Control method and device of immersion liquid cooling heat exchange system, terminal and program product

Abstract

The application provides a control method and device of an immersed liquid cooling heat exchange system, a terminal and a program product. The heat exchange system comprises a cabinet, a first cooling loop and a flow rate adjusting device. The control method comprises the following steps: acquiring a real-time temperature of a specified object in a server; determining an adjustment strategy of a current adjustment period based on a relationship between the real-time temperature and a preset temperature safety threshold and in combination with a compensation condition; when the compensation condition is met, the compensation condition comprises that the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object; obtaining a final adjustment value according to a compensation adjustment value and a basic adjustment value, wherein the basic adjustment value is determined by using a basic temperature difference adjustment strategy and based on at least a temperature difference between an inlet liquid temperature and a return liquid temperature of the first cooling loop; and adjusting a working state of the flow rate adjusting device by using the final adjustment value to adjust a cooling liquid delivery flow rate of the first cooling loop per unit time. The above scheme can reduce the hysteresis of heat dissipation and meet the timely heat dissipation demand when the power consumption of the server suddenly changes.

Description

Technical Field

[0001] The present invention relates to an immersion liquid cooling heat exchange system, and more particularly to a control method, device, terminal, and program product for an immersion liquid cooling heat exchange system. Background Technology

[0002] In data center immersion liquid cooling systems, the heat exchange system is the core unit for ensuring the safe operation of servers and controlling energy consumption. Its control strategy directly determines the heat dissipation reliability of the servers installed in the immersion liquid cooling system and the energy efficiency level of the entire system. Currently, mainstream immersion liquid cooling heat exchange systems in the industry all adopt a single feedback control mode based on the temperature difference between the inlet and outlet of the coolant: by collecting the temperature signals at the inlet and outlet of the heat exchange system, comparing the difference between the outlet temperature and the inlet temperature with the set target temperature, the circulation pump speed or system fan speed is adjusted in real time to achieve passive adjustment of the cooling capacity.

[0003] However, the above methods of cooling the server cannot cope with the server's transient heat peaks, causing some high-power chips and other components in the server to overheat locally, affecting the server's performance. Summary of the Invention

[0004] The technical problem solved by the embodiments of the present invention is that the heat dissipation of the server based on the temperature difference between the inlet and outlet of the coolant is unable to cope with the transient heat peaks of the server, causing some high-power chips and other components in the server to overheat locally, affecting the performance of the server.

[0005] To address the aforementioned technical problems, this invention provides a control method for an immersion liquid cooling heat exchange system. The immersion liquid cooling heat exchange system includes a server rack, a first cooling circuit, and a flow rate adjustment device. The first cooling circuit is fluidly connected to the server rack and is used to supply coolant to the server rack to cool a server immersed in the coolant within the rack. The control method for the immersion liquid cooling heat exchange system includes: acquiring the real-time temperature of a specified object within the server; determining an adjustment strategy for the current adjustment cycle based on the relationship between the acquired real-time temperature and a preset temperature safety threshold of the specified object, combined with compensation conditions; wherein, when the compensation conditions are met, the adjustment strategy is determined to be a compensation adjustment strategy, and the compensation conditions include the... If the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object, the compensation adjustment strategy is configured as follows: determine the compensation adjustment value, and obtain the final adjustment value based on the compensation adjustment value and the basic adjustment value. The basic adjustment value is determined using a basic temperature difference adjustment strategy based at least on the temperature difference between the inlet temperature and the return temperature of the first cooling circuit. The inlet temperature is the temperature of the coolant input from the first cooling circuit to the cabinet, and the return temperature is the temperature of the coolant output from the cabinet to the first cooling circuit. The final adjustment value obtained by the adjustment strategy determined in the current adjustment cycle is used to adjust the working state of the flow rate adjustment device to adjust the coolant delivery flow rate of the first cooling circuit per unit time.

[0006] Optionally, determining the compensation adjustment value includes: determining the adjustment increment value according to the first adjustment step size, and obtaining the compensation adjustment value based on the sum of the compensation adjustment value of the previous adjustment cycle and the adjustment increment value.

[0007] Optionally, the first adjustment step size is a preset fixed step size; or, the first adjustment step size is determined in the following way: based on the first mapping relationship between the difference and the adjustment step size, the first adjustment step size is determined according to the difference between the real-time temperature and the preset temperature safety threshold of the specified object, wherein the difference is positively correlated with the first adjustment step size.

[0008] Optionally, after adjusting the working state of the flow rate adjustment device using the final adjustment value obtained by the adjustment strategy determined in the current adjustment cycle, the method further includes: obtaining the actual rotation speed of the flow rate adjustment device; if the actual rotation speed is less than the first rotation speed threshold of the flow rate adjustment device, then continuing to obtain the real-time temperature of the specified object, and in the next adjustment cycle, determining the adjustment strategy for the next adjustment cycle based on the relationship between the obtained real-time temperature and the preset temperature safety threshold of the specified object.

[0009] Optionally, the control method for the immersion liquid cooling heat exchange system further includes: outputting an alarm reminder if the actual rotational speed is greater than or equal to a first rotational speed threshold of the flow rate adjustment device; and / or, locking the final adjustment value to a maximum adjustment value if the actual rotational speed is greater than or equal to a second rotational speed threshold, wherein the maximum adjustment value is used to control the flow rate adjustment device to rotate at the maximum rotational speed, and the second rotational speed threshold is greater than or equal to the first rotational speed threshold.

