Data center digital twin refrigeration system oriented to various cold sources

Abstract

The invention discloses a data center digital twin refrigeration system oriented to multiple cold sources, and relates to the technical field of data center refrigeration systems, which comprises the following steps: S1, data acquisition; s2, constructing a model; and S3, analysis and decision. According to the data center digital twinborn refrigeration system oriented to the multiple cold sources, a complete data center refrigeration system digital twinborn model is constructed, and a refrigeration system temperature control strategy is given, that is, when the outdoor temperature is lower than the lowest temperature in the transition season, a water chilling unit does not need to be started, and the external cold sources are fully utilized; when the outdoor temperature ranges from the lowest temperature to the highest temperature in the transition season, the fluorine pump is used for refrigeration, and refrigeration energy consumption is reduced; and when the outdoor temperature is higher than the highest temperature in the transition season, the water chilling unit and related refrigeration equipment are started, and through intelligent temperature control, the total energy consumption of the refrigeration system is reduced on the premise that the refrigeration requirement of the machine room is met.

Description

Technical Field

[0001] The present invention relates to the technical field of data center refrigeration systems, and in particular to a data center digital twin refrigeration system for multiple cold sources. Background Art

[0002] The data center refrigeration system is an important support system for maintaining the data center room and key equipment such as servers within an appropriate temperature range. The efficient and stable operation of the data center is inseparable from the support of the refrigeration system, but at the same time, the energy consumption of the refrigeration system itself cannot be ignored. At present, the data center refrigeration system has become the second largest high-energy consumption system after IT equipment. In order to improve the energy efficiency of the data center, building an efficient and intelligent refrigeration system can ensure that on the basis of meeting the daily refrigeration needs of the data center, the overall energy consumption of the refrigeration system is further reduced, the energy efficiency level of the data center is improved, and a green and low-carbon data center is created. In recent years, with the increasingly stringent requirements for the green and low-carbon development of data centers, the industry has actively promoted the development of intelligent and efficient temperature control technology for refrigeration systems, and reduced the overall energy consumption of the refrigeration system while ensuring sufficient supply of cold sources.

[0003] At present, the refrigeration energy-saving method adopted by most data centers during the operation stage is mainly to dynamically adjust the cooling output of the corresponding refrigeration equipment to achieve on-demand distribution of cooling capacity and achieve energy saving. For example: PLC refrigeration machine group control technology of the cold source system: realize variable capacity cooling control and switch the refrigeration system operation mode according to changes in outdoor temperature, and realize energy efficiency optimization based on supply and demand balance. Terminal precision air-conditioning group control technology: the air-conditioning room adopts the master-slave mode to establish a control relationship, and realizes joint control of the air-conditioning room by analyzing the temperature of the air-conditioning supply and return air, such as the addition and subtraction of the air-conditioning machine operation, and the rotation operation to achieve the energy-saving goal. The above two technologies are energy efficiency control optimization technologies for single equipment and local systems, and cannot realize the linkage optimization of the global refrigeration system of the data center.

[0004] Therefore, in view of this, the existing structure and deficiencies are studied and improved, and a digital twin refrigeration system for data centers for multiple cold sources is proposed. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a digital twin refrigeration system for a data center for multiple cold sources, which solves the problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a data center digital twin refrigeration system for multiple cold sources, the data center digital twin refrigeration system for multiple cold sources comprising the following steps:

[0007] S1. Data collection:

[0008] Temperature and humidity sensors, as well as IT equipment load monitors, are deployed in various parts of the data center. For servers, the operating status data of the servers is collected, and sensor data is collected through IoT technology and sent to the central data processing platform;

[0009] S2. Model construction:

[0010] Establish an external cooling source model. The external cooling source refers to outdoor air and water cooling resources, using outdoor air as the cooling source;

[0011] The second step is to establish a cooling system model in the data center computer room. The internal cooling system includes a chiller, a dry cooler or evaporator, and a precision terminal air conditioner.

