Data center fan regulation method

Abstract

A method for controlling the speed of fans of a dry cooling unit of a data center, the method comprising: estimating (202) a heat load (Q) of a rack-mounted data processing assembly (120A-120N); estimating (206) an air flow (AF) as a function of the heat load (Q) and an ambient temperature (T) of the air flow; estimating (208) a fan rotational speed (N) of at least one fan assembly (110A) as a function of the estimated air flow (AF); controlling (210) a voltage (V) of the at least one fan assembly (110A) as a function of the estimated fan rotational speed (N), wherein the process (200) comprises a step (205) of modeling the air flow (AF) as a polynomial function of the ambient temperature.

Description

Cross-reference to related applications

[0001] This patent application claims priority to European Patent Application No. 24307256.8, filed on 20 December 2024, the entire disclosure of which is incorporated herein by reference. Technical Field

[0002] This technology generally relates to the field of liquid cooling systems for data centers. Background Technology

[0003] Data centers and many computer processing facilities house a large number of rack-mounted electronic processing devices. During operation, these devices generate a significant amount of heat, which must be dissipated to prevent electronic component failure and ensure continuous, efficient processing.

[0004] To this end, various liquid cooling measures have been adopted to facilitate the dissipation of heat generated by electronic processing devices. One such measure employs liquid block cooling technology to directly cool one or more heat-generating processing components. This technology utilizes a liquid cooling block having internal channels that receive cooling liquid from a cooling liquid source, and these internal channels are in thermal contact with the heat-generating processing components.

[0005] Depending on the availability of water sources, in many cases, the preferred source of cooling liquid includes a dry cooling unit. The dry cooling unit supplies cooling liquid to the rack-mounted electronic processing equipment via a pump, and receives heated liquid from the electronic processing equipment. The dry cooling unit is configured to re-cool the received heated liquid for recirculation back to the electronic processing equipment.

[0006] Energy consumption in data centers has become a significant issue due to a variety of factors. Data centers require substantial amounts of electricity to power servers and storage devices. This demand is growing rapidly, driven by the increasing prevalence of technologies such as cloud computing, big data analytics, and artificial intelligence. Furthermore, cooling systems themselves consume significant amounts of energy. The high energy consumption of data centers exacerbates greenhouse gas emissions, thus having a significant impact on climate change.

[0007] Improving energy efficiency is crucial to mitigating these impacts.

[0008] Therefore, there remains an interest in trying to minimize the energy consumption of data center cooling systems.

[0009] The topics discussed in the background section should not be considered prior art simply because they are mentioned therein. Similarly, the problems mentioned in the background section, or those related to the topics in the background section, should not be considered as previously known in the prior art. The topics in the background section merely represent different approaches. Summary of the Invention

[0010] The specific implementation of this technology was developed based on some drawbacks associated with traditional dry cooling technology, and the implementation method is designed to improve energy efficiency.

[0011] This technology relates to a liquid cooling method for a data center liquid cooling system, wherein the liquid data center liquid cooling system is used to cool rack-mounted data processing components, and the data center liquid cooling system includes a cooling device, the cooling device being provided with: A cooling unit configured to supply cooling fluid to the rack-mounted data processing assembly and to receive heated fluid from the rack-mounted data processing assembly. A forward liquid distribution circuit, comprising a temperature sensor and / or at least one first pump, and / or a pump pressure sensor, and / or a liquid flow / volume sensor, and / or a valve, the forward liquid distribution circuit being used to deliver cooling liquid from the cooling unit to the rack-mounted data processing assembly. A return liquid distribution circuit, comprising a temperature sensor and / or at least one first pump, and / or a pump pressure sensor, and / or a liquid flow / volume sensor, and / or a valve, the return liquid distribution circuit being used to deliver heated liquid from the rack-mounted data processing assembly back to the cooling unit and the temperature sensor. The control module panel is connected to the temperature sensor, pump pressure sensor, liquid flow rate / volume sensor, and temperature sensor via communication to receive data from them. Each rack-mounted data processing assembly includes at least one heat-generating electronic processing element and at least one liquid cooling block, the at least one liquid cooling block being arranged to make corresponding thermal contact with the at least one heat-generating electronic processing element and to be fluidly connected to the forward liquid distribution loop and / or intelligent flow valve, the intelligent flow valve being correspondingly arranged to be fluidly connected to the at least one liquid cooling block of the respective rack-mounted data processing assembly. The cooling unit includes at least one fan assembly and at least one heat exchanger assembly, and is used to cool the liquid circulating in the system and entering the cooling unit through the return liquid distribution loop via airflow. The method includes the following steps: - Estimate the thermal load of the rack-mounted data processing unit. - Estimate the airflow rate based on the heat load and ambient temperature of the airflow. - Based on the estimated airflow, the fan rotation speed of the at least one fan assembly is estimated, and - Control the voltage of the at least one fan assembly based on the estimated fan rotation speed. The voltage depends on the minimum of two function values: a first function and a second function. The first function depends on the estimated fan speed, and the second function depends on the liquid inlet temperature.