[0010] Optionally, the compensation conditions further include: the real-time temperature is less than the preset temperature safety threshold of the specified object, and the compensation adjustment value of the previous adjustment cycle is greater than zero; determining the compensation adjustment value includes: determining the adjustment reduction value according to the second adjustment step size, and obtaining the compensation adjustment value based on the difference between the compensation adjustment value of the previous adjustment cycle and the adjustment reduction value.

[0011] Optionally, the second adjustment step size is a preset fixed step size; or, the second adjustment step size is determined in the following way: based on the second mapping relationship between the absolute value difference and the adjustment step size, the second adjustment step size is determined according to the absolute value difference between the real-time temperature and the preset temperature safety threshold of the specified object.

[0012] Optionally, determining the adjustment strategy for the current adjustment cycle based on the relationship between the acquired real-time temperature and the preset temperature safety threshold of the specified object includes: if the real-time temperature is less than the preset temperature safety threshold of the specified object and the compensation adjustment value of the previous adjustment cycle is zero, then the adjustment strategy for the current adjustment cycle is determined to be a basic temperature difference adjustment strategy; the basic temperature difference adjustment strategy is configured as follows: based on the temperature difference, the temperature difference setpoint, and the set deviation value, the basic adjustment value is determined using a PID algorithm, and the basic adjustment value is used as the final adjustment value determined by the adjustment strategy for the current adjustment cycle, wherein the set deviation value is a value that allows the temperature difference to fluctuate around the temperature difference setpoint.

[0013] Optionally, determining the basic adjustment value using a PID algorithm based on the temperature difference, the temperature difference setpoint, and the set deviation value includes: if the temperature difference is greater than or equal to a first value, increasing the basic adjustment value based on the current basic adjustment value according to the PID algorithm, where the first value is the sum of the temperature difference setpoint and the set deviation value; if the temperature difference is less than or equal to a second value, decreasing the basic adjustment value based on the current basic adjustment value according to the PID algorithm, where the second value is the difference between the temperature difference setpoint and the set deviation value; and if the temperature difference is greater than the second value and less than the first value, maintaining the current basic adjustment value.

[0014] Optionally, obtaining the real-time temperature of a specified object within the server includes any of the following: sending an HTTP GET request to the baseboard management controller of the server based on the Redfish protocol, wherein the HTTP GET request carries a temperature acquisition instruction, the temperature acquisition instruction being used to read the real-time temperature of the specified object within the server from the baseboard management controller of each server; or sending an SDR command to the baseboard management controller of each server in the rack using the IPMI protocol, the SDR command being used to query and obtain the real-time temperature data of the specified object within the server from the baseboard management controller.

[0015] This application also provides a control device for an immersion liquid cooling heat exchange system. The immersion liquid cooling heat exchange system includes: a cabinet, a first cooling circuit, and a flow rate adjustment device. The first cooling circuit is in fluid communication with the cabinet and is used to provide coolant to the cabinet to cool the server immersed in the coolant within the cabinet. The control device for the immersion liquid cooling heat exchange system includes: an acquisition unit for acquiring the real-time temperature of a specified object within the server; and an adjustment strategy determination unit for determining an adjustment strategy for the current adjustment cycle based on the relationship between the acquired real-time temperature and a preset temperature safety threshold of the specified object, combined with compensation conditions. When the compensation conditions are met, the adjustment strategy is determined to be a compensation adjustment strategy, and the compensation conditions include the... If the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object, the compensation adjustment strategy is configured as follows: determine the compensation adjustment value, and obtain the final adjustment value based on the compensation adjustment value and the basic adjustment value. The basic adjustment value is determined by using a basic temperature difference adjustment strategy based at least on the temperature difference between the inlet temperature and the return temperature of the first cooling circuit. The inlet temperature is the temperature of the coolant input from the first cooling circuit to the cabinet, and the return temperature is the temperature of the coolant output from the cabinet to the first cooling circuit. A processing unit is used to adjust the working state of the flow rate adjustment device using the final adjustment value obtained by the adjustment strategy determined in the current adjustment cycle, so as to adjust the coolant delivery flow rate of the first cooling circuit per unit time.

[0016] This application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, performs the steps of any of the above-described control methods for an immersion liquid cooling heat exchange system.

[0017] This application also provides a terminal, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the steps of any of the above-described control methods for an immersion liquid cooling heat exchange system when running the computer program.

[0018] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of any of the above-described control methods for an immersion liquid cooling heat exchange system.

[0019] Compared with the prior art, the technical solution of the embodiments of the present invention has the following beneficial effects: In existing technologies, when the power consumption of a single or multiple servers suddenly changes, the cooling capacity adjustment is only implemented after the overall power consumption of the servers increases and the accumulated heat causes the temperature difference between the inlet and outlet coolant to rise. Before the temperature difference increases, the heat exchange system maintains the original coolant flow rate, failing to proactively prevent transient heat peaks and exhibiting a significant lag in temperature control response. In contrast, this application obtains the real-time temperature of a specified object within the server and, based on the relationship between the obtained real-time temperature and a preset temperature safety threshold for the specified object, determines the adjustment strategy for the current adjustment cycle in conjunction with compensation conditions. When the real-time temperature is greater than or equal to the preset temperature safety threshold for the specified object, i.e., when the compensation conditions are met, a compensation adjustment strategy is used to determine the compensation adjustment value. The final adjustment value is obtained based on the compensation adjustