Energy-saving liquid cooling system and method for machine room and communication machine room

Abstract

The invention relates to the technical field of liquid cooling systems, and discloses an energy-saving liquid cooling system and method for a machine room and a communication machine room, and the system comprises a thermal analysis module, a conventional circulation module, an emergency response module and an emergency circulation module. The method and the communication machine room correspond to the system. According to the application, the thermal analysis module is combined with the constructed model to generate thermal distribution real-time data to provide a basis for cooling liquid flow direction planning, so that the conventional circulation module generates energy-saving cooling liquid flow direction planning, and heat dissipation of machine room equipment and energy saving of liquid cooling equipment are both considered; the emergency response module generates an emergency cooling liquid flow direction plan when the heat distribution is abnormal, so that the equipment safety is guaranteed; and the emergency circulation module calculates a cooling surplus value and generates a compensation type cooling liquid flow direction plan, so that the utilization rate of the cooling liquid and the energy-saving performance of the system are improved, the defects of the existing liquid cooling technology in the aspects of energy consumption control, environment change coping and the like are overcome, the overall performance of the liquid cooling system is improved, and the requirements of a modern data center for energy conservation and efficient heat dissipation are met.

Description

Technical Field

[0001] The present application relates to the technical field of liquid cooling systems, and specifically to an energy-saving liquid cooling system, method, and communication computer room for a computer room. Background Art

[0002] With the rapid development of information technology, the importance of data centers has become increasingly prominent. The operation of core technologies such as cloud computing, big data, and artificial intelligence carried by them will generate a large amount of heat. At the same time, the continuous increase in the demand for computing and storage capabilities has further increased the generation of heat. In this context, computer room heat dissipation has become a key issue. When the traditional air-cooled heat dissipation method is difficult to meet the heat dissipation requirements of high-density and high-performance devices, liquid cooling technology has emerged.

[0003] Although liquid cooling computer rooms have advantages such as high heat dissipation efficiency, low noise, and low energy consumption, there are still many deficiencies in existing liquid cooling technologies. These deficiencies are further amplified in the context where the scale of the computer room increases and the scale of the liquid cooling system also correspondingly increases. Specifically, the specific liquid cooling system lacks effective means in energy consumption control, resulting in low heat dissipation control efficiency, high overall energy consumption, and inability to meet the strict energy-saving requirements of modern data centers.

[0004] Chinese Patent No. CN118741983B discloses a liquid cooling computer room heat dissipation energy-saving control method. When evaluating energy consumption efficiency, it overly relies on specific machine learning models, and these models have strict requirements for the amount and quality of data. Once the data is defective or the device operating environment changes, the evaluation accuracy will drop significantly; further, when mining the heat dissipation control correlation domain, the scene similarity judgment method is not accurate enough, easily introducing inappropriate historical solutions and interfering with subsequent decisions; the designed optimization algorithm is too complex, consuming a large amount of computing resources and having a slow response, making it difficult to cope with the rapid changes in the operating state of the computer room.

[0005] At the same time, existing technologies generally ignore the influence of factors such as coolant characteristics, thermal interference between devices, and environmental dynamic changes on the heat dissipation and energy-saving effect, further restricting the improvement of the performance of liquid cooling systems.

[0006] In summary, there is an urgent need for a new technical solution for energy-saving liquid cooling in a computer room to solve the above technical problems. Summary of the Invention

[0007] The purpose of the present application is to provide an energy-saving liquid cooling system, method, and communication computer room for a computer room to solve the technical problems raised in the above background art.

[0008] To achieve the above purpose, the present application discloses the following technical solutions:

[0009] In a first aspect, the present application discloses an energy-saving liquid cooling system for a computer room. The system includes a thermal analysis module, a conventional circulation module, an emergency response module, and an emergency circulation module. Among them, the thermal analysis module is communicatively connected to the conventional circulation module, the thermal analysis module is also communicatively connected to the emergency response module, and the emergency response module is communicatively connected to the emergency circulation module.

[0010] The thermal analysis module is configured to input real-time working data into a thermal analysis model to generate real-time thermal distribution data. Among them, the real-time working data includes real-time status data of the operation of computer room equipment. The thermal analysis model is constructed based on machine learning technology and combined with historical working data, corresponding heat generation historical data, and thermal interference historical data. And the thermal analysis model is used to generate the real-time thermal distribution data based on the real-time working data. The real-time thermal distribution data is used to characterize the real-time situation of the thermal distribution in the computer room. This real-time situation is the regional thermal value corresponding to each area in the computer room. The historical working data includes historical status data of the operation of computer room equipment. The heat generation historical data is the historical data of the heat generation of the computer room equipment corresponding to the historical working data. The thermal interference historical data is the thermal interference situation between the computer room equipment and between the computer room equipment and the computer room environment corresponding to the historical working data.

[0011] The conventional circulation module is configured to generate an energy-saving coolant flow plan based on the real-time thermal distribution data. Among them, the energy-saving coolant flow plan is the optimal plan for the coolant flow after comprehensively considering the heat dissipation effect of computer room equipment and the energy-saving effect of liquid cooling equipment. And the energy-saving coolant flow plan is used to conventionally control the flow direction and flow path of the coolant in the computer room.

[0012] The emergency response module is configured to generate an emergency coolant flow plan when there is abnormal thermal distribution data in the real-time thermal distribution data. Among them, the abnormal thermal distribution data is the real-time thermal distribution data that does not meet the preset thermal distribution threshold. The thermal distribution threshold includes an independent thermal threshold for characterizing the normal real-time thermal distribution data of a single computer room equipment and a regional thermal threshold for characterizing the normal real-time thermal distribution data of an area composed of multiple computer room equipment. The emergency coolant flow plan is the optimal plan for the coolant flow corresponding to the best heat dissipation effect of computer room equipment. And the emergency coolant flow plan is used to emergently control the flow direction and flow path of the coolant in the computer room.

