A flexible liquid cooling control method and system for data center storage

Abstract

The present invention discloses a flexible liquid cooling control method and system for data center storage, which can obtain the total power consumption and total heat generation of the data center in real time; control the input power of the data center and the total cooling capacity of the liquid cooling system based on the total power consumption and the total heat generation, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency; control the cooling capacity output of each liquid cooling unit according to the optimal cooling capacity; and achieve real-time adjustment of the input power of the data center and the cooling capacity of the liquid cooling system according to the total power consumption and the total heat generation of the data center, further adjust the cooling capacity of each liquid cooling unit, improve the utilization efficiency of liquid cooling resources and the cooling efficiency, optimize the overall thermal management efficiency of the data center, effectively reduce the temperature of the data center, and improve the stability and reliability of the data center.

Description

Technical Field

[0001] The present invention relates to the field of thermal management technology, and in particular to a flexible liquid cooling control method and system for data center storage. Background Art

[0002] With the development of data centers, more and more cabinets use energy storage systems as backup power or part of the power supply. However, the current centralized cabinets and centralized energy storage systems are not flexible enough in terms of energy distribution and capacity configuration. This inflexibility may result in the ineffective use of energy, affecting the overall operating efficiency and economy of the data center.

[0003] In existing data center architectures, the liquid cooling system and the energy storage system are usually designed separately. This design leads to low thermal efficiency because the two systems cannot work together to optimize heat transfer and utilization. At the same time, the separate systems increase construction and operating costs, which is not economical.

[0004] In summary, existing data centers have deficiencies in the energy allocation and capacity configuration of energy storage systems and the design of liquid cooling systems. A system that can achieve flexible control, make full use of battery and liquid cooling system resources, and efficiently configure is needed to improve the safety and economy of data centers. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a flexible liquid cooling control method and system for data center storage, so as to achieve efficient and reasonable utilization of liquid cooling resources.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] A flexible liquid cooling control method for data center storage, wherein the data center includes a liquid cooling system, and the liquid cooling system includes multiple groups of liquid cooling units, including the following steps:

[0008] S1. Obtaining the total power consumption and total heat generation of the data center in real time;

[0009] The data center further includes an electrical cabinet system, a cabinet system, multiple groups of DC loads, and multiple groups of AC loads, wherein the electrical cabinet system includes multiple groups of electrical cabinets, and the cabinet system includes multiple groups of cabinets;

[0010] Step S1 is specifically as follows:

[0011] Obtain the power consumption of each cabinet, each liquid cooling unit, each DC load, and each AC load in real time, and obtain the heat generation of each cabinet and each electrical cabinet;

[0012] S2. Controlling the input power of the data center and the total cooling capacity of the liquid cooling system based on the total power consumption and the total heat generation, and calculating the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency;

[0013] Each electric cabinet, each cabinet, each DC load and each AC load is connected to the DC bus through a power converter, and is connected to an AC source and a DC source through the DC bus and the power converter. The AC source includes an AC photovoltaic, a diesel generator and a power grid, and the DC source includes a DC photovoltaic;

[0014] Step S2 comprises the steps of:

[0015] Control the total cooling capacity of the liquid cooling system based on the total heat generation of the data center, and detect the current power consumption period in real time. If it is currently in the off-peak period, execute step A1; if it is currently in the peak period, execute step A2;

[0016] A1. Control the input power of the data center to be provided entirely by the power grid, control the AC photovoltaic and the DC photovoltaic to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0017] A2. Control the input power of the data center to be provided by the electric cabinet system, and control the DC photovoltaic, the AC photovoltaic and the diesel generator to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0018] S3. Control the cooling capacity output of each liquid cooling unit according to the optimal cooling capacity.

[0019] In order to solve the above problem, another technical solution adopted by the present invention is:

[0020] A flexible liquid cooling control system for data center storage, the data center includes a liquid cooling system, and the liquid cooling system includes multiple groups of liquid cooling units;

[0021] The flexible liquid cooling control system for data center storage implements the following steps:

[0022] S1. Obtaining the total power consumption and total heat generation of the data center in real time;

[0023] The data center further includes an electrical cabinet system, a cabinet system, multiple groups of DC loads, and multiple groups of AC loads, wherein the electrical cabinet system includes multiple groups of electrical cabinets, and the cabinet system includes multiple groups of cabinets;

[0024] Step S1 is specifically as follows:

[0025] Obtain the power consumption of each cabinet, each liquid cooling unit, each DC load, and each AC load in real time, and obtain the heat generation of each cabinet and each electrical cabinet;

[0026] S2. Controlling the input power of the data center and the total cooling capacity of the liquid cooling system based on the total power consumption and the total heat generation, and calculating the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency;

[0027] Each electric cabinet, each cabinet, each DC load and each AC load is connected to the DC bus through a power converter, and is connected to an AC source and a DC source through the DC bus and the power converter. The AC source includes an AC photovoltaic, a diesel generator and a power grid, and the DC source includes a DC photovoltaic;

[0028] Step S2 comprises the steps of:

[0029] Control the total cooling capacity of the liquid cooling system based on the total heat generation of the data center, and detect the current power consumption period in real time. If it is currently in the off-peak period, execute step A1; if it is currently in the peak period, execute step A2;

[0030] A1. Control the input power of the data center to be provided entirely by the power grid, control the AC photovoltaic and the DC photovoltaic to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0031] A2. Control the input power of the data center to be provided by the electric cabinet system, and control the DC photovoltaic, the AC photovoltaic and the diesel generator to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0032] S3. Control the cooling capacity output of each liquid cooling unit according to the optimal cooling capacity.