[0012] The fan control method of this technology ensures reduced energy consumption and better water cooling.

[0013] In some implementations, the voltage is proportional to the minimum of the two function values.

[0014] In some implementations, the first function is proportional to the estimated fan rotation speed.

[0015] In some implementations, the second function is a linear function of the liquid inlet temperature.

[0016] In some implementations, the fan is activated when the air temperature is below a set point, which depends on the ambient temperature and / or the heat load. The difference between the set point and the activation temperature is called a margin.

[0017] In some implementations, the setpoint is: - If the ambient temperature is lower than or equal to the threshold temperature, and the sampled heat load is lower than or equal to the threshold heat load, then the setpoint is a low setpoint; - If the ambient temperature is lower than or equal to the threshold temperature, and the sampled heat load is higher than the threshold heat load, then the setpoint is a high setpoint, which is higher than the low setpoint; - If the ambient temperature is higher than the threshold temperature and the sampled heat load is lower than or equal to the threshold heat load, the setpoint depends on the ambient temperature and the first temperature value; - If the ambient temperature is higher than the threshold temperature and the sampled heat load is higher than the threshold heat load, the setpoint depends on the ambient temperature and a second temperature value that is higher than the first temperature value.

[0018] In some implementations, the margin depends on the value of the first function.

[0019] In some implementations, the margin is a linear function of the first function value.

[0020] In some implementations, the second function is:

[0021] Wherein, a is a coefficient that depends on the magnitude of the voltage value that can be applied to at least one of the fan components, and b is a coefficient that depends on the start-up temperature.

[0022] In some implementations... ,and ,in, It is the minimum voltage value applied to at least one of the fan components. It is the maximum voltage value applied to at least one of the fan components. The startup temperature, and It is the predetermined difference between the maximum temperature and the start-up temperature.

[0023] In some implementations, the method includes the step of modeling airflow (AF) as a polynomial function of the ambient temperature.

[0024] In some implementations, the polynomial function is a sixth-degree polynomial: .

[0025] Among them, all coefficients All are based on a set of values ​​( It was calculated.

[0026] In some embodiments, the method includes the step of comparing an electrical power calculated based on an estimated fan speed with the actual electrical power of a fan in at least one of the fan assemblies.

[0027] This technology also relates to a computer program that includes instructions that, when executed by a computer, cause the computer to perform the steps of a claimed method.

[0028] This technology also relates to a computer-readable medium comprising instructions that, when executed by a computer, cause the computer to perform the steps of a claimed method.

[0029] In the context of this specification, unless otherwise expressly stated, computer system may refer to, but is not limited to, “electronic device,” “operating system,” “system,” “computer-based system,” “controller unit,” “monitoring device,” “control device,” and / or any combination thereof suitable for the present and relevant task.

[0030] In the context of this specification, unless otherwise expressly stated, the terms "computer-readable medium" and "memory" are intended to include media of any nature and kind, with non-limiting examples including RAM, ROM, disks (CD-ROM, DVD, floppy disk, hard disk, etc.), USB keys, flash memory cards, solid-state drives, and tape drives. Still within the context of this specification, "a" computer-readable medium and "the" computer-readable medium should not be construed as the same computer-readable medium. Rather, where appropriate, "a" computer-readable medium and "the" computer-readable medium may also be construed as a first computer-readable medium and a second computer-readable medium.

[0031] In the context of this specification, unless otherwise expressly stated, the words “first,” “second,” “third,” etc., are used as adjectives only to distinguish the nouns they modify from one another, and not to describe any particular relationship between these nouns.

[0032] The implementations of this technology each have at least one of the purposes and / or aspects described above, but not necessarily all of them. It should be understood that some aspects of this technology resulting from attempts to achieve the purposes mentioned above may not satisfy those purposes and / or may satisfy other purposes not specifically described herein.

[0033] Additional and / or alternative features, aspects, and advantages of the implementation of this technology will become apparent from the following description, the accompanying drawings, and the appended claims. Attached Figure Description

[0034] To better understand this technology and its other aspects and additional features, please refer to the following description, which should be used in conjunction with the accompanying drawings, wherein: Figure 1 A high-level functional block diagram of a data center liquid cooling system including a liquid device according to a non-limiting embodiment of the present technology is shown; Figure 2 Non-limiting embodiments according to the present technology are shown. Figure 1 A flowchart of the system's fan adjustment process; Figure 3 It shows Figure 2 The flowchart of the fan regulation subprocess; Figure 4 It shows the use of Figure 1 A high-level function block diagram of the controller for a liquid cooling system.