[0013] The emergency circulation module is configured to calculate the remaining cooling value of the coolant after implementing the emergency coolant flow plan, and generate a compensated coolant flow plan based on the remaining cooling value; wherein, the remaining cooling value is used to characterize the degree of the cooling function that the coolant still has after implementing the emergency coolant flow plan, the compensated coolant flow plan is the optimal plan for the coolant flow after comprehensively considering the heat dissipation effect of the computer room equipment, the energy-saving effect of the liquid cooling equipment, and the remaining cooling value, and the compensated coolant flow plan is used to compensate and control the flow direction and flow path of the coolant in the computer room.

[0014] Preferably, the construction of the thermal analysis model includes:

[0015] Collect the working historical data, the corresponding heat generation historical data, and the heat interference historical data of the computer room equipment;

[0016] Using machine learning technology, taking the working historical data, the corresponding heat generation historical data, and the heat interference historical data as input features, and the corresponding heat distribution historical data as output labels, training the model, adjusting the parameters of the trained model, so that the error between the prediction result of the model and the actual heat distribution historical data meets the preset prediction error threshold, and obtaining the thermal analysis model.

[0017] Preferably, the generation of the real-time heat distribution data includes:

[0018] Divide the computer room into several areas, calculate the regional thermal values of each area respectively, and define the regional thermal values as the real-time heat distribution data based on the sorting result of the spatial sequence of the areas; wherein, the calculation of the regional thermal value is:

[0019] Obtain the real-time working data of the computer room equipment in the area, input the real-time working data into the thermal analysis model, and match the corresponding thermal weight parameter and thermal correction value for the real-time working data; wherein, the thermal weight parameter and the thermal correction value are the model parameters obtained by regression analysis based on the working historical data, the corresponding heat generation historical data, and the heat interference historical data of the computer room equipment during the training of the thermal analysis model

[0020] Obtain the regional thermal value based on the real-time working data of the area, the corresponding thermal weight parameter, and the thermal correction value.

[0021] Preferably, the generation of the energy-saving coolant flow plan includes:

[0022] Use the energy-saving planning formula to generate the energy-saving coolant flow plan, wherein the energy-saving planning formula is:

[0023]

[0024] Among them, Q i represents the flow rate Q of the coolant flowing into area i i , n represents the total number of areas and i ∈ n, E(Q i ) represents the energy consumption value of the liquid cooling device flowing into area i at the flow rate Q i obtained based on the energy consumption function of the liquid cooling device. The energy consumption function of the liquid cooling device is the energy consumption value at different flow rates obtained based on the energy consumption situation of the liquid cooling device. R i represents the heat dissipation demand coefficient of area i and R i = k 1 *H i + k 2 , k 1 and k 2 are preset heat dissipation adjustment common senses, H i is the area thermal value of area i, C(Q i ) represents the cooling capacity that the coolant flow rate Q i flowing into area i can provide. This cooling capacity is obtained by fitting based on the physical properties and flow rate of the coolant. max() is the maximum value calculation operator, λ is a preset heat dissipation weight coefficient, γ is an energy consumption discrimination coefficient and γ = 1 in the energy-saving planning formula, and F(Q) is the calculated energy-saving planning;

[0025] An energy-saving coolant flow direction plan is obtained based on solving the minimum value of the energy-saving planning F(Q).

[0026] Preferably, the presetting of the heat distribution threshold includes:

[0027] An independent thermal threshold H τ_x of the single computer room device x is obtained based on the technical parameters of the single computer room device x;

[0028] The independent thermal thresholds of several computer room devices in the area are obtained, and the area thermal threshold is calculated using the area thermal threshold calculation formula. The area thermal threshold calculation formula is:

[0029]

[0030] Among them, x ∈ i means that the computer room device x belongs to area i. min(H τ_x ) is the minimum value of the independent thermal threshold of area i, the average value of the independent thermal threshold of area i, || is the absolute value calculation operator, and β i is the thermal interference compensation value of area i obtained by regression analysis based on the thermal interference historical data.

[0031] Preferably, the generation of the emergency coolant flow direction plan includes:

[0032] Generate an emergency coolant flow plan using an emergency planning formula, where the emergency planning formula is:

[0033]

[0034] where ω i is the preset thermal emergency coefficient for area i, which is obtained based on the importance of the computer room equipment in area i, H i_F is the abnormal regional thermal value of area i, H i_tar is the target thermal value obtained based on the regional thermal threshold, || is the absolute value operator, F(Q) is the calculated energy-saving plan and γ = 0 in the emergency planning formula, and G(Q') is the calculated emergency plan;

[0035] Obtain the emergency coolant flow plan based on solving the minimum value of the emergency plan G(Q').

[0036] Preferably, the calculation of the cooling remaining value is as follows:

[0037] Calculate the cooling remaining value using the cooling remaining value calculation formula, and the cooling remaining value calculation formula is:

[0038] C res = C 0 - Q abs

[0039] where C 0 is the initial cooling capacity of the coolant, which is obtained from the mass, specific heat capacity, and initial temperature of the coolant, Q abs is the total heat absorbed by the coolant when executing the emergency coolant flow plan, and this total heat is obtained from the flow rate of the coolant in the thermally abnormal area, the temperature difference between the inlet and outlet, the specific heat capacity, and the density of the coolant.

[0040] Preferably, the generation of the compensated coolant flow plan includes:

[0041] Generate a compensated coolant flow plan using a compensation planning formula, where the compensation planning formula is:

[0042] S(Q”) = η * C res * F(Q)

[0043] where η is the regression coefficient of the cooling remaining value obtained by regression analysis, C res is the cooling remaining value, F(Q) is the calculated energy-saving plan and γ = 1 in the compensation planning formula, and S(Q”) is the calculated compensation plan;

[0044] The compensated coolant flow planning is obtained based on solving the minimum value of the compensation planning S(Q").