[0012] The simulation of the refrigeration unit needs to consider the modeling of the refrigerant cycle, compressor, condenser, and evaporator. The data that needs to be collected include refrigeration power, compressor power consumption, COP, chilled water inlet and outlet temperature difference, and cooling water inlet and outlet temperature difference;

[0013] Terminal precision air conditioning needs to consider modeling of supply air temperature control, dehumidification control, air flow organization, and filtration system. The data that needs to be collected include supply air temperature, supply air humidity, wind speed, and filtration efficiency.

[0014] The evaporative cooler model needs to model the evaporative cooling process, and the collected data include direct evaporative cooling efficiency, indirect evaporative cooling efficiency, wet bulb temperature, and dry bulb temperature;

[0015] Based on the above description, by sensing and simulating the outdoor temperature, indoor temperature, and the working status and temperature of each refrigeration system, a relationship model between the working status of the data center refrigeration system and the outdoor temperature and indoor temperature is constructed. By controlling the temperature of the refrigeration equipment, the temperature of the computer room can be accurately controlled, and vice versa. While maintaining the temperature at the specified value, the overall energy consumption of the refrigeration equipment is ensured to be optimal.

[0016] S3. Analysis and decision making:

[0017] The corresponding cooling method is determined according to the indoor and outdoor temperatures and the transition season temperatures. According to the cooling status, the status data of the refrigeration equipment that needs to be monitored is first monitored. When the outdoor temperature is higher than the preset value, the chiller is used for cooling. It is necessary to collect the chiller's chilled water inlet and outlet temperatures, cooling water inlet and return temperatures, chiller power, cooling tower power, outdoor temperature, and indoor temperature of the computer room to build a temperature and power model; when the outdoor temperature is suitable or it is a transition season, turn off the chiller and turn on the fluorine pump for cooling to reduce energy consumption. It is necessary to collect the fluorine pump power, outdoor temperature of the computer room, and indoor temperature to build a temperature and power model; when the outdoor temperature reaches the preset value or it is winter, turn off the refrigeration equipment and use the outdoor cold source for cooling.

[0018] Furthermore, in step S1, the operating status data of the server includes but is not limited to CPU usage, memory usage, temperature, and power consumption.

[0019] Furthermore, in step S2, when outdoor air is used as a cooling source, it is necessary to consider temperature and humidity factors, and it is necessary to collect and monitor the key data of the average temperature value, fluctuation range, and wind speed of the outdoor air and water. Therefore, a monitoring perception module is used to collect data center cooling information, analyze data, and provide real-time feedback functional components.

[0020] Furthermore, in step S2, COP is the cooling efficiency, and in the relationship model between the working state of the refrigeration system and the indoor and outdoor temperatures of the computer room, the control strategy is to dynamically adjust the outlet water temperature of the chiller according to the deviation between the computer room temperature and the set point.

[0021] Furthermore, in step S3, taking into account the working conditions of various refrigeration equipment under different ambient temperatures, the refrigeration equipment of the data center includes but is not limited to chillers and fluorine pumps. Under summer working conditions, the chilled water is cooled by the refrigerant in the chiller, and then heat is exchanged with the return air of the computer room through the terminal air handling unit. Then the fan sends the low-temperature air to the floor outlet, and the cooling water exchanges heat with the high-temperature refrigerant of the chiller, and the heat is dissipated to the outside through the cooling tower. This cycle realizes the refrigeration function.

[0022] Furthermore, in step S3, in transitional seasons or when the outdoor temperature is suitable, the chiller is turned off and a fluorine pump is used to cool the data center room. The fluorine pump is a device that can absorb heat from the environment and transfer it to a desired location.

[0023] Furthermore, in step S3, under mild climate conditions, the fluorine pump can be better used because it utilizes external free energy, namely outdoor air and water cooling resources, which makes its energy efficiency ratio, namely COP, greater than 1.

[0024] Furthermore, in step S3, in winter or when the outdoor temperature is low, the cooling energy consumption is reduced by shutting down various types of cooling machines and using the outdoor cold source to exchange heat with the indoor temperature.