[0035] It should be understood that, unless otherwise expressly stated herein, the accompanying drawings are not drawn to scale. Detailed Implementation

[0036] The examples and conditions described herein are primarily intended to aid the reader's understanding of the principles of this technology, rather than limiting its scope to the specific examples and conditions described herein. It will be understood that those skilled in the art can devise various arrangements that, while not explicitly described or shown herein, still embody the principles of this technology.

[0037] Furthermore, to aid understanding, the following description may depict a relatively simplified implementation of this technology. Those skilled in the art will understand that various implementations of this technology may involve greater complexity.

[0038] In some cases, examples that are considered helpful to modifications of the present technology may also be illustrated. This is done merely to aid understanding and is not intended to limit the scope of the present technology or to define its boundaries. These modifications are not exhaustive, and those skilled in the art can make other modifications while still remaining within the scope of the present technology. Furthermore, the absence of examples illustrating modifications should not be construed as impossibility of modification or as the description being the only way to implement that element of the present technology.

[0039] Furthermore, all statements herein describing the principles, aspects, and implementations of the present technology, and specific examples thereof, are intended to cover both their structural and functional equivalents, whether they are currently known or will be developed in the future. Therefore, for example, those skilled in the art will understand that any block diagram herein represents a conceptual view of an exemplary circuit embodying the principles of the present technology. Similarly, it will be understood that any flowchart, operation diagram, state transition diagram, pseudocode, etc., represents various processes that can be substantially represented in a non-transitory computer-readable medium and therefore executed by a computer or processor, whether or not such a computer or processor is explicitly shown.

[0040] The functionality of the various elements shown in the figure (including any functional blocks labeled "processor") can be provided using dedicated hardware and hardware capable of executing software in conjunction with appropriate software. When these functions are provided by a processor, they can be provided by a single dedicated processor, a single shared processor, or several separate processors that can be shared. In some implementations of this technology, the processor can be a general-purpose processor (such as a central processing unit (CPU)) or a processor dedicated to a specific purpose (such as a digital signal processor (DSP)). Furthermore, the term "processor" as explicitly used should not be construed as specifically referring to hardware capable of executing software and may implicitly include, but is not limited to, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), read-only memory (ROM) for storing software, random access memory (RAM), and non-volatile memory. Other conventional and / or custom hardware may also be included.

[0041] A software module, or simply a module implied as software, may be represented herein as a flowchart element or any combination of other elements and / or textual descriptions indicating the execution of process steps. These modules may be executed by hardware, whether explicitly or implicitly shown. Furthermore, it should be understood that a module may include, for example, but not limited to, computer program logic, computer program instructions, software, stacks, firmware, hardware circuitry, or combinations thereof, providing the required capabilities.

[0042] It is also worth noting that while the various operations of this invention can be represented by flowchart elements arranged in a specific order, it should be understood that these steps can be combined, subdivided, reordered, or modified to run in parallel without departing from the teachings of this technique. In fact, at least some processing steps can be executed in parallel or serially. Therefore, the order, sequence, and grouping of processing steps are not limitations of this technique.

[0043] Based on this fundamental understanding, the disclosed embodiments are intended to provide a system and method configured to minimize the damaging effects of failure of a critical flow component of a single liquid cooling device by sharing liquid cooling components / sources of other liquid cooling devices.

[0044] Figure 1 A high-level functional block diagram of a data center liquid cooling system 10 according to a non-limiting embodiment of the present technology is shown. The data center liquid cooling system includes a liquid cooling device 100 for cooling corresponding rack-mounted data processing components.

[0045] As shown, the liquid cooling device 100 of system 10 includes: a cooling unit 110 for cooling the liquid circulating in the system, for example, a dry cooling unit; multiple rack-mounted data processing components 120A-120N; multiple smart valves 122A-122N, wherein each smart valve is fluidly connected to a corresponding processing component; and a forward liquid distribution circuit 115 including at least one pump 112A. Figure 1 The system includes two pumps 112A and 112B for supplying cooling liquid from the dry cooling unit 110; a return liquid distribution circuit 125 for returning heated liquid to the dry cooling unit 110; and optionally a control module panel 150 that is communicatively connected to various components and sensors.

[0046] The dry cooling unit 110 can be positioned on any suitable stable supporting surface, such as the top of a data center / computer processing facility building. The dry cooling unit 110 is used to dissipate the heat energy of the heated liquid circulating therein to the surrounding environment. For example, in a data center or similar facility, the dry cooling unit 110 operates to receive heated liquid from the respective rack-mounted data processing components 120A-120N and extract heat energy from the heated liquid by dissipating the heat energy to the surrounding environment via at least one corresponding fan assembly 110A, thus recooling the heated liquid. The dry cooling unit 110 is then operated to supply the recooled liquid back to the respective rack-mounted data processing components 120A-120N.