[0045] In a second aspect, the present application discloses an energy-saving liquid cooling method for a computer room, which is applicable to the energy-saving liquid cooling system for a computer room as described above. The method includes:

[0046] Inputting the working real-time data into a thermal analysis model to generate thermal distribution real-time data; wherein, the working real-time data includes the real-time state data of the computer room equipment working, the thermal analysis model is constructed based on machine learning technology and combined with the working historical data, the corresponding heat generation historical data and the thermal interference historical data, and the thermal analysis model is used to generate the thermal distribution real-time data based on the working real-time data, the thermal distribution real-time data is used to characterize the real-time situation of the thermal distribution in the computer room, the real-time situation is the regional thermal value corresponding to each area in the computer room, the working historical data includes the historical state data of the computer room equipment working, the heat generation historical data is the historical data of the heat generation of the computer room equipment corresponding to the working historical data, and the thermal interference historical data is the thermal interference situation between the computer room equipment and between the computer room equipment and the computer room environment corresponding to the working historical data;

[0047] Generating an energy-saving coolant flow planning based on the thermal distribution real-time data; wherein, the energy-saving coolant flow planning is the optimal planning of the coolant flow direction after comprehensively considering the heat dissipation effect of the computer room equipment and the energy-saving effect of the liquid cooling equipment, and the energy-saving coolant flow planning is used to conventionally control the flow direction and the flow path of the coolant in the computer room;

[0048] Generating an emergency coolant flow planning when there is thermal distribution abnormal data in the thermal distribution real-time data; wherein, the thermal distribution abnormal data is the thermal distribution real-time data that does not meet the preset thermal distribution threshold, the thermal distribution threshold includes an independent thermal threshold for characterizing the threshold of the normal thermal distribution real-time data of a single computer room equipment and a regional thermal threshold for characterizing the threshold of the normal thermal distribution real-time data of an area composed of multiple computer room equipment, the emergency coolant flow planning is the optimal planning of the coolant flow direction corresponding to the best heat dissipation effect of the computer room equipment, and the emergency coolant flow planning is used to emergently control the flow direction and the flow path of the coolant in the computer room;

[0049] Calculate the remaining cooling value of the coolant after implementing the emergency coolant flow plan, and generate a compensatory coolant flow plan based on the remaining cooling value; wherein, the remaining cooling value is used to characterize the degree of the cooling function that the coolant still has after implementing the emergency coolant flow plan, and the compensatory coolant flow plan is the optimal plan for the coolant flow after comprehensively considering the heat dissipation effect of the computer room equipment, the energy-saving effect of the liquid cooling equipment, and the remaining cooling value, and the compensatory coolant flow plan is used to compensate and control the flow direction and flow path of the coolant in the computer room.

[0050] In a third aspect, the present application discloses a communication computer room, which includes electronic devices. The electronic devices include a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it implements the energy-saving liquid cooling method for the computer room as described above.

[0051] Beneficial effects: For the energy-saving liquid cooling system, method, and communication computer room of the present application, the thermal analysis module constructs a model based on machine learning technology and combines various historical data to accurately generate real-time thermal distribution data, providing a reliable basis for subsequent coolant flow planning; the conventional circulation module generates an energy-saving coolant flow plan based on the real-time thermal distribution data, taking into account both the heat dissipation of computer room equipment and the energy saving of liquid cooling equipment; the emergency response module quickly generates an emergency coolant flow plan when the thermal distribution is abnormal to ensure equipment safety; the emergency circulation module calculates the remaining cooling value and generates a compensatory coolant flow plan to further improve the coolant utilization rate and system energy efficiency, thus solving the deficiencies of existing liquid cooling technologies in aspects such as energy consumption control and coping with environmental changes, improving the overall performance of the liquid cooling system, and meeting the requirements of modern data centers for energy conservation and efficient heat dissipation. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0053] Figure 1 It is a structural block diagram of the energy-saving liquid cooling system for the computer room provided by the embodiment of the present application;

[0054] Figure 2 It is a flow block diagram of the energy-saving liquid cooling method for the computer room provided by the embodiment of the present application. Detailed Embodiments

[0055] The technical solutions in the embodiments of the present application will be described clearly and completely below. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0056] In this document, the term "including" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, the elements defined by the statement "including..." do not exclude the existence of additional identical elements in the process, method, article or device including the said elements.

[0057] The first aspect of this embodiment discloses an energy-saving liquid cooling system for a computer room as shown in Figure 1 Figure [not shown]. The system includes a thermal analysis module, a conventional circulation module, an emergency response module, and an emergency circulation module; wherein, the thermal analysis module is communicatively connected to the conventional circulation module, the thermal analysis module is also communicatively connected to the emergency response module, and the emergency response module is communicatively connected to the emergency circulation module;

[0058] The thermal analysis module is configured to input real-time working data into a thermal analysis model to generate real-time thermal distribution data; wherein, the real-time working data includes real-time status data of the computer room equipment, the thermal analysis model is constructed based on machine learning technology and combined with historical working data and corresponding heat generation historical data and thermal interference historical data, and the thermal analysis model is used to generate real-time thermal distribution data based on the real-time working data, the real-time thermal distribution data is used to characterize the real-time situation of the thermal distribution in the computer room, this real-time situation is the regional thermal value corresponding to each area in the computer room, the historical working data includes historical status data of the computer room equipment, the heat generation historical data is the historical heat generation data of the computer room equipment corresponding to the historical working data, and the thermal interference historical data is the thermal interference situation between the computer room equipment and between the computer room equipment and the computer room environment corresponding to the historical working data;

[0059] The conventional circulation module is configured to generate an energy-saving coolant flow plan based on the real-time thermal distribution data; wherein, the energy-saving coolant flow plan is the optimal plan of the coolant flow direction after comprehensively considering the heat dissipation effect of the computer room equipment and the energy-saving effect of the liquid cooling equipment, and the energy-saving coolant flow plan is used to conventionally control the flow direction and the flow path of the coolant in the computer room;