[0025] Furthermore, in step S3, when the corresponding cooling method is used, the following formula is specifically used for selection:

[0026] When T out <T threshold_low At this time, the outdoor cold source of the computer room is directly used for cooling, and there is no need to turn on the chiller and related refrigeration equipment, which minimizes energy consumption;

[0027] When T threshold_low ≤T out ≤Tthreshold_high , use fluorine pump;

[0028] When T out >T threshold_high , the outdoor temperature is higher than the maximum temperature range of the transition season, that is, it is necessary to turn on the chiller to compensate for cooling. At this time, the chiller is turned on;

[0029] Among them, T threshold_low represents the minimum temperature in the transition season, T threshold_high represents the maximum temperature in the transition season, T out Represents the outdoor temperature.

[0030] The present invention provides a data center digital twin refrigeration system for multiple cold sources, which has the following beneficial effects:

[0031] 1. The digital twin refrigeration system for data centers with multiple cold sources constructs a complete digital twin model of the data center refrigeration system and provides a temperature control strategy for the refrigeration system, that is, when the outdoor temperature is lower than the minimum temperature in the transition season, there is no need to turn on the chiller and make full use of the external cold source; when the outdoor temperature is between the minimum and maximum temperatures in the transition season, a fluorine pump is used for refrigeration to reduce refrigeration energy consumption; when the outdoor temperature is higher than the maximum temperature in the transition season, the chiller and related refrigeration equipment are turned on, and through intelligent temperature control, the overall energy consumption of the refrigeration system is reduced while meeting the cooling needs of the computer room. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a schematic diagram of the working principle of a digital twin refrigeration system for a data center facing multiple cold sources according to the present invention;

[0033] Figure 2 A schematic diagram of common refrigeration system-related data of a digital twin refrigeration system for a data center with multiple cold sources according to the present invention;

[0034] Figure 3 This is a schematic diagram of the system architecture of a digital twin refrigeration system for a data center for multiple cold sources according to the present invention;

[0035] Figure 4 This is a schematic diagram of the steps of a digital twin refrigeration system for a data center for multiple cold sources according to the present invention. DETAILED DESCRIPTION

[0036] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0037] like Figure 1-Figure 4As shown, the present invention provides a technical solution: a digital twin refrigeration system for a data center for multiple cold sources, and the digital twin refrigeration system for a data center for multiple cold sources includes the following steps:

[0038] S1. Data collection:

[0039] Temperature and humidity sensors, as well as IT equipment load monitors, are deployed in various parts of the data center. For servers, the operating status data of the servers is collected, and sensor data is collected through IoT technology and sent to the central data processing platform. The operating status data of the servers includes but is not limited to CPU usage, memory usage, temperature, and power consumption;

[0040] S2. Model construction:

[0041] Establish an external cooling source model. The external cooling source refers to outdoor air and water cooling resources, using outdoor air as the cooling source;

[0042] The second step is to establish a cooling system model in the data center computer room. The internal cooling system includes a chiller, a dry cooler or evaporator, and a precision terminal air conditioner.

[0043] The simulation of the refrigeration unit needs to consider the modeling of the refrigerant cycle, compressor, condenser, and evaporator. The data that needs to be collected include refrigeration power, compressor power consumption, COP (refrigeration efficiency), chilled water inlet and outlet temperature difference, and cooling water inlet and outlet temperature difference;

[0044] Terminal precision air conditioning needs to consider modeling of supply air temperature control, dehumidification control, air flow organization, and filtration system. The data that needs to be collected include supply air temperature, supply air humidity, wind speed, and filtration efficiency.