[0033] The beneficial effects of the present invention are as follows: a flexible liquid cooling control method and system for data center storage of the present invention adjusts the input power of the data center and the cooling capacity of the liquid cooling system in real time according to the total power consumption and total heat generation of the data center, further adjusts the cooling capacity of each liquid cooling unit, and improves the utilization efficiency of liquid cooling resources and the cooling efficiency; at the same time, for peak and valley power periods, the data center adopts different control strategies to optimize the overall thermal management efficiency of the data center, effectively reduce the temperature of the data center, and improve the stability and reliability of the data center. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1This is a flow chart of a flexible liquid cooling control method for data center storage in this embodiment;

[0035] Figure 2 This is a framework diagram of a flexible liquid cooling control system for data center storage in this embodiment;

[0036] Figure 3 This is an overall structural diagram of a flexible liquid cooling control system for data center storage in this embodiment;

[0037] Figure 4 An extended connection diagram of a flexible liquid cooling control system for multiple data centers in this embodiment;

[0038] Description of labels:

[0039] 1. Flexible liquid cooling control system for data center storage. DETAILED DESCRIPTION

[0040] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0041] Please refer to Figure 1 A flexible liquid cooling control method for data center storage, wherein the data center includes a liquid cooling system, and the liquid cooling system includes multiple groups of liquid cooling units, including the steps of:

[0042] S1. Obtaining the total power consumption and total heat generation of the data center in real time;

[0043] The data center further includes an electrical cabinet system, a cabinet system, multiple groups of DC loads, and multiple groups of AC loads, wherein the electrical cabinet system includes multiple groups of electrical cabinets, and the cabinet system includes multiple groups of cabinets;

[0044] Step S1 is specifically as follows:

[0045] Obtain the power consumption of each cabinet, each liquid cooling unit, each DC load, and each AC load in real time, and obtain the heat generation of each cabinet and each electrical cabinet;

[0046] S2. Controlling the input power of the data center and the total cooling capacity of the liquid cooling system based on the total power consumption and the total heat generation, and calculating the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency;

[0047] Each electric cabinet, each cabinet, each DC load and each AC load is connected to the DC bus through a power converter, and is connected to an AC source and a DC source through the DC bus and the power converter. The AC source includes an AC photovoltaic, a diesel generator and a power grid, and the DC source includes a DC photovoltaic;

[0048] Step S2 comprises the steps of:

[0049] Control the total cooling capacity of the liquid cooling system based on the total heat generation of the data center, and detect the current power consumption period in real time. If it is currently in the off-peak period, execute step A1; if it is currently in the peak period, execute step A2;

[0050] A1. Control the input power of the data center to be provided entirely by the power grid, control the AC photovoltaic and the DC photovoltaic to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0051] A2. Control the input power of the data center to be provided by the electric cabinet system, and control the DC photovoltaic, the AC photovoltaic and the diesel generator to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0052] S3. Control the cooling capacity output of each liquid cooling unit according to the optimal cooling capacity.

[0053] From the above description, it can be seen that the beneficial effects of the present invention are: a flexible liquid cooling control method for data center storage of the present invention adjusts the input power of the data center and the cooling capacity of the liquid cooling system in real time according to the total power consumption and total heat generation of the data center, and further adjusts the cooling capacity of each liquid cooling unit to improve the utilization efficiency of liquid cooling resources and the cooling efficiency; at the same time, for peak and valley power periods, the data center adopts different control strategies to optimize the overall thermal management efficiency of the data center, effectively reduce the temperature of the data center, and improve the stability and reliability of the data center.

[0054] Furthermore, step S3 includes the following steps:

[0055] S4. Control the data center to enable a first preset number of electrical cabinets and a second preset number of cabinets, and cool the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units.

[0056] From the above description, it can be seen that according to actual demand, a specific number of electrical cabinets and cabinets and a certain number of refrigeration units are enabled. Each refrigeration unit outputs the optimal cooling capacity to ensure that the total cooling capacity is accurately matched with the actual heat dissipation needs of the data center. On the one hand, it avoids overheating of equipment caused by insufficient cooling, affecting performance and life; on the other hand, it also prevents energy waste caused by excessive cooling, reduces the burden on a single refrigeration unit, and improves refrigeration utilization.