[0060] It should be noted that in the translation of , since the figure number is not shown in the original text, it is marked as "Figure [not shown]" in the translation. You can adjust it according to the actual figure number if available.The emergency response module is configured to generate an emergency coolant flow plan when there is abnormal heat distribution data in the real-time heat distribution data; wherein, the abnormal heat distribution data is real-time heat distribution data that does not meet the preset heat distribution threshold, and the heat distribution threshold includes an independent thermal threshold for characterizing the threshold of the normal real-time heat distribution data of a single computer room device and a regional thermal threshold for characterizing the threshold of the normal real-time heat distribution data of an area composed of multiple computer room devices. The emergency coolant flow plan is the optimal plan for the coolant flow corresponding to the best heat dissipation effect of the computer room device, and the emergency coolant flow plan is used for emergency control of the flow direction and flow path of the coolant in the computer room;

[0061] The emergency circulation module is configured to calculate the remaining cooling value of the coolant after executing the emergency coolant flow plan, and generate a compensation coolant flow plan based on the remaining cooling value; wherein, the remaining cooling value of the coolant is used to characterize the degree of the cooling function that the coolant still has after executing the emergency coolant flow plan, and the compensation coolant flow plan is the optimal plan for the coolant flow after comprehensively considering the heat dissipation effect of the computer room device, the energy-saving effect of the liquid cooling device, and the remaining cooling value, and the compensation coolant flow plan is used for compensating and controlling the flow direction and flow path of the coolant in the computer room.

[0062] Through the above, this embodiment realizes the efficient energy-saving control of the computer room liquid cooling system by the collaborative work of the heat analysis module, the conventional circulation module, the emergency response module, and the emergency circulation module. The heat analysis module constructs a model based on the existing machine learning technology and combines various historical data to accurately generate real-time heat distribution data, providing a reliable basis for the subsequent coolant flow plan. The conventional circulation module generates an energy-saving coolant flow plan based on the real-time heat distribution data, taking into account both the heat dissipation of the computer room device and the energy saving of the liquid cooling device. The emergency response module quickly generates an emergency coolant flow plan when the heat distribution is abnormal to ensure the safety of the equipment. The emergency circulation module calculates the remaining cooling value and generates a compensation coolant flow plan to further improve the utilization rate of the coolant and the energy saving of the system. Based on the close cooperation of each module, the deficiencies of the existing liquid cooling technology in energy consumption control, coping with environmental changes, etc. are solved, the overall performance of the liquid cooling system is improved, and the requirements of modern data centers for energy saving and efficient heat dissipation are met.

[0063] Specifically, the construction of the heat analysis model includes:

[0064] Collect the working historical data of the computer room device and the corresponding heat generation historical data and heat interference historical data;

[0065] Using machine learning techniques, taking the work history data and the corresponding heat generation history data and heat interference history data as input features, and the corresponding heat distribution history data as output labels, training the model, adjusting the parameters of the trained model, so that the error between the prediction result of the model and the actual heat distribution history data meets the preset prediction error threshold, and obtaining a heat analysis model.

[0066] Through the above, this embodiment uses the collected work history data, heat generation history data and heat interference history data of the computer room equipment, and combines the existing machine learning techniques (such as deep learning models) to achieve the accurate construction of the heat analysis model. Using these historical data as input features and output labels to train the model, and adjusting the parameters to make the prediction error meet the preset threshold, ensuring that the model can accurately reflect the heat distribution in the computer room. This model provides core support for the generation of real-time heat distribution data, and further lays a foundation for the coolant flow planning, avoiding the problems of over-reliance on specific models and being greatly affected by data quality, improving the accuracy and stability of the heat distribution analysis in the computer room, helping the subsequent modules to more reasonably plan the coolant flow, and enhancing the overall performance of the system.

[0067] Specifically, the generation of real-time heat distribution data includes:

[0068] Dividing the computer room into several regions, calculating the regional heat values of each region respectively, and defining the sorted results of the regional heat values based on the spatial sequence of the regions as real-time heat distribution data; among them, the calculation of the regional heat value is:

[0069] Obtaining the real-time work data of the computer room equipment in the region, inputting the real-time work data into the heat analysis model and matching the corresponding heat weight parameters and heat correction values for the real-time work data; among them, the heat weight parameters and heat correction values are model parameters obtained by regression analysis based on the work history data, the corresponding heat generation history data and heat interference history data of the computer room equipment during the training of the heat analysis model

[0070] Obtaining the regional heat value based on the real-time work data of the region and the corresponding heat weight parameters and heat correction values.

[0071] It can be understood that in the actually divided regions, there will be multiple computer room equipment. Therefore, when actually calculating the real-time work data, the heat weight parameters should be matched corresponding to the computer room equipment. The heat weight parameters can be but are not limited to the importance of the equipment during the training of the heat analysis model. The importance directly affects the degree of consideration of its heat. Further, the matched heat weight parameters and the corresponding real-time work data need to be summed up to fully represent the heat situation of the region.

[0072] Through the above, in this embodiment, the computer room is divided into several areas, and the precise calculation of real-time data of heat distribution is realized by using the heat analysis model and related model parameters. After obtaining the real-time data of the operation of the computer room equipment in the area, the corresponding thermal weight parameters and thermal correction values are matched, and the thermal values of each area are calculated and sorted to obtain the real-time data of heat distribution, so as to more carefully reflect the thermal conditions of different areas in the computer room. Compared with the prior art, the characterization of the heat distribution in the computer room is more accurate, providing more precise data support for the subsequent generation of an energy-saving coolant flow plan for the conventional circulation module, making the coolant flow plan more in line with the actual heat dissipation requirements, and improving the heat dissipation efficiency and energy-saving effect.