[0045] The evaporative cooler model needs to model the evaporative cooling process. The collected data include direct evaporative cooling efficiency, indirect evaporative cooling efficiency, wet bulb temperature, and dry bulb temperature. Figure 2 As shown in the figure, some common refrigeration system related data that need to be monitored are listed;

[0046] Based on the above description, by sensing and simulating the outdoor temperature, indoor temperature, and the working status and temperature of each refrigeration system, a relationship model between the working status of the data center refrigeration system and the outdoor temperature and indoor temperature is constructed. By controlling the temperature of the refrigeration equipment, the temperature of the computer room can be accurately controlled, and vice versa. While maintaining the temperature at a specified value, the overall energy consumption of the refrigeration equipment is ensured to be optimal. In the relationship model between the working status of the refrigeration system and the indoor and outdoor temperatures of the computer room, the control strategy is to dynamically adjust the outlet water temperature of the chiller according to the deviation between the computer room temperature and the set point;

[0047] When using outdoor air as a cooling source, it is necessary to consider factors such as temperature and humidity. It is necessary to collect and monitor the key data of the average temperature, fluctuation range, and wind speed of outdoor air and water. Therefore, a monitoring and sensing module is used to collect data center cooling information, analyze data, and provide real-time feedback.

[0048] The refrigeration system temperature control and refrigeration energy efficiency optimization algorithm can be specifically described as using a predictive control algorithm, using a machine learning algorithm to predict future load changes and adjust the operating parameters of the refrigeration system in advance. When a significant increase in the power consumption of a server is detected, its heat increase ΔQ is calculated, and the algorithm is used to predict the impact of this heat increase on the overall temperature of the computer room.

[0049] S3. Analysis and decision making:

[0050] The corresponding cooling method is determined according to the indoor and outdoor temperatures and the transition season temperature. According to the cooling status, the status data of the refrigeration equipment that needs to be monitored is first monitored. When the outdoor temperature is higher than the preset value, the chiller is used for cooling. It is necessary to collect the chiller's chilled water inlet and outlet temperatures, cooling water inlet and return temperatures, chiller power, cooling tower power, outdoor temperature, and indoor temperature of the computer room to build a temperature and power model; when the outdoor temperature is suitable or it is a transition season, turn off the chiller and turn on the fluorine pump for cooling to reduce energy consumption. It is necessary to collect the fluorine pump power, outdoor temperature of the computer room, and indoor temperature to build a temperature and power model; when the outdoor temperature reaches the preset value or it is winter, turn off the refrigeration equipment and use outdoor cold source for cooling;

[0051] Considering the working status of various refrigeration equipment under different ambient temperatures, data center refrigeration equipment includes but is not limited to chillers and fluorine pumps. Under summer working conditions, after the chilled water is cooled by the refrigerant in the chiller, it exchanges heat with the return air of the computer room through the terminal air handling unit, and then the fan sends the low-temperature air to the floor outlet. The cooling water exchanges heat with the high-temperature refrigerant of the chiller, and the heat is dissipated to the outside through the cooling tower. This cycle realizes the refrigeration function;

[0052] In transitional seasons or when the outdoor temperature is suitable, the data center room is cooled by shutting down the chiller and using a fluorine pump, a device that absorbs heat from the environment and transfers it to the desired location;

[0053] In mild climate conditions, fluorine pumps can be better used because they use external free energy, namely outdoor air and water cooling resources, which makes their energy efficiency ratio, namely COP, greater than 1, while the energy efficiency ratio of chillers is usually less than 1 because they do not directly use external heat energy;

[0054] In winter or when the outdoor temperature is low, the cooling energy consumption can be reduced by shutting down various types of cooling machines and using outdoor cold sources to exchange heat with indoors.

[0055] When using the corresponding cooling method, the following formula is used for selection:

[0056] When T out <T threshold_low At this time, the outdoor cold source of the computer room is directly used for cooling, and there is no need to turn on the chiller and related refrigeration equipment, which minimizes energy consumption;

[0057] When T threshold_low ≤T out ≤T threshold_high , use fluorine pump;

[0058] When T out >T threshold_high , the outdoor temperature is higher than the maximum temperature range of the transition season, that is, it is necessary to turn on the chiller to compensate for cooling. At this time, the chiller is turned on;

[0059] Among them, T threshold_low represents the minimum temperature in the transition season, T threshold_high represents the maximum temperature in the transition season, T out Represents the outdoor temperature;

[0060] The key to the analysis and decision-making of this system lies in the determination of the transition season temperature range. The selection of the transition season temperature range needs to be determined based on the heat resistance range and heat dissipation of the server in the computer room. When the heat dissipation of the server in the computer room is small and the server's heat resistance is strong, the transition season temperature range is wider, otherwise it is smaller. The transition season temperature range selection will be based on the industry's common practices and the previous working status data of the data center. In order to improve the accuracy of the temperature control and energy efficiency control of the refrigeration system, the transition season temperature range can be flexibly changed, mainly based on the server temperature and indoor temperature.