[0047] As shown, the dry cooling unit 110 includes at least one heat exchanger 110B and at least one fan assembly 110A. The heat exchanger 110B can take various configurations, such as an air-liquid heat exchanger, etc., and the heat exchanger 110B may also include an evaporative cooling pad. For the purposes of this disclosure, the exact configuration of the dry cooling unit 110 and the heat exchanger 110B is not limited, as various configurations can be employed without departing from the concept of this disclosure.

[0048] As also shown, the forward liquid distribution circuit 115 includes at least one pump 112A.

[0049] Optionally, the forward fluid distribution loop 115 also includes a forward "smart" control valve 119A. For the purposes of this disclosure, the term "smart" valve refers to a valve that is independent of pressure and susceptible to temperature effects, including differential pressure regulators that automatically regulate system pressure changes and valves that close under specific operating conditions. Such smart valves may include PICVs, ABQMs, other similarly functional valves, or combinations of valves, such as a combination of a solenoid valve and a control valve. In this implementation, the smart control valve 119A is configured to sense and regulate the flow rate of the cooling / recooling fluid supplied to the processing units 120A-120N. Advantageously, the forward smart control valve 119A is also communicatively coupled to a control module panel to provide notification of fluid flow.

[0050] The thermoplastic fluid from the rack-mounted data processing units 120A-120N is returned to the dry cooling unit 110 for recooling via a corresponding return fluid distribution circuit 125. The return fluid distribution circuit 125 may also include a return intelligent control valve 119B, which is configured to sense and regulate the flow rate of the thermoplastic fluid returning to the dry cooling unit 110.

[0051] As depicted, the dry cooling unit 110 supplies coolant / recoolant to the rack-mounted data processing units 120A-120N at a nominal temperature T, and the reheated liquid returning to the dry cooling units 110, 210 is at a nominal temperature T+ T, of which T represents the temperature difference between the cooling / recooling liquid and the heated liquid.

[0052] The liquid cooling device 100 of system 10 includes multiple rack-mounted data processing units 120A-120N, which receive supplied cooling liquid / recooling liquid through corresponding forward liquid distribution loops 115 to guide the cooling liquid internally to the heat treatment components (e.g., water circulated through water blocks) and deliver the heated liquid from the heat treatment components to the return liquid distribution loops 125.

[0053] The rack-mounted data processing units 120A-120N may or may not be configured with similar heat-generating components. Therefore, the temperature and flow requirements for normal operation of each rack-mounted data processing unit in the rack-mounted data processing units 120A-120N may differ.

[0054] It should be understood that although the rack-mounted data processing units 120A-120N are depicted as arranged in a parallel configuration, this is not intended to be limiting, as the processing units 120A-120N can also be arranged in a series configuration or a combination of parallel and series configurations without departing from the concept of this disclosure.

[0055] As shown, each rack-mounted data processing unit in the rack-mounted data processing units 120A-120N is fluidly connected to a smart valve 122A-122N, which dynamically controls the flow rate of the corresponding processing unit 120A-120N based on the detected liquid temperature.

[0056] Along the forward liquid distribution loop 115, the liquid cooling device 100 also includes a temperature sensor 126 for measuring the temperature T of the supplied cooling liquid. C ; a flow-pressure sensor 127 for measuring the pressure P of the supplied liquid flow; and a volume sensor 128 for measuring the flow rate V of the supplied cooling liquid. C .

[0057] For the return liquid distribution loop 125, the liquid cooling device 100 also includes a temperature sensor 130, which is used to measure the temperature T of the return heated liquid. H .

[0058] As will be described in detail below, another parameter of note is the "pinch" value of the heat exchanger assembly 110B. That is, each heat exchanger assembly 110B has a "hot side" and a "cold side." For the hot side, the pinch value... Thotpinch is defined as the temperature difference between the hot air leaving the heat exchanger and the hot liquid entering the heat exchanger. Additionally, for the cold side, the pinch value... Tcoldpinch is defined as the temperature difference between the temperature of the fresh air entering a heat exchanger and the temperature of the cold liquid leaving the heat exchanger. (Pinch value) Thotpinch and Both Tcoldpinch and Tcoldpinch are positive numbers.

[0059] The measured T C V C T H and T DC (T) DC = T amb Each of these values ​​is provided to the control panel 150. As will be described in detail below, based on these measurements, the module 150 functions to determine an estimate of the power consumed by the rack-mounted data processing components 120A-120N, and to dynamically control the rotational speed of the dry cooling unit fan assembly 110A to improve the efficiency of the cooling system.

[0060] In summary, Figure 2 A flowchart of a fan control process 200 for a liquid cooling system of a rack-mounted data processing assembly according to a non-limiting embodiment of the present technology is shown. The power estimation and fan control process 200 can be executed by a control module 150. Execution by module 150 can be performed by a controller 600.

[0061] For example, Figure 4A high-level functional block diagram describes such a controller. As shown, controller 600 includes one or more cooperating processors (collectively referred to as processor 610 for simplicity), one or more storage devices (collectively referred to as storage device 630 for simplicity), and input / output interface 620 (or separate input and output interfaces) that allows controller 600 to communicate with certain components of liquid cooling device 100. Processor 610 is operatively connected to storage device 630 and input / output interface 620. Storage device 630 includes storage for storing parameters 634, including, for example, but not limited to, the predetermined conductivity thresholds described above. Storage device 630 may include a non-transitory computer-readable medium for storing code instructions 632 executable by processor 610 to allow controller 600 to perform various tasks assigned to controller 600.

[0062] The controller 600 is operably connected to various components of the liquid cooling device 100 via an input / output interface 620. These components include, for example, those used to measure the temperature T of the cooling liquid. C Temperature sensor 126, used to measure the temperature T of an ambient dry cooling unit. DC Temperature sensor 132, used to measure the temperature T of a thermotropic liquid. H Temperature sensor 130 and for measuring coolant flow rate V C The volume sensor 128. The controller 600 executes code instructions 632 stored in the storage device 630 to implement the various functions of the control module 150 described above.

[0063] However, it should be understood that in other embodiments, the power estimation and fan control process 200, or parts thereof, may be performed by relevant components such as temperature sensors, volume sensors, and pumps. For the purposes of this disclosure, the specific one or more entities performing process 200 are not intended to limit the innovative concepts presented herein.

[0064] Back Figure 2 Process 200 begins at step 202, where the heat load Q of the forward liquid distribution circuit 115 is estimated. The heat load Q of the forward liquid distribution circuit 115 corresponds to the power consumed by the rack-mounted data processing units 120A-120N. Therefore, process 200 determines... T, the T represents the temperature of the supplied coolant (measured as T). C ) and the temperature of the returned heat-transforming liquid (measured as T) H The temperature difference between them. Then, process 200 calculates the estimated heat load Q of the forward liquid distribution loop 115 according to the following relationship: heat load ,in, Represents the mass flow rate of water. This represents the specific heat capacity calculated based on the average temperature of the fluid.

[0065] In some implementations, process 200 then proceeds to step 203 to discretize the estimated heat load Q by sampling the heat load Q within a subsequent regular range, the sampled heat load being the maximum value within each range. For example, the load levels can increase in increments of 50 kW, from 50 kW to 800 kW. In this case, if the estimated heat load is 425 kW, the sampled estimated heat load is 450 kW. Alternatively, the load levels can increase in increments of 25 kW. Note that subsequent steps use the heat load measured by the heat exchanger.

[0066] Process 200 then proceeds to step 204 to measure the temperature ΔT of the dry cooling unit 110 and the ambient temperature T. DC The assessment will be conducted, and the ambient temperature will depend on whether an evaporative cooling pad is used. Advantageously, the ambient temperature is considered to be between -30°C and 22°C.

[0067] Then, as will be detailed hereafter, process 200 evaluates the operating parameters of the dry cooling unit fan assembly 110A. These operating parameters may include, but are not limited to, fan speed n. fan (in revolutions per minute: rpm), fan control voltage U (approximately between 0 and 10V), overall efficiency at the operating point (also known as the fan efficiency parameter, η (in %)), and estimated fan power consumption P. elec_calc It should be understood that these operating parameters may be affected by current conditions, such as heat load Q and ambient temperature T of the dry cooling unit. DC Outdoor humidity and other factors.

[0068] In step 206, the ambient temperature, The airflow rate AF is determined by the values ​​of T and the heat load Q of each sampled heat exchanger. Advantageously, process 200 includes a preliminary step 205: expressing the airflow rate as a polynomial function of the ambient temperature. The coefficients of each power of the ambient temperature depend at least on... T and the collected heat load.

[0069] For example, airflow is a sixth-order polynomial of ambient temperature: Among them, all coefficients All are based on a set of values ​​( It was calculated.

[0070] coefficient Favorably between to Between, for example, between to between.

[0071] coefficient Favorably between to Between, for example, between to between.

[0072] coefficient Favorably between to Between, for example, between to between.

[0073] coefficient Favorably between to between.

[0074] coefficient Favorably between 1 to Between, for example, between 1 to between.

[0075] coefficient Favorably between 1 to Between, for example, between 50 to between.

[0076] coefficient Favorably between 1000 to Between, for example, between 4000 to between.

[0077] coefficient The calculation is recalculated in each query of process step 205, so the results may change with each recalculation.

[0078] Then, in step 208, process 200 evaluates the fan speed based on the airflow rate AF and the fan configuration. (in revolutions per minute: rpm).

[0079] Advantageously, process 200 includes a preliminary step 207: determining the fan speed as a function of the airflow AF for the fan configuration, which depends on the pressure drop in the dry cooling unit (particularly the pressure drop caused by a specific heat exchanger assembly 110B of the dry cooling unit) and the efficiency parameter η.

[0080] For example, for each heat exchanger, the fan speed It is a quadratic polynomial of airflow AF: .

[0081] Fan speed Between 0 and Between, for example, between 0 PRM and 980 PRM.

[0082] coefficient Favorably between to between.

[0083] coefficient Favorably between to between.

[0084] coefficient Favorably between to between.

[0085] coefficient Between 0% and 100%.

[0086] Process 200 includes subprocess 210: calculating the voltage applied to the fan. Subprocess 210 will be described in detail later. Then, at step 212, based on the fan speed... Calculate electrical power .

[0087] At step 214, process 200 includes measuring the actual electrical power Pel_meas of fan assembly 110A. If system 10 includes more than one fan, at step 216, process 200 checks whether all fans have electrical power equal to a given tolerance (e.g., ±10%, or ±20%). If not all fans have electrical power equal to the given tolerance (e.g., ±10%, or ±20%), process 200 issues a fan imbalance alarm. If all fans have electrical power equal to the given tolerance (e.g., ±10%, or ±20%), at step 218, the calculated electrical power Pel_calc is compared with the measured electrical power Pel_meas. If the measured electrical power Pel_meas is lower than the calculated electrical power, process 200 exits. If the measured electrical power is higher than the calculated electrical power, a warning of excessive fan power consumption is issued.

[0088] Process 200 is advantageously repeated at a fixed frequency, such as once every 30 minutes or once every 3 minutes.

[0089] Here is an example: the heat load estimated in step 202 is 366 kW. T=18K. In step 203, the estimated heat load Q is sampled to 375kW, considering an increment of 25kW. In step 206, the ambient temperature and temperature difference are used to... The airflow rate is determined by the values ​​of T and the sampled heat load Q. In this example, the airflow rate is a sixth-order polynomial of the ambient temperature: , Among them, all coefficients All are based on a set of values ​​( It was calculated.

[0090] The ambient temperature was measured at 21℃, therefore the airflow rate was calculated as AF(18,300,21) = 41968. Then, at step 208, process 200 evaluates the fan speed based on the airflow rate AF and the fan configuration. (In revolutions per minute: rpm), where the static pressure increases by 78.11 Pa, and the efficiency parameter η is 71.4%. The fan speed is estimated to be... The voltage is 7.8 V. The associated power Pel_calc is 1003.3 W.

[0091] Subprocess 210 is designed to smooth the voltage U applied to the fan assembly 110A, rather than making it reach 0V or the maximum value (e.g., 9V or 10V).

[0092] like Figure 3As shown, subprocess 400 includes step 402: determining the air temperature, referred to as the start-up temperature or minimum temperature, so that the fan operates at the set temperature. It started at a temperature a few degrees Celsius previously. Startup temperature. , recorded as , It can be represented as: , Where M is the margin, which is calculated at step 404 and will be explained in detail later.

[0093] At step 406, subprocess 400 includes adjusting based on ambient temperature. The setpoint is determined by the sampled heat load Q. .

[0094] Advantageously, the setpoint can be selected in the following manner. : - If the ambient temperature is below or equal to the threshold temperature (e.g., 22°C), and the sampled heat load Q is below or equal to the threshold heat load (e.g., 500kW), then the setpoint is the lower setpoint. (For example, 25°C) (This means the pinch point is 3K).

[0095] This design utilizes cold ambient air, which, due to its low heat load, can easily cool the liquid entering the dry cooling system. - If the ambient temperature is below or equal to the threshold temperature (e.g., 22°C), and the sampled heat load Q is higher than the threshold heat load (e.g., 500 kW), then the setpoint is the higher setpoint. For example, 27℃ (meaning the pinch point is 5K).

[0096] This design utilizes cold ambient air, but due to the higher heat load (compared to the design described above), the setpoint is chosen to be higher; - If the ambient temperature is above the threshold temperature (e.g., 22°C) and the sampled heat load Q is less than or equal to the threshold heat load (e.g., 500kW), then the setpoint is selected as follows: ,in, For example, 3K. This design takes into account the insufficient freshness of the air and the low heat load; - If the ambient temperature is higher than the threshold temperature (e.g., 22°C) and the sampled heat load Q is higher than the threshold heat load (e.g., 500kW), then the setpoint is selected as follows: ,in, For example, 5K; This design takes into account the insufficient freshness of the air and the high heat load.