[0057] Furthermore, step S4 is specifically as follows:

[0058] If it is currently a valley power period, controlling the data center to enable a second preset number of cabinets, and cooling the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units;

[0059] Otherwise, the data center is controlled to enable a first preset number of electrical cabinets and a second preset number of cabinets, and the data center is cooled based on the optimal cooling capacity by using a third preset number of liquid cooling units.

[0060] From the above description, it can be seen that during the off-peak period, the power grid is used for power supply, the power cabinet is not enabled, and the liquid cooling system only needs to cool the cabinet; during the peak period, the power cabinet is turned on and charged by other low-cost methods except the power grid. At this time, the liquid cooling system needs to cool the cabinet and the power cabinet. That is, different control measures are adopted according to different power consumption periods to control the reasonable allocation of equipment resources, reduce electricity costs, improve energy utilization efficiency, and enhance the overall operating efficiency of the data center.

[0061] Furthermore, during the off-peak period, the optimal cooling capacity is calculated as follows:

[0062] ;

[0063] ;

[0064] in:

[0065] … ;

[0066] ;

[0067] ;

[0068] … ;

[0069] ;

[0070] ;

[0071] get:

[0072] ;

[0073] ;

[0074] Among them, A1~A i is the power consumption of cabinet 1~i, E1~E i is the heat generated by cabinet 1~i, O is a constant, indicating A i With E iThere is a conversion relationship between C1 and C j is the power consumption of water cooling units 1~j, G1~G j is the cooling capacity of water cooling unit 1~j, Q is a constant, indicating C j With G j There is a conversion relationship between them, H1 is the total power consumption of DC load, H2 is the total power consumption of AC load, and D1 is the output power of the power grid.

[0075] From the above description, it can be seen that due to the different power consumption periods, the control strategies adopted by the data center are also different. During the off-peak period, the data center does not use the power cabinet, and its input power is directly provided by the power grid. Therefore, the total power consumption needs to consider the cabinet, liquid cooling unit, DC load and AC load, and the total heat generation only considers the cabinet.

[0076] Furthermore, during the peak electricity period, the optimal cooling capacity is calculated as follows:

[0077] ;

[0078] ;

[0079] in:

[0080] … ;

[0081] ;

[0082] ;

[0083] … ;

[0084] ;

[0085] ;

[0086] … ;

[0087] ;

[0088] ;

[0089] Right now:

[0090] ;

[0091]

[0092] Bring in:

[0093] ;

[0094] ;

[0095] ;

[0096] get:

[0097] ;

[0098] ;

[0099] When the heat output of n electrical cabinets and i cabinets is the same, the thermal efficiency of the water cooling unit is the highest, that is, ,but:

[0100] ;

[0101] Among them, B1~B n is the power consumption of the electric cabinet 1~n, F1~F n is the heat generated by the cabinet 1~n, P is a constant, indicating B n With F n There is a conversion relationship between A1 and A i is the power consumption of cabinet 1~i, E1~E i is the heat generated by cabinet 1~i, O is a constant, indicating A i With E i There is a conversion relationship between C1 and C j is the power consumption of water cooling units 1~j, G1~G j is the cooling capacity of water cooling unit 1~j, Q is a constant, indicating C j With G j There is a conversion relationship between them, H1 is the total power consumption of DC load, and H2 is the total power consumption of AC load.

[0102] From the above description, it can be seen that due to the different power consumption periods, the control strategies adopted by the data center are also different. During the peak power period, the data center does not use the power grid to provide input power but uses the switch cabinet. Therefore, all input power is provided by the switch cabinet. The total power consumption needs to consider the cabinet, liquid cooling unit, DC load and AC load, and the total heat generation needs to consider the switch cabinet and cabinet.

[0103] Please refer to Figure 2 , a flexible liquid cooling control system for data center storage, the data center includes a liquid cooling system, the liquid cooling system includes a plurality of liquid cooling units;

[0104] The flexible liquid cooling control system for data center storage implements the following steps:

[0105] S1. Obtaining the total power consumption and total heat generation of the data center in real time;

[0106] The data center further includes an electrical cabinet system, a cabinet system, multiple groups of DC loads, and multiple groups of AC loads, wherein the electrical cabinet system includes multiple groups of electrical cabinets, and the cabinet system includes multiple groups of cabinets;

[0107] Step S1 is specifically as follows:

[0108] Obtain the power consumption of each cabinet, each liquid cooling unit, each DC load, and each AC load in real time, and obtain the heat generation of each cabinet and each electrical cabinet;

[0109] S2. Controlling the input power of the data center and the total cooling capacity of the liquid cooling system based on the total power consumption and the total heat generation, and calculating the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency;

[0110] Each electric cabinet, each cabinet, each DC load and each AC load is connected to the DC bus through a power converter, and is connected to an AC source and a DC source through the DC bus and the power converter. The AC source includes an AC photovoltaic, a diesel generator and a power grid, and the DC source includes a DC photovoltaic;