[0073] Specifically, the generation of the energy-saving coolant flow plan includes:

[0074] Using the energy-saving planning formula to generate the energy-saving coolant flow plan, where the energy-saving planning formula is:

[0075]

[0076] Among them, Q i represents the flow rate Q of the coolant flowing to area i i , n represents the total number of areas and i ∈ n, E(Q i ) represents the energy consumption value of the liquid cooling equipment flowing to area i at the flow rate Q i obtained based on the liquid cooling equipment energy consumption function. The liquid cooling equipment energy consumption function is the energy consumption value at different flow rates obtained based on the energy consumption situation of the liquid cooling equipment. R i represents the heat dissipation demand coefficient of area i and R i = k 1 *H i + k 2 , k 1 and k 2 are preset heat dissipation adjustment common senses, H i is the area thermal value of area i, C(Q i ) represents the cooling capacity that the coolant flow rate Q i flowing to area i can provide. This cooling capacity is obtained by fitting based on the physical properties and flow rate of the coolant. max() is the maximum value calculation operator, λ is a preset heat dissipation weight coefficient, γ is a preset energy consumption discrimination coefficient and γ = 1 in the energy-saving planning formula, and F(Q) is the calculated energy-saving plan;

[0077] Based on solving the minimum value of the energy-saving plan F(Q), the energy-saving coolant flow plan is obtained.

[0078] Through the above, this embodiment realizes the scientific generation of the energy-saving coolant flow direction planning by using the energy-saving planning formula. Based on comprehensively considering factors such as coolant flow rate, energy consumption of liquid cooling equipment, regional heat dissipation requirements, and cooling capacity, the flow direction planning is determined by solving the minimum value of the formula. Different from the existing optimization algorithms that are complex and consume a large amount of computing resources, this formula is simple and efficient, and can quickly obtain the optimal coolant flow direction planning that takes into account both the heat dissipation effect of the computer room equipment and the energy-saving effect of the liquid cooling equipment, accurately control the flow direction and flow path of the coolant in the computer room, reduce system energy consumption, and improve the energy-saving performance and overall operation efficiency of the liquid cooling system.

[0079] Specifically, the preset of the heat distribution threshold includes:

[0080] Obtain the independent thermal threshold H of a single computer room equipment x based on the technical parameters of the single computer room equipment x τ_x ;

[0081] Obtain the independent thermal thresholds of several computer room equipment in the area, and use the regional thermal threshold calculation formula to calculate the regional thermal threshold. The regional thermal threshold calculation formula is:

[0082]

[0083] Among them, x∈i means that the computer room equipment x belongs to the area i, and min(H τ_x ) is the minimum value of the independent thermal thresholds of the area i, The average value of the independent thermal thresholds of the area i, || is the absolute value operator, and β i is the thermal interference compensation value of the area i obtained by regression analysis based on the historical thermal interference data.

[0084] Through the above, this embodiment realizes the reasonable preset of the heat distribution threshold by calculating based on the technical parameters of a single computer room equipment and the independent thermal thresholds of the equipment in the area. Determine the independent thermal threshold according to the technical parameters of a single equipment, and then use the regional thermal threshold calculation formula, combined with the thermal interference compensation value, to obtain the regional thermal threshold. This method fully considers the individual differences of the equipment and the regional thermal interference situation, thus paying attention to the thermal interference between the equipment and between the equipment and the environment, and can more accurately judge whether the heat distribution is abnormal, providing an accurate basis for the emergency response module to detect heat anomalies in time and generate a reasonable emergency coolant flow direction planning, and ensuring the stable operation of the computer room equipment.

[0085] Specifically, the generation of the emergency coolant flow direction planning includes:

[0086] Use the emergency planning formula to generate the emergency coolant flow direction planning. Among them, the emergency planning formula is:

[0087]

[0088] Among them, ωi is the preset thermal emergency coefficient for region i, which is obtained based on the importance of the computer room equipment in region i, H i_F is the abnormal regional thermal value of region i, H i_tar is the target thermal value obtained based on the regional thermal threshold, || is the absolute value operator, F(Q) is the calculated energy-saving plan and γ = 0 in the emergency plan formula, and G(Q') is the calculated emergency plan;

[0089] Based on solving the minimum value of the emergency plan G(Q'), the emergency-type coolant flow direction plan is obtained.

[0090] It should be noted that the goal of the emergency-type coolant flow direction plan is to eliminate the thermal anomaly as soon as possible and restore the thermal value of the thermal anomaly area to the normal range. Therefore, in this embodiment, γ in the calculated energy-saving plan F(Q) is set to 0, so as to release the cooling capacity of the liquid cooling equipment.

[0091] Through the above, this embodiment uses the emergency plan formula to effectively generate the emergency-type coolant flow direction plan. Based on factors such as the importance of the computer room equipment in the considered area, the abnormal thermal value, and the target thermal value, the emergency plan is obtained by solving the minimum value of the formula. This enables the system to quickly generate the coolant flow direction plan with the best heat dissipation effect according to the actual situation of each area when the heat distribution is abnormal, timely respond to the thermal anomaly situation, avoid equipment damage due to overheating, thus avoiding the problem of slow response when dealing with the rapid change of the computer room operation state, and realizing a faster and more accurate adjustment of the coolant flow direction to ensure the safe and stable operation of the computer room equipment.

[0092] Specifically, the calculation of the cooling remaining value is as follows:

[0093] The cooling remaining value is calculated using the cooling remaining value calculation formula, and the cooling remaining value calculation formula is:

[0094] C res = C 0 - Q abs

[0095] where C 0 is the initial cooling capacity of the coolant, which is obtained from the mass, specific heat capacity, and initial temperature of the coolant, and Q abs is the total heat absorbed by the coolant when implementing the emergency-type coolant flow direction plan, and this total heat is obtained from the flow rate, inlet and outlet temperature difference, specific heat capacity, and density of the coolant in the thermal anomaly area.