[0097] Control the water temperature to ensure it does not exceed the maximum temperature. The highest temperature is recorded as equal to Difference It was pre-ordered. You can choose between 1K and 15K, for example, between 5K and 10K. .

[0098] Now for reference Figure 3 Describe subprocess 210.

[0099] Sub-process 210 includes step 210-1: calculating the voltage U as a function, for example, this function is proportional to the minimum of two functions (the first function is called Umax and the second function is called Ucalc), as described below: .

[0100] Where C is the selected coefficient, which can be selected between 0.01 and 100, or between 1 and 10, for example, it can be 1.

[0101] Sub-process 210 includes step 210-2, which precedes step 210-1: calculation .function With fan speed Proportional. For example, if U is chosen to be between 0V and 10V, it can be expressed as: , in, It is the maximum voltage that can be applied to the fan, such as 10V, and This is the maximum speed value. For example, as well as : .

[0102] Sub-process 210 includes step 210-3, which precedes step 210-1: calculation .function Depends on the temperature of the liquid at the inlet of the dry cooling unit and temperature difference Advantageously, the function It is the temperature increment Linear functions: , in, ,as well as .

[0103] It is the minimum voltage that can be applied to the fan.

[0104] The advantage is that Alternative or additional land may be selected. .

[0105] Advantageously, the margin M is A function (expressed as a percentage). For example, M is... A linear function. For example, , where c and d are coefficients.

[0106] The margin M is a value used to calculate the fan kick-on setpoint; therefore, it is in degrees Celsius.

[0107] The coefficient c can depend on the amplitude of the fan voltage, while the coefficient d can depend on the coefficient c. For example ,as well as , in, It is the percentage of the maximum voltage that can be applied to the fan. It is the percentage of the minimum voltage that can be applied to the fan. For example, Between 0V and 2V Between 9V and 10V Between 5% and 15%, and Between 90% and 100%.

[0108] Advantageously, subprocess 210 is executed at a fixed frequency, such as once per minute, thereby ensuring optimized control of the fan.

[0109] Now, for example, the estimated heat load at step 202 is 600 kW, and the heat load of each heat exchanger is 300 kW. T=20K. In step 203, the estimated heat load Q is sampled to 300kW. In step 206, the ambient temperature and temperature difference are used to... The airflow rate is determined by the values ​​of T and the sampled heat load Q, such as Figure 3 As shown. In this example, the airflow rate is a sixth-order polynomial of the ambient temperature: , Among them, coefficient yes ,coefficient yes ,coefficient yes ,coefficient yes ,coefficient yes ,coefficient yes ,coefficient yes .

[0110] The ambient temperature measured upstream of the heat exchanger was 13.1℃, from which the air flow rate was calculated to be [value missing]. Then, at step 208, process 200 evaluates the fan speed based on the airflow rate AF and fan configuration, with a static pressure increase of 107 Pa. (In revolutions per minute: rpm). The estimated fan speed is... Maximum voltage The voltage is 6.2 V (step 210-2). The associated power... It is 587W.

[0111] The parameters are as follows: (Pre-booked) and (Planned). Inlet water temperature measurement: .

[0112] Given the ambient temperature and Then select the grip point as 5K, and .

[0113] pass and To calculate the margin M, such that and and .

[0114] Therefore, the lowest temperature equals and .

[0115] The voltage applied to the fan is U calc and U max The minimum value in, i.e., U calc =2.1V, which means the fan speed is... equal 21%.

[0116] While the above implementations have been described and illustrated with reference to specific steps performed in a particular order, it should be understood that these steps can be combined, subdivided, or reordered without departing from the teachings of this art. At least some of the steps can be performed in parallel or sequentially. Therefore, the order and grouping of steps are not limitations of this art.

[0117] Please note that steps 202, 206, and 208 can be achieved through modeling or by measuring the heat load, airflow, and fan speed separately.

[0118] The heat exchanger can be any suitable type, such as a cooling device, a cooling tower, or even a plate heat exchanger.

[0119] Modifications and improvements to the above-described implementation of this technology will be apparent to those skilled in the art. The above description is intended to be exemplary and not restrictive. Therefore, the scope of this technology is limited only by the scope of the appended claims.