[0111] Step S2 comprises the steps of:

[0112] Control the total cooling capacity of the liquid cooling system based on the total heat generation of the data center, and detect the current power consumption period in real time. If it is currently in the off-peak period, execute step A1; if it is currently in the peak period, execute step A2;

[0113] A1. Control the input power of the data center to be provided entirely by the power grid, control the AC photovoltaic and the DC photovoltaic to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0114] A2. Control the input power of the data center to be provided by the electric cabinet system, and control the DC photovoltaic, the AC photovoltaic and the diesel generator to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0115] S3. Control the cooling capacity output of each liquid cooling unit according to the optimal cooling capacity.

[0116] From the above description, it can be seen that the beneficial effects of the present invention are: a flexible liquid cooling control system for data center storage of the present invention adjusts the input power of the data center and the cooling capacity of the liquid cooling system in real time according to the total power consumption and total heat generation of the data center, and further adjusts the cooling capacity of each liquid cooling unit to improve the utilization efficiency of liquid cooling resources and the cooling efficiency; at the same time, for peak and valley power periods, the data center adopts different control strategies to optimize the overall thermal management efficiency of the data center, effectively reduce the temperature of the data center, and improve the stability and reliability of the data center.

[0117] Furthermore, step S3 includes the following steps:

[0118] S4. Control the data center to enable a first preset number of electrical cabinets and a second preset number of cabinets, and cool the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units.

[0119] From the above description, it can be seen that according to actual demand, a specific number of electrical cabinets and cabinets and a certain number of refrigeration units are enabled. Each refrigeration unit outputs the optimal cooling capacity to ensure that the total cooling capacity is accurately matched with the actual heat dissipation needs of the data center. On the one hand, it avoids overheating of equipment caused by insufficient cooling, affecting performance and life; on the other hand, it also prevents energy waste caused by excessive cooling, reduces the burden on a single refrigeration unit, and improves refrigeration utilization.

[0120] Furthermore, step S4 is specifically as follows:

[0121] If it is currently a valley power period, controlling the data center to enable a second preset number of cabinets, and cooling the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units;

[0122] Otherwise, the data center is controlled to enable a first preset number of electrical cabinets and a second preset number of cabinets, and the data center is cooled based on the optimal cooling capacity by using a third preset number of liquid cooling units.

[0123] From the above description, it can be seen that during the off-peak period, the power grid is used for power supply, the power cabinet is not enabled, and the liquid cooling system only needs to cool the cabinet; during the peak period, the power cabinet is turned on and charged by other low-cost methods except the power grid. At this time, the liquid cooling system needs to cool the cabinet and the power cabinet. That is, different control measures are adopted according to different power consumption periods to control the reasonable allocation of equipment resources, reduce electricity costs, improve energy utilization efficiency, and enhance the overall operating efficiency of the data center.

[0124] Furthermore, during the off-peak period, the optimal cooling capacity is calculated as follows:

[0125] ;

[0126] ;

[0127] in:

[0128] … ;

[0129] ;

[0130] ;

[0131] … ;

[0132] ;

[0133] ;

[0134] get:

[0135] ;

[0136] ;

[0137] Among them, A1~A i is the power consumption of cabinet 1~i, E1~E i is the heat generated by cabinet 1~i, O is a constant, indicating A i With E i There is a conversion relationship between C1 and C j is the power consumption of water cooling units 1~j, G1~G j is the cooling capacity of water cooling unit 1~j, Q is a constant, indicating C j With G j There is a conversion relationship between them, H1 is the total power consumption of DC load, H2 is the total power consumption of AC load, and D1 is the output power of the power grid.

[0138] From the above description, it can be seen that due to the different power consumption periods, the control strategies adopted by the data center are also different. During the off-peak period, the data center does not use the power cabinet, and its input power is directly provided by the power grid. Therefore, the total power consumption needs to consider the cabinet, liquid cooling unit, DC load and AC load, and the total heat generation only considers the cabinet.

[0139] Furthermore, during the peak electricity period, the optimal cooling capacity is calculated as follows:

[0140] ;

[0141] ;

[0142] in:

[0143] … ;

[0144] ;

[0145] ;

[0146] … ;

[0147] ;

[0148] ;

[0149] … ;

[0150] ;

[0151] ;

[0152] Right now:

[0153] ;

[0154]

[0155] Bring in:

[0156] ;

[0157] ;

[0158] ;

[0159] get:

[0160] ;

[0161] ;

[0162] When the heat output of n electrical cabinets and i cabinets is the same, the thermal efficiency of the water cooling unit is the highest, that is, ,but:

[0163] ;

[0164] Among them, B1~B n is the power consumption of the electric cabinet 1~n, F1~F n is the calorific value of the electric cabinet 1~n, P is a constant, indicating that there is a conversion relationship between Bn and Fn, A1~A iis the power consumption of cabinet 1~i, E1~E i is the heat generated by cabinet 1~i, O is a constant, indicating A i With E i There is a conversion relationship between C1 and C j is the power consumption of water cooling units 1~j, G1~G j is the cooling capacity of water cooling unit 1~j, Q is a constant, indicating C j With G j There is a conversion relationship between them, H1 is the total power consumption of DC load, and H2 is the total power consumption of AC load.