[0096] Through the above, this embodiment realizes the accurate calculation of the cooling remaining value of the coolant by using the cooling remaining value calculation formula. By considering the initial cooling capacity of the coolant and the total heat absorbed by implementing the emergency coolant flow plan, the cooling remaining value is obtained. This calculation method quantifies the cooling capacity of the coolant after the emergency, provides key data support for generating the compensatory coolant flow plan subsequently, enables the system to make full use of the remaining cooling function of the coolant, improves the utilization rate of the coolant, further optimizes the energy-saving effect of the liquid cooling system, and solves the problem of ignoring the influence of coolant characteristics on heat dissipation and energy saving.

[0097] Specifically, the generation of the compensatory coolant flow plan includes:

[0098] Using the compensation plan formula to generate the compensatory coolant flow plan, where the compensation plan formula is:

[0099] S(Q”) = η * C res * F(Q)

[0100] where η is the regression coefficient of the cooling remaining value obtained by regression analysis, C res is the cooling remaining value, F(Q) is the energy-saving plan calculated and γ = 1 in the compensation plan formula, and S(Q”) is the calculated compensation plan;

[0101] Based on solving the minimum value of the compensation plan S(Q”), the compensatory coolant flow plan is obtained.

[0102] Through the above, this embodiment realizes the reasonable generation of the compensatory coolant flow plan by using the compensation plan formula. Based on factors such as the regression coefficient of the cooling remaining value and the cooling remaining value considered, the compensation plan is obtained by solving the minimum value of the formula, thereby integrating the heat dissipation effect of the computer room equipment, the energy-saving effect of the liquid cooling equipment, and the cooling remaining value, making full use of the remaining cooling capacity of the coolant, optimizing and compensating the control of the coolant flow direction, further improving the energy-saving performance and heat dissipation effect of the liquid cooling system, avoiding waste of coolant resources, and perfecting the energy-saving control system of the entire liquid cooling system.

[0103] The second aspect of this embodiment discloses an energy-saving liquid cooling method for a computer room as Figure 2 shown, which is applicable to the energy-saving liquid cooling system of the computer room as described above. The method includes:

[0104] Input real-time working data into a thermal analysis model to generate real-time thermal distribution data; wherein, the real-time working data includes real-time status data of the operation of computer room equipment, the thermal analysis model is constructed based on machine learning technology and combined with historical working data, corresponding historical heat generation data, and historical thermal interference data, and the thermal analysis model is used to generate real-time thermal distribution data based on the real-time working data. The real-time thermal distribution data is used to characterize the real-time situation of the thermal distribution in the computer room, and this real-time situation is the regional thermal value corresponding to each area in the computer room. The historical working data includes historical status data of the operation of computer room equipment, the historical heat generation data is the historical heat generation data of the computer room equipment corresponding to the historical working data, and the historical thermal interference data is the thermal interference situation between the computer room equipment and between the computer room equipment and the computer room environment corresponding to the historical working data;

[0105] Generate an energy-saving coolant flow direction plan based on the real-time thermal distribution data; wherein, the energy-saving coolant flow direction plan is the optimal plan for the coolant flow direction after comprehensively considering the heat dissipation effect of computer room equipment and the energy-saving effect of liquid cooling equipment, and the energy-saving coolant flow direction plan is used to regularly control the flow direction and flow path of the coolant in the computer room;

[0106] Generate an emergency coolant flow direction plan when there is abnormal thermal distribution data in the real-time thermal distribution data; wherein, the abnormal thermal distribution data is the real-time thermal distribution data that does not meet the preset thermal distribution threshold. The thermal distribution threshold includes an independent thermal threshold for characterizing the threshold of the normal real-time thermal distribution data of a single computer room equipment and a regional thermal threshold for characterizing the threshold of the normal real-time thermal distribution data of an area composed of multiple computer room equipment. The emergency coolant flow direction plan is the optimal plan for the coolant flow direction corresponding to the best heat dissipation effect of computer room equipment, and the emergency coolant flow direction plan is used to emergently control the flow direction and flow path of the coolant in the computer room;

[0107] Calculate the remaining cooling value of the coolant after implementing the emergency coolant flow direction plan, and generate a compensatory coolant flow direction plan based on the remaining cooling value; wherein, the remaining cooling value is used to characterize the degree of the remaining cooling function of the coolant after implementing the emergency coolant flow direction plan. The compensatory coolant flow direction plan is the optimal plan for the coolant flow direction after comprehensively considering the heat dissipation effect of computer room equipment, the energy-saving effect of liquid cooling equipment, and the remaining cooling value, and the compensatory coolant flow direction plan is used to compensatorily control the flow direction and flow path of the coolant in the computer room.

[0108] It should be noted that the energy-saving liquid cooling method for the computer room in this embodiment corresponds to the aforementioned energy-saving liquid cooling system for the computer room. Therefore, for the content not specifically described in the energy-saving liquid cooling system for the computer room in this embodiment, such as but not limited to function definitions, working principles, and technical effects, etc., reference can be made to the records of the aforementioned energy-saving liquid cooling system for the computer room, and this text will not elaborate here.

[0109] In the third aspect of this embodiment, a communication machine room is disclosed. The communication machine room includes electronic devices, and the electronic devices include a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it implements the energy-saving liquid cooling method for the machine room as described above.

[0110] Similarly, it should be noted that the communication machine room in this embodiment corresponds to the energy-saving liquid cooling system of the aforementioned machine room. Therefore, the content not specifically described in the communication machine room of this embodiment, such as, but not limited to, function definition, working principle, and technical effects, etc., can refer to the records of the energy-saving liquid cooling system of the aforementioned machine room, and will not be elaborated in this text.

[0111] In summary, for the energy-saving liquid cooling system, method, and communication machine room of this embodiment, the thermal analysis module constructs a model based on machine learning technology and combines various historical data to accurately generate real-time thermal distribution data, providing a reliable basis for the subsequent coolant flow planning; the conventional circulation module generates an energy-saving coolant flow plan based on the real-time thermal distribution data, taking into account both the heat dissipation of the machine room equipment and the energy saving of the liquid cooling equipment; the emergency response module quickly generates an emergency coolant flow plan when the thermal distribution is abnormal to ensure the safety of the equipment; the emergency circulation module calculates the remaining cooling value and generates a compensatory coolant flow plan to further improve the coolant utilization rate and system energy saving. Thus, the deficiencies of the existing liquid cooling technology in aspects such as energy consumption control and response to environmental changes are solved, the overall performance of the liquid cooling system is improved, and the requirements of modern data centers for energy saving and efficient heat dissipation are met.