Claims

1. A liquid cooling method (50) for a data center liquid cooling system (10), the data center liquid cooling system being used to cool rack-mounted data processing components (120A-120N), the data center liquid cooling system comprising a cooling device (100), the cooling device (100) being provided with: Cooling unit (110), configured to supply coolant to the rack-mounted data processing units (120A-120N) and receive heated coolant from the rack-mounted data processing units (120A-120N), A forward liquid distribution circuit (115) including a temperature sensor (126) is provided for distributing the cooling liquid from the cooling unit to the rack-mounted data processing unit (120A-120N). A return liquid distribution circuit (125) is provided for delivering the heated liquid from the rack-mounted data processing unit (120A-120N) back to the cooling unit (110) and the temperature sensor (130). A control module panel (150) is communicatively connected to the temperature sensor (126) to receive data from the temperature sensor (126). Each rack-mounted data processing unit (120A-120N) includes at least one heat-generating electronic processing element and at least one liquid cooling block, the at least one liquid cooling block being arranged to make corresponding thermal contact with the at least one heat-generating electronic processing element and to be fluidly connected to the forward liquid distribution circuit (115). The cooling unit (110) includes at least one fan assembly (110A) and at least one heat exchanger assembly (110B), the cooling unit being used to cool the liquid circulating in the system (10) and entering the cooling unit (110) through the return liquid distribution loop (125) by means of an airflow. The method includes the following steps: - Estimate the heat load (Q) of the rack-mounted data processing unit (120A-120N) (202). - Based on the heat load (Q) and the ambient temperature of the airflow ( ), to estimate airflow (AF) (206). - Based on the estimated airflow rate (AF), the fan rotation speed of the at least one fan assembly (110A) is estimated (208), and - Based on the estimated fan rotation speed ( ), to control the voltage of the at least one fan assembly (110A) of (210), The voltage depends on the minimum of two function values: a first function (Umax) and a second function (Ucalc). The first function (Umax) depends on the estimated fan speed (nfan), and the second function (Ucalc) depends on the temperature of the liquid inlet.

2. The method of claim 1, wherein, The voltage is proportional to the minimum of the two function values ​​(Umax, Ucalc).

3. The method of claim 1 or 2, wherein, The first function (Umax) is proportional to the estimated fan rotation speed (nfan).

4. The method according to one of the preceding claims, wherein, The second function (Ucalc) is a linear function of the liquid inlet temperature.

5. The method according to any one of the preceding claims, wherein, The fan is activated when the air temperature is below a set point, which depends on the ambient temperature. The difference between the setpoint and the start-up temperature is called the margin (M) and / or the heat load (Q).

6. The method according to the preceding claim, wherein, The set point is: - If the ambient temperature ( If the setpoint is lower than or equal to the threshold temperature and the sampled heat load (Q) is lower than or equal to the threshold heat load, then the setpoint is a low setpoint (TiL). - If the ambient temperature ( If the setpoint is lower than or equal to the threshold temperature and the sampled heat load (Q) is higher than the threshold heat load, then the setpoint is a high setpoint (TiH), which is higher than the low setpoint (TiL). - If the ambient temperature ( If the ambient temperature is higher than the threshold temperature and the sampled heat load (Q) is lower than or equal to the threshold heat load, then the setpoint depends on the ambient temperature and the first temperature value. - If the ambient temperature ( If the ambient temperature is higher than the threshold temperature and the sampled heat load (Q) is higher than the threshold heat load, then the setpoint depends on the ambient temperature and a second temperature value, where the second temperature value is higher than the first temperature value.

7. The method according to claim 5 or 6, wherein, The margin depends on the value of the first function.

8. The method according to the preceding claim, wherein, The margin is a linear function of the first function value.

9. The method according to any one of claims 5 to 8, wherein the second function is: in, a is a coefficient that depends on the magnitude of the voltage value that can be applied to at least one of the fan components (110A), and b is a coefficient that depends on the start-up temperature.

10. The method according to the preceding claim, wherein, ,and ,in, It is the minimum voltage value applied to at least one of the fan components (110A). It is the maximum voltage value applied to at least one of the fan components (110A). The startup temperature, and It is a predetermined difference between the maximum temperature and the start-up temperature.

11. The method according to any one of the preceding claims, the method comprising the step (205): modeling the air flow rate (AF) as a polynomial function of the ambient temperature.

12. The method according to the preceding claim, wherein, The polynomial function is a sixth-degree polynomial: 。 13. The method according to any one of the preceding claims, the method comprising the step (218): comparing an electrical power calculated based on an estimated fan speed with the actual electrical power of a fan in at least one of the fan assemblies.

14. A computer program product comprising instructions that, when executed by a computer, cause the computer to perform the steps of the method according to any one of claims 1 to 13.

15. A computer-readable medium comprising instructions that, when executed by a computer, cause the computer to perform the steps of the method according to any one of claims 1 to 13.

Images