extraction from the return air of IT equipment. We have corrected the errors in the diagram and provided detailed explanations about the system, as flowers:

“2.1. System description

…

…Fig. 1(a) displays heat extraction from lake water after heat exchange with chilled return water (free cooling mode and hybrid cooling mode); Fig. 1(b) shows the extraction of heat from the chilled return water after heat exchange with the IT equipment return air; Fig. 1(c) indicates heat extraction from the return air of IT equipment. The temperature of waste heat at several locations is different. The temperature of return air is the highest, followed by the chilled return water, and the temperature of the lake water is the lowest….”

2. The assumptions for the simulation model have been missed.

Response: Thank you for your professional comment.

Table 4 lists the simulation condition settings and we have added the limitations assumptions and the applicability, as follows：

“3. Results and discussions

…

…Due to the location of the DC in a 5A-level scenic area, hotel construction is planned and it requires year-round heating. The hotel has 130 rooms with an area of 5840 m², and when calculating its heating load, it is assumed that its occupancy rate is 100%. Simulation condition parameters are listed in Table 4. In this paper, METEORORM software is used to export the dry bulb temperature, wet bulb temperature, and other data of typical meteorological years in Chenzhou, and the meteorological component in TRNSYS is employed to read them. The heat loss, pipeline resistance loss, and equipment performance degradation are ignored….”

3. Whether the simulation results have been validated against any experimental data? or partially validated against the published experimental data? 

Response: Thank you for your professional question. We have tested the case data center and the calculated PUE based on the test data and simulation results is close. We added this sentence in the article, as follows:

“3. Results and discussions

…

…It is worth noting that we have tested the case DC and the calculated PUE based on the test data and simulation results is close.”

4. It is suggested to add and compare the COPs with various scenarios in the result discussion. 

Response: Thank you for your useful suggestion.

COP is indeed an important indicator for evaluating the efficiency of cooling systems. We have added the definition of COP in Section 2.5 and supplemented the discussion on the impact of different waste heat recovery locations and levels on the COP of data center cooling systems, as follows:

“2.5. COP analysis

Coefficient of performance (COP) refers to the cooling capacity that can be obtained per unit of power consumption and is a significant technical indicator of refrigeration systems. A high coefficient of refrigeration performance indicates a high energy utilization efficiency of the refrigeration system. It can be calculated using Eq. (13).

                                                                                               

where L_{DC, cooling} is the cooling load of the DC, kWh; P_cooling is the energy consumption of the cooling system of the DC, kWh.

…

3.1. Comparative analysis of energy consumption of waste heat recovery locations

…

…Similarly, the COP of the cooling system in Scenarios a1 and a2 is 9.14, while Scenarios a3 and a4 achieve COP of 9.85 and 10.17 respectively due to the recovery of waste heat from the cooling system.

…

3.2. Comparative analysis of waste heat recovery volumes

…

Fig. 7 indicates the trends of PUE and COP changes in Scenarios a3 and a4 at different levels of waste heat recovery. Whether it is recovering waste heat from chilled water return or return air of IT equipment, with the enhancement of the degree of waste heat utilization, the implementer of the PUE and COP of the DC. Specifically, recovering waste heat from chilled water return can achieve an increase in PUE from 1.199 to 1.186 and an increase in COP from 9.54 to 10.73. The effect of recovering waste heat from IT equipment return air is more significant. It can increase PUE from 1.194 to 1.173 and COP from 9.92 to 12.36.”

3.3. Comparative analysis of heat extraction methods

…

…under the two heat extraction methods, as the waste heat recovery level increases, the heating system’s energy consumption becomes higher due to the growth amount of heating. The energy consumption of the DC cooling system drops continuously. For the same heating demand, the cooling and heating energy consumption of Scenario c1 is lower than that of Scenario c2. Therefore, localized heat recovery of IT equipment waste heat will achieve higher COP….”

5. There is no description of Fig.3 in the paragraph. 

Response: Thank you for your careful review.

We have added a description of Figure 3, as flowers:

“3. Results and discussions

…

…, the lake water temperature remains within the range of 10.86 to 16.66°C throughout the year, as shown in Fig. 3. Therefore, Dongjiang Lake contains abundant cooling resources….”

6. Please quantify all your key discussion results in the conclusion, including the energy consumption and COP. 

Response: Thank you for your professional suggestion.

We have added conclusions about COP in the conclusion section, as follows：

“4. Conclusions

…

(1)          The scenario that the waste heat recovered from the return air of IT equipment achieves the lowest cooling energy consumption, with PUE, cooling system energy savings, and COP of 1.19, 10.06%, and 10.17, respectively.

(2)          In the scenario of maximizing the utilization of waste heat, the energy-saving rate and COP of the cooling system from recovering waste heat from the return air of IT equipment are 19.80% and 12.36, respectively.

(3)…with the best COP of 12.36….”

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The majority of the comments have been carefully addressed and reflected in the revised manuscript. So i suggest the paper can be accepted in the present form.