[0112] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any suitable combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, or other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. The computer-readable storage medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transmission of a computer program from one place to another. The storage media can be any available medium that can be accessed by a computer. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM, or other optical disk storage, magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0113] Finally, it should be noted that the above are only the preferred embodiments of the present application and are not used to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An energy-saving liquid cooling system for a computer room, characterized in that: The system includes a thermal analysis module, a conventional circulation module, an emergency response module and an emergency circulation module; wherein the thermal analysis module is communicatively connected to the conventional circulation module, the thermal analysis module is also communicatively connected to the emergency response module, and the emergency response module is communicatively connected to the emergency circulation module; The thermal analysis module is configured to input the real-time working data into the thermal analysis model to generate real-time data of thermal distribution; wherein the real-time working data includes real-time working status data of the equipment in the computer room, the thermal analysis model is constructed based on machine learning technology and in combination with the working history data and the corresponding heat generation history data and thermal interference history data, and the thermal analysis model is used to generate the real-time data of thermal distribution based on the real-time working data, the real-time data of thermal distribution is used to characterize the real-time situation of thermal distribution in the computer room, and the real-time situation is the regional thermal value corresponding to each area in the computer room, the working history data includes the historical working status data of the equipment in the computer room, the heat generation history data is the historical data of heat generation of the equipment in the computer room corresponding to the working history data, and the thermal interference history data is the thermal interference situation between the equipment in the computer room corresponding to the working history data and the thermal interference situation between the equipment in the computer room and the computer room environment; The conventional circulation module is configured to generate an energy-saving coolant flow direction plan based on the real-time data of the heat distribution; wherein the energy-saving coolant flow direction plan is an optimal plan for the coolant flow direction after comprehensively considering the heat dissipation effect of the equipment in the computer room and the energy-saving effect of the liquid cooling equipment, and the energy-saving coolant flow direction plan is used for conventionally controlling the flow direction and flow path of the coolant in the computer room; The emergency response module is configured to generate an emergency coolant flow direction plan when abnormal heat distribution data exists in the real-time heat distribution data; wherein the abnormal heat distribution data is real-time heat distribution data that does not meet a preset heat distribution threshold, and the heat distribution threshold includes an independent thermal threshold for characterizing the threshold of normal heat distribution real-time data of a single computer room device and a regional thermal threshold for the threshold of normal heat distribution real-time data of a region composed of multiple computer room devices, and the emergency coolant flow direction plan is an optimal plan for the coolant flow direction corresponding to the best heat dissipation effect of the computer room equipment, and the emergency coolant flow direction plan is used to emergency control the flow direction and flow path of the coolant in the computer room; The emergency circulation module is configured as follows: calculating the cooling surplus value of the coolant after executing the emergency coolant flow direction planning, and generating a compensatory coolant flow direction planning based on the cooling surplus value; wherein the cooling surplus value is used to characterize the degree of cooling function of the coolant after executing the emergency coolant flow direction planning, and the compensatory coolant flow direction planning is an optimal planning of the coolant flow direction after comprehensively considering the heat dissipation effect of the computer room equipment, the energy-saving effect of the liquid cooling equipment and the cooling surplus value, and the compensatory coolant flow direction planning is used to compensate and control the flow direction and flow path of the coolant in the computer room.

2. The energy-saving liquid cooling system for a computer room according to claim 1, characterized in that: The construction of the thermal analysis model includes: Collecting the working history data of the equipment in the computer room and the corresponding heat generation history data and thermal interference history data; Using machine learning technology, the working history data and the corresponding heat generation history data and the thermal interference history data are used as input features, and the corresponding heat distribution history data are used as output labels. The model is trained and the parameters of the training model are adjusted so that the error between the prediction result of the model and the actual heat distribution history data meets the preset prediction error threshold, thereby obtaining a thermal analysis model.

3. The energy-saving liquid cooling system for a computer room according to claim 1, characterized in that: The generation of the real-time data of heat distribution includes: The computer room is divided into several areas, and the regional thermal value of each area is calculated respectively. The sorting result of the regional thermal value based on the spatial sequence of the area is defined as the real-time data of heat distribution; wherein the calculation of the regional thermal value is: Acquire the real-time working data of the equipment in the computer room in the area, input the real-time working data into the thermal analysis model, and match the corresponding thermal weight parameters and thermal correction values ​​for the real-time working data; wherein the thermal weight parameters and the thermal correction values ​​are model parameters obtained by regression analysis based on the working history data of the equipment in the computer room and the corresponding heat generation history data and thermal interference history data in the training of the thermal analysis model. The regional thermal value is obtained based on the real-time working data of the region and the corresponding thermal weight parameter and the thermal correction value.

4. The energy-saving liquid cooling system for a computer room according to claim 1, characterized in that: The generation of the energy-saving coolant flow planning includes: The energy-saving planning formula is used to generate the energy-saving coolant flow planning, wherein the energy-saving planning formula is: Among them, Q i represents the flow rate Q of the coolant flowing to area i i , n represents the total number of regions and i∈n, E(Q i ) represents the flow rate Q obtained based on the energy consumption function of the liquid cooling equipment i The energy consumption value of the liquid cooling device flowing to area i, the energy consumption function of the liquid cooling device is the energy consumption value under different flow rates obtained based on the energy consumption of the liquid cooling device, R i represents the heat dissipation demand coefficient of area i and R i =k1*H i +k2, k1 and k2 are the default cooling adjustment common sense, H i is the regional thermal value of region i, C(Q i ) represents the coolant flow Q flowing to area i i The cooling capacity that can be provided is obtained by fitting based on the physical properties and flow rate of the coolant. max() is the maximum value calculation symbol, λ is the preset heat dissipation weight coefficient, γ is the preset energy consumption discrimination coefficient and γ=1 in the energy-saving planning formula, and F(Q) is the calculated energy-saving plan; Based on solving the minimum value of the energy-saving planning F(Q), the energy-saving coolant flow planning is obtained.