[0165] From the above description, it can be seen that due to the different power consumption periods, the control strategies adopted by the data center are also different. During the peak power period, the data center does not use the power grid to provide input power but uses the switch cabinet. Therefore, all input power is provided by the switch cabinet. The total power consumption needs to consider the cabinet, liquid cooling unit, DC load and AC load, and the total heat generation needs to consider the switch cabinet and cabinet.

[0166] A flexible liquid cooling control method and system for data center storage provided by the present invention are suitable for flexible control of an overall liquid cooling system of a data center equipped with an energy storage system.

[0167] Please refer to Figure 1 and Figure 2 , Embodiment 1 of the present invention is:

[0168] A flexible liquid cooling control method for data center storage, wherein the data center includes a liquid cooling system, and the liquid cooling system includes multiple groups of liquid cooling units, including the following steps:

[0169] S1. Obtaining the total power consumption and total heat generation of the data center in real time;

[0170] The data center further includes an electrical cabinet system, a cabinet system, multiple groups of DC loads, and multiple groups of AC loads, wherein the electrical cabinet system includes multiple groups of electrical cabinets, and the cabinet system includes multiple groups of cabinets;

[0171] Step S1 is specifically as follows:

[0172] Obtain the power consumption of each cabinet, each liquid cooling unit, each DC load, and each AC load in real time, and obtain the heat generation of each cabinet and each electrical cabinet;

[0173] S2. Controlling the input power of the data center and the total cooling capacity of the liquid cooling system based on the total power consumption and the total heat generation, and calculating the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency;

[0174] Each electric cabinet, each cabinet, each DC load and each AC load is connected to the DC bus through a power converter, and is connected to an AC source and a DC source through the DC bus and the power converter. The AC source includes an AC photovoltaic, a diesel generator and a power grid, and the DC source includes a DC photovoltaic;

[0175] In this embodiment, the architecture of the data center of the liquid cooling system is composed of multiple groups of electrical cabinets, multiple groups of cabinets, a water cooling system, an AC source, and a DC source, which are hung on a 750V / 1500V DC bus. The source end is composed of n groups of electrical cabinets, several AC sources (grid, diesel generation system and AC photovoltaic) and a DC source (DC photovoltaic), the load end is composed of j groups of water cooling units, a data center, other DC loads and other AC loads, and the data center cabinets have i groups of cabinets.

[0176] In the distribution of the water cooling system, the electrical cabinets and cabinets have been connected to the water cooling system in groups and in parallel. Through design and deployment, the balance of heat load is guaranteed, the optimization of thermal design and the rationalization of pipeline design are guaranteed.

[0177] Step S2 comprises the steps of:

[0178] Control the total cooling capacity of the liquid cooling system based on the total heat generation of the data center, and detect the current power consumption period in real time. If it is currently in the off-peak period, execute step A1; if it is currently in the peak period, execute step A2;

[0179] A1. Control the input power of the data center to be provided entirely by the power grid, control the AC photovoltaic and the DC photovoltaic to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0180] In this embodiment, when the data center area is in a valley power state (battery as backup power), the power cabinet system is used as a DC backup power system instead of a power supply system. The power supply of the data center is guaranteed by the power grid, and its output power is D1. The following conditions should be met to meet the requirements in order to meet the daily work of the data center, as follows:

[0181] Power consumption conditions:

[0182] ;

[0183] Heat dissipation conditions:

[0184] ;

[0185] in:

[0186] … ;

[0187] ;

[0188] ;

[0189] and:

[0190] … ;

[0191] ;

[0192] ;

[0193] That is, the optimal cooling capacity of a single water-cooling unit is:

[0194] ;

[0195] ;

[0196] Among them, A1~A i is the power consumption of cabinet 1~i, E1~E i is the heat generated by cabinet 1~i, O is a constant, indicating A i With E i There is a conversion relationship between C1 and C j is the power consumption of water cooling units 1~j, G1~G j is the cooling capacity of water cooling unit 1~j, Q is a constant, indicating C j With G j There is a conversion relationship between them, H1 is the total power consumption of DC load, H2 is the total power consumption of AC load, and D1 is the output power of the power grid.

[0197] In this embodiment, the cooling object of the water cooling system is only the cabinets, and the cabinets have the same heat generation and heat load, which is already in a high efficiency state for the water cooling system. It is only necessary to ensure that each cabinet and the corresponding water cooling unit meet the above relationship, and then the corresponding number of cabinets and the corresponding number of water cooling units can be combined into one group to achieve efficient management and efficient deployment.