[0061] Based on the above description, the present invention constructs a complete digital twin model of the data center refrigeration system and provides a temperature control strategy for the refrigeration system, that is, when the outdoor temperature is lower than the minimum temperature of the transition season, there is no need to turn on the chiller and make full use of the external cold source; when the outdoor temperature is between the minimum and maximum temperatures of the transition season, a fluorine pump is used for refrigeration to reduce refrigeration energy consumption; when the outdoor temperature is higher than the maximum temperature of the transition season, the chiller and related refrigeration equipment are turned on, and through intelligent temperature control, the overall energy consumption of the refrigeration system is reduced while meeting the cooling needs of the computer room.

[0062] In summary, when using the digital twin refrigeration system for data centers with multiple cold sources, after the sensors are deployed and the relationship model is constructed, the corresponding cooling method is determined according to the indoor and outdoor temperatures and the transition season temperatures. According to the cooling status, the status data of the refrigeration equipment that needs to be monitored is first monitored. When the outdoor temperature is higher than the preset value, the chiller is used for cooling. It is necessary to collect the chiller's chilled water inlet and outlet temperatures, cooling water inlet and return temperatures, chiller power, cooling tower power, outdoor temperature, and computer room indoor temperature data to build a temperature and power model; when the outdoor temperature is suitable or it is a transition season, turn off the chiller and turn on the fluorine pump for cooling to reduce energy consumption. It is necessary to collect the fluorine pump power, outdoor temperature of the computer room, and indoor temperature to build a temperature and power model; when the outdoor temperature reaches the preset value or it is winter, turn off the refrigeration equipment and use the outdoor cold source for cooling.

[0063] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A digital twin refrigeration system for a data center for multiple cold sources, characterized by: The digital twin refrigeration system for a data center with multiple cold sources includes the following steps: S1. Data collection: Temperature and humidity sensors, as well as IT equipment load monitors, are deployed in various parts of the data center. For servers, the operating status data of the servers is collected, and sensor data is collected through IoT technology and sent to the central data processing platform; S2. Model construction: Establish an external cooling source model. The external cooling source refers to outdoor air and water cooling resources, using outdoor air as the cooling source; The second step is to establish a cooling system model in the data center computer room. The internal cooling system includes a chiller, a dry cooler or evaporator, and a precision terminal air conditioner. The simulation of the refrigeration unit needs to consider the modeling of the refrigerant cycle, compressor, condenser, and evaporator. The data that needs to be collected include refrigeration power, compressor power consumption, COP, chilled water inlet and outlet temperature difference, and cooling water inlet and outlet temperature difference; Terminal precision air conditioning needs to consider modeling of supply air temperature control, dehumidification control, air flow organization, and filtration system. The data that needs to be collected include supply air temperature, supply air humidity, wind speed, and filtration efficiency. The evaporative cooler model needs to model the evaporative cooling process, and the collected data include direct evaporative cooling efficiency, indirect evaporative cooling efficiency, wet bulb temperature, and dry bulb temperature; Based on the above description, by sensing and simulating the outdoor temperature, indoor temperature, and the working status and temperature of each refrigeration system, a relationship model between the working status of the data center refrigeration system and the outdoor temperature and indoor temperature is constructed. By controlling the temperature of the refrigeration equipment, the temperature of the computer room can be accurately controlled, and vice versa. While maintaining the temperature at the specified value, the overall energy consumption of the refrigeration equipment is ensured to be optimal. S3. Analysis and decision making: The corresponding cooling method is determined according to the indoor and outdoor temperatures and the transition season temperatures. According to the cooling status, the status data of the refrigeration equipment that needs to be monitored is first monitored. When the outdoor temperature is higher than the preset value, the chiller is used for cooling. It is necessary to collect the chiller's chilled water inlet and outlet temperatures, cooling water inlet and return temperatures, chiller power, cooling tower power, outdoor temperature, and indoor temperature of the computer room to build a temperature and power model; when the outdoor temperature is suitable or it is a transition season, turn off the chiller and turn on the fluorine pump for cooling to reduce energy consumption. It is necessary to collect the fluorine pump power, outdoor temperature of the computer room, and indoor temperature to build a temperature and power model; when the outdoor temperature reaches the preset value or it is winter, turn off the refrigeration equipment and use the outdoor cold source for cooling.