5. The energy-saving liquid cooling system for a computer room according to claim 1, characterized in that: The preset heat distribution threshold includes: Based on the technical parameters of a single equipment room device x, the independent thermal threshold H of a single equipment room device x is obtained. τ_x ; The independent thermal thresholds of several equipment rooms in the area are obtained, and the regional thermal threshold is calculated using the regional thermal threshold calculation formula. The regional thermal threshold calculation formula is: Among them, x∈i means that the equipment x in the computer room belongs to area i, min(H τ_x ) is the minimum value of the independent thermal threshold of region i, The average value of the independent thermal threshold of region i, || is the absolute value operator, β i is the thermal interference compensation value of region i obtained based on the regression analysis of the thermal interference historical data.

6. The energy-saving liquid cooling system for a computer room according to claim 4, characterized in that: The generation of the emergency coolant flow planning includes: The emergency planning formula is used to generate the emergency coolant flow planning, wherein the emergency planning formula is: Among them, ω i is the preset thermal emergency coefficient of area i, which is obtained based on the importance of the equipment in the computer room in area i, H i_F is the abnormal regional thermal value of region i, H i_tar is the target thermal value obtained based on the regional thermal threshold, || is the absolute value operator, F(Q) is the calculated energy-saving plan and γ=0 in the emergency planning formula, G(Q') is the calculated emergency plan; The emergency coolant flow planning is obtained by solving the minimum value of the emergency planning G(Q').

7. The energy-saving liquid cooling system for a computer room according to claim 4, characterized in that: The calculation of the cooling residual value is: The cooling surplus value is calculated using the cooling surplus value calculation formula, and the cooling surplus value calculation formula is: C res =C0-Q abs Where C0 is the initial cooling capacity of the coolant, which is obtained from the mass, specific heat capacity and initial temperature of the coolant, Q abs It is the total heat absorbed by the coolant when executing the emergency coolant flow planning, and the total heat is obtained by the flow rate of the coolant in the thermal abnormality area, the inlet and outlet temperature difference, the specific heat capacity and density of the coolant.

8. The energy-saving liquid cooling system for a computer room according to claim 7, characterized in that: The generation of the compensatory coolant flow planning includes: The compensation planning formula is used to generate the compensation type coolant flow planning, wherein the compensation planning formula is: S(Q”)=η*C res *F(Q) Among them, η is the regression coefficient of cooling surplus value obtained by regression analysis, C res is the cooling surplus value, F(Q) is the calculated energy saving plan and γ=1 in the compensation plan formula, S(Q”) is the calculated compensation plan; The compensated coolant flow planning is obtained based on solving the minimum value of the compensation planning S(Q”).

9. An energy-saving liquid cooling method for a computer room, the method being applicable to the energy-saving liquid cooling system for a computer room as claimed in any one of claims 1 to 8, characterized in that: The method includes: Input the real-time working data into the thermal analysis model to generate real-time data of thermal distribution; wherein, the real-time working data includes real-time working status data of the equipment in the computer room, the thermal analysis model is constructed based on machine learning technology and in combination with working history data and corresponding heat generation history data and thermal interference history data, and the thermal analysis model is used to generate the real-time data of thermal distribution based on the real-time working data, the real-time data of thermal distribution is used to characterize the real-time situation of thermal distribution in the computer room, and the real-time situation is the regional thermal value corresponding to each area in the computer room, the working history data includes historical working status data of the equipment in the computer room, the heat generation history data is the heat generation history data of the equipment in the computer room corresponding to the working history data, and the thermal interference history data is the thermal interference situation between the equipment in the computer room corresponding to the working history data and the thermal interference situation between the equipment in the computer room and the computer room environment; Generate an energy-saving coolant flow direction plan based on the real-time data of heat distribution; wherein the energy-saving coolant flow direction plan is an optimal plan for the coolant flow direction after comprehensively considering the heat dissipation effect of the equipment in the computer room and the energy-saving effect of the liquid cooling equipment, and the energy-saving coolant flow direction plan is used for conventionally controlling the flow direction and flow path of the coolant in the computer room; When abnormal heat distribution data exists in the real-time heat distribution data, an emergency coolant flow direction planning is generated; wherein, the abnormal heat distribution data is real-time heat distribution data that does not meet a preset heat distribution threshold, and the heat distribution threshold includes an independent thermal threshold used to characterize the threshold of normal heat distribution real-time data of a single computer room device and a regional thermal threshold of the threshold of normal heat distribution real-time data of a region composed of multiple computer room devices, and the emergency coolant flow direction planning is the optimal planning of the coolant flow direction corresponding to the best heat dissipation effect of the computer room equipment, and the emergency coolant flow direction planning is used to emergency control the flow direction and flow path of the coolant in the computer room; The cooling surplus value of the coolant after executing the emergency coolant flow direction planning is calculated, and a compensatory coolant flow direction planning is generated based on the cooling surplus value; wherein the cooling surplus value is used to characterize the degree of cooling function of the coolant after executing the emergency coolant flow direction planning, and the compensatory coolant flow direction planning is an optimal planning of the coolant flow direction after comprehensively considering the heat dissipation effect of the equipment in the computer room, the energy-saving effect of the liquid cooling equipment and the cooling surplus value, and the compensatory coolant flow direction planning is used to compensate and control the flow direction and flow path of the coolant in the computer room.

10. A communication room, characterized in that: The communication room includes electronic equipment, which includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the energy-saving liquid cooling method for the room as claimed in claim 9 is implemented.

Images