[0198] A2. Control the input power of the data center to be provided by the electric cabinet system, and control the DC photovoltaic, the AC photovoltaic and the diesel generator to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity;

[0199] In this embodiment, when the data center area is in peak power state (battery as the main power supply), the power cabinet supplies power to the data center, and the DC photovoltaic, AC photovoltaic and diesel generation systems charge the power cabinet at the same time. Therefore, the following requirements must be met to ensure the daily work of the data center, as follows:

[0200] Power consumption conditions:

[0201] ;

[0202] Heat dissipation conditions:

[0203] ;

[0204] in:

[0205] … ;

[0206] ;

[0207] ;

[0208] … ;

[0209] ;

[0210] ;

[0211] and:

[0212] … ;

[0213] ;

[0214] ;

[0215] Right now:

[0216] ;

[0217]

[0218] Substitute:

[0219] ;

[0220] ;

[0221] ;

[0222] get:

[0223] ;

[0224] ;

[0225] When the heat output of n electrical cabinets and i cabinets is the same, the thermal efficiency of the small system equipped with a water cooling unit is the highest, that is, , then the optimal cooling capacity of a single water cooling unit is:

[0226] ;

[0227] When the cooling capacity is controlled as in the above formula, the highest cooling efficiency can be achieved;

[0228] Among them, B1~B n is the power consumption of the electric cabinet 1~n, F1~F n is the heat generated by the cabinet 1~n, P is a constant, indicating B n With F n There is a conversion relationship between A1 and A i is the power consumption of cabinet 1~i, E1~E i is the heat generated by cabinet 1~i, O is a constant, indicating A i With E i There is a conversion relationship between C1 and C j is the power consumption of water cooling units 1~j, G1~G j is the cooling capacity of water cooling unit 1~j, Q is a constant, indicating C j With G j There is a conversion relationship between them, H1 is the total power consumption of DC load, and H2 is the total power consumption of AC load.

[0229] S3. Control the cooling capacity output of each liquid cooling unit according to the optimal cooling capacity.

[0230] Please refer to Figures 1 to 4 , Embodiment 2 of the present invention is:

[0231] A flexible liquid cooling control method for data center storage, based on the first embodiment, further comprising the following steps after step S3:

[0232] S4, controlling the data center to enable a first preset number of electrical cabinets and a second preset number of cabinets, and cooling the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units;

[0233] Step S4 is specifically as follows:

[0234] If it is currently a valley power period, controlling the data center to enable a second preset number of cabinets, and cooling the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units;

[0235] Otherwise, the data center is controlled to enable a first preset number of electrical cabinets and a second preset number of cabinets, and the data center is cooled based on the optimal cooling capacity by using a third preset number of liquid cooling units.

[0236] In this embodiment, Figure 3 As shown, when the data center area is in a valley power state (batteries are used as backup power), i groups of cabinets are enabled, and correspondingly, j groups of water cooling units output cooling capacity to the cabinets, other AC loads and other DC loads for heat dissipation. At this time, the i groups of cabinets and the j groups of water cooling units can be combined into a group, and each group of water cooling outputs cooling capacity at an optimal cooling capacity, which can achieve efficient management and efficient liquid cooling resource allocation; when the data center area is in a peak power state (batteries are used as the main power supply), if i cabinets are started, n power cabinets are started to power them, and j groups of water cooling units are started at the same time, and each group of water cooling units outputs cooling capacity at an optimal cooling capacity, thereby achieving optimization of thermal management.

[0237] Furthermore, if Figure 4 , which can be expanded based on multiple data center storage systems to provide more sufficient computing resources.

[0238] Please refer to Figures 2 to 4 , Embodiment 3 of the present invention is:

[0239] A flexible liquid cooling control system for data center storage, the data center includes a liquid cooling system, the liquid cooling system includes multiple groups of liquid cooling units, and the flexible liquid cooling control system for data center storage implements the steps in a flexible liquid cooling control method for data center storage described in embodiment one or two.

[0240] In summary, the present invention provides a flexible liquid cooling control method and system for data center storage, which adjusts the input power of the data center and the cooling capacity of the liquid cooling system in real time according to the total power consumption and total heat generation of the data center, and further adjusts the cooling capacity of each liquid cooling unit to improve the utilization efficiency of liquid cooling resources and the cooling efficiency; at the same time, for peak and valley power periods, the data center adopts different control strategies to optimize the overall thermal management efficiency of the data center, effectively reduce the temperature of the data center, and improve the stability and reliability of the data center.