2. According to claim 1, a digital twin refrigeration system for a data center for multiple cold sources is characterized in that: In step S1, the operating status data of the server includes but is not limited to CPU usage, memory usage, temperature, and power consumption.

3. According to claim 1, a digital twin refrigeration system for a data center facing multiple cold sources is characterized in that: In step S2, when outdoor air is used as a cooling source, it is necessary to consider temperature and humidity factors, and it is necessary to collect and monitor the key data of the average temperature value, fluctuation range, and wind speed of outdoor air and water. Therefore, a monitoring perception module is used to collect data center cooling information, analyze data, and provide real-time feedback functional components.

4. According to claim 1, a digital twin refrigeration system for a data center for multiple cold sources is characterized in that: In step S2, COP is the cooling efficiency, and in the relationship model between the working state of the refrigeration system and the indoor and outdoor temperatures of the computer room, the control strategy is to dynamically adjust the outlet water temperature of the chiller according to the deviation between the computer room temperature and the set point.

5. According to claim 1, a digital twin refrigeration system for a data center for multiple cold sources is characterized in that: In step S3, taking into account the working conditions of various refrigeration equipment under different ambient temperatures, the refrigeration equipment of the data center includes but is not limited to chillers and fluorine pumps. Under summer working conditions, the chilled water is cooled by the refrigerant in the chiller, and then heat is exchanged with the return air of the computer room through the terminal air handling unit. Then the fan sends the low-temperature air to the floor outlet, and the cooling water and the high-temperature refrigerant of the chiller are heat exchanged. The heat is dissipated to the outside through the cooling tower. The refrigeration function is realized in this cycle.

6. According to claim 5, a digital twin refrigeration system for a data center facing multiple cold sources is characterized in that: In step S3, in the transition season or when the outdoor temperature is suitable, the chiller is turned off and a fluorine pump is used to cool the data center room. The fluorine pump is a device that can absorb heat from the environment and transfer it to a desired location.

7. A digital twin refrigeration system for a data center with multiple cold sources according to claim 6, characterized in that: In step S3, under mild climate conditions, the fluorine pump can be better used because it utilizes external free energy, namely outdoor air and water cooling resources, which makes its energy efficiency ratio, namely COP, greater than 1.

8. A digital twin refrigeration system for a data center with multiple cold sources according to claim 7, characterized in that: In step S3, in winter or when the outdoor temperature is low, the cooling energy consumption is reduced by shutting down various types of cooling machines and using the outdoor cold source to exchange heat with the indoor temperature.

9. A digital twin refrigeration system for a data center with multiple cold sources according to claim 8, characterized in that: In step S3, when the corresponding cooling method is used, the following formula is specifically used for selection: When T out <T threshold_low At this time, the outdoor cold source of the computer room is directly used for cooling, and there is no need to turn on the chiller and related refrigeration equipment, which minimizes energy consumption; When T threshold_low ≤T out ≤T threshold_high , use fluorine pump; When T out >T threshold_high , the outdoor temperature is higher than the maximum temperature range of the transition season, that is, it is necessary to turn on the chiller to compensate for cooling. At this time, the chiller is turned on; Among them, T threshold_low represents the minimum temperature in the transition season, T threshold_high represents the maximum temperature in the transition season, T out Represents the outdoor temperature.

Images