[0241] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A flexible liquid cooling control method for data center storage, wherein the data center includes a liquid cooling system, and the liquid cooling system includes multiple groups of liquid cooling units, characterized in that: Includes steps: S1. Obtaining the total power consumption and total heat generation of the data center in real time; The data center further includes an electrical cabinet system, a cabinet system, multiple groups of DC loads, and multiple groups of AC loads, wherein the electrical cabinet system includes multiple groups of electrical cabinets, and the cabinet system includes multiple groups of cabinets; Step S1 is specifically as follows: Obtain the power consumption of each cabinet, each liquid cooling unit, each DC load, and each AC load in real time, and obtain the heat generation of each cabinet and each electrical cabinet; S2. Controlling the input power of the data center and the total cooling capacity of the liquid cooling system based on the total power consumption and the total heat generation, and calculating the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency; Each electric cabinet, each cabinet, each DC load and each AC load is connected to the DC bus through a power converter, and is connected to an AC source and a DC source through the DC bus and the power converter. The AC source includes an AC photovoltaic, a diesel generator and a power grid, and the DC source includes a DC photovoltaic; Step S2 comprises the steps of: Control the total cooling capacity of the liquid cooling system based on the total heat generation of the data center, and detect the current power consumption period in real time. If it is currently in the off-peak period, execute step A1; if it is currently in the peak period, execute step A2; A1. Control the input power of the data center to be provided entirely by the power grid, control the AC photovoltaic and the DC photovoltaic to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity; During the off-peak period, the optimal cooling capacity is calculated as follows: D1≥{A1+A2+…+A i }+{C1+C2+…+C j }+H1+H2; {G1+G2+…+G j }≥{E1+E2+…+E i }; in: <h2 style=";text-align:left;direction:ltr">A1=A2=…=A<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ; E1=E2=…=E i ; A1=O×E1; C1=C2=…=C j ; G1=G2=…=G j ; C1=Q×G1; get: D1≥i·A1+j·C1; Among them, A1~Ai is the power consumption of cabinets 1~i, E1~Ei is the calorific value of cabinets 1~i, O is a constant, indicating that there is a conversion relationship between Ai and Ei, C1~Cj is the power consumption of water cooling units 1~j, G1~Gj is the cooling capacity of water cooling units 1~j, Q is a constant, indicating that there is a conversion relationship between Cj and Gj, H1 is the total power consumption of DC loads, H2 is the total power consumption of AC loads, and D1 is the output power of the power grid; A2. Control the input power of the data center to be provided by the electric cabinet system, and control the DC photovoltaic, the AC photovoltaic and the diesel generator to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity; During the peak electricity period, the optimal cooling capacity is calculated as follows: {B1+B2+…+B n }≥{A1+A2+…+A i }+{C1+C2+…+C j }+H1+H2; {G1+G2+…+G i }≥{E1+E2+…+E j }+{F1+F2+…+F n }; in: <h2 style=";text-align:left;direction:ltr">A1=A2=…=A<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ; E1=E2=…=E i ; A1=O×E1; C1=C2=…=C j ; G1=G2=…=G j ; C1=Q×G1; <h2 style=";text-align:left;direction:ltr">B1=B2=…=B<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> ; F1=F2=…=F n ; B1=P×F1; Right now: n·B1≥i·A1+j·C1+H1+H2; i·G1≥j·E1+n·F1; Bring in: A1=O×E1; B1=P×F1; C1=Q×G1; get: When the heat output of n electrical cabinets and i cabinets is the same, the thermal efficiency of the water cooling unit is the highest, that is, n·F1=i·E1, then: Among them, B1~Bn is the power consumption of cabinets 1~n, F1~Fn is the calorific value of cabinets 1~n, P is a constant, indicating that there is a conversion relationship between Bn and Fn, A1~Ai is the power consumption of cabinets 1~i, E1~Ei is the calorific value of cabinets 1~i, O is a constant, indicating that there is a conversion relationship between Ai and Ei, C1~Cj is the power consumption of water cooling units 1~j, G1~Gj is the cooling capacity of water cooling units 1~j, Q is a constant, indicating that there is a conversion relationship between Cj and Gj, H1 is the total power consumption of DC loads, and H2 is the total power consumption of AC loads; S3. Control the cooling capacity output of each liquid cooling unit according to the optimal cooling capacity.

2. A flexible liquid cooling control method for data center storage according to claim 1, characterized in that: Step S3 and subsequent steps include: S4. Control the data center to enable a first preset number of electrical cabinets and a second preset number of cabinets, and cool the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units.

3. A flexible liquid cooling control method for data center storage according to claim 1, characterized in that: Step S4 is specifically as follows: If it is currently a valley power period, controlling the data center to enable a second preset number of cabinets, and cooling the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units; Otherwise, the data center is controlled to enable a first preset number of electrical cabinets and a second preset number of cabinets, and the data center is cooled based on the optimal cooling capacity by using a third preset number of liquid cooling units.

4. A flexible liquid cooling control system for data center storage, characterized in that: The data center includes a liquid cooling system, which includes multiple groups of liquid cooling units. The flexible liquid cooling control system of the data center implements the following steps: S1. Obtaining the total power consumption and total heat generation of the data center in real time; The data center further includes an electrical cabinet system, a cabinet system, multiple groups of DC loads, and multiple groups of AC loads, wherein the electrical cabinet system includes multiple groups of electrical cabinets, and the cabinet system includes multiple groups of cabinets; Step S1 is specifically as follows: Obtain the power consumption of each cabinet, each liquid cooling unit, each DC load, and each AC load in real time, and obtain the heat generation of each cabinet and each electrical cabinet; S2. Controlling the input power of the data center and the total cooling capacity of the liquid cooling system based on the total power consumption and the total heat generation, and calculating the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency; Each electric cabinet, each cabinet, each DC load and each AC load is connected to the DC bus through a power converter, and is connected to an AC source and a DC source through the DC bus and the power converter. The AC source includes an AC photovoltaic, a diesel generator and a power grid, and the DC source includes a DC photovoltaic; Step S2 comprises the steps of: Control the total cooling capacity of the liquid cooling system based on the total heat generation of the data center, and detect the current power consumption period in real time. If it is currently in the off-peak period, execute step A1; if it is currently in the peak period, execute step A2; A1. Control the input power of the data center to be provided entirely by the power grid, control the AC photovoltaic and the DC photovoltaic to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity; During the off-peak period, the optimal cooling capacity is calculated as follows: D1≥{A1+A2+…+A i }+{C1+C2+…+C j }+H1+H2; {G1+G2+…+G j }≥{E1+E2+…+E i }; in: <h2 style=";text-align:left;direction:ltr">A1=A2=…=A<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ; E1=E2=…=E i ; A1=O×E1; C1=C2=…=C j ; G1=G2=…=G j ; C1=Q×G1; get: D1≥i·A1+j·C1; Among them, A1~Ai is the power consumption of cabinets 1~i, E1~Ei is the calorific value of cabinets 1~i, O is a constant, indicating that there is a conversion relationship between Ai and Ei, C1~Cj is the power consumption of water cooling units 1~j, G1~Gj is the cooling capacity of water cooling units 1~j, Q is a constant, indicating that there is a conversion relationship between Cj and Gj, H1 is the total power consumption of DC loads, H2 is the total power consumption of AC loads, and D1 is the output power of the power grid; A2. Control the input power of the data center to be provided by the electric cabinet system, and control the DC photovoltaic, the AC photovoltaic and the diesel generator to charge the electric cabinet system, and calculate the optimal cooling capacity of each liquid cooling unit under the highest cooling efficiency according to the total power consumption, the input power, the total heat generation and the total cooling capacity; During the peak electricity period, the optimal cooling capacity is calculated as follows: {B1+B2+…+B n }≥{A1+A2+…+A i }+{C1+C2+…+C j }+H1+H2; {G1+G2+…+G i }≥{E1+E2+…+E j }+{F1+F2+…+F n }; in: <h2 style=";text-align:left;direction:ltr">A1=A2=…=A<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> ; E1=E2=…=E i ; A1=O×E1; C1=C2=…=C j ; G1=G2=…=G j ; C1=Q×G1; <h2 style=";text-align:left;direction:ltr">B1=B2=…=B<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> ; F1=F2=…=F n ; B1=P×F1; Right now: n·B1≥i·A1+j·C1+H1+H2; i·G1≥j·E1+n·F1; Bring in: A1=O×E1; B1=P×F1; C1=Q×G1; get: When the heat output of n electrical cabinets and i cabinets is the same, the thermal efficiency of the water cooling unit is the highest, that is, n·F1=i·E1, then: Among them, B1~Bn is the power consumption of cabinets 1~n, F1~Fn is the calorific value of cabinets 1~n, P is a constant, indicating that there is a conversion relationship between Bn and Fn, A1~Ai is the power consumption of cabinets 1~i, E1~Ei is the calorific value of cabinets 1~i, O is a constant, indicating that there is a conversion relationship between Ai and Ei, C1~Cj is the power consumption of water cooling units 1~j, G1~Gj is the cooling capacity of water cooling units 1~j, Q is a constant, indicating that there is a conversion relationship between Cj and Gj, H1 is the total power consumption of DC loads, and H2 is the total power consumption of AC loads; S3. Control the cooling capacity output of each liquid cooling unit according to the optimal cooling capacity.

5. A flexible liquid cooling control system for data center storage according to claim 4, characterized in that: Step S3 and subsequent steps include: S4. Control the data center to enable a first preset number of electrical cabinets and a second preset number of cabinets, and cool the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units.

6. A flexible liquid cooling control system for data center storage according to claim 4, characterized in that: Step S4 is specifically as follows: If it is currently a valley power period, controlling the data center to enable a second preset number of cabinets, and cooling the data center based on the optimal cooling capacity by using a third preset number of liquid cooling units; Otherwise, the data center is controlled to enable a first preset number of electrical cabinets and a second preset number of cabinets, and the data center is cooled based on the optimal cooling capacity by using a third preset number of liquid cooling units.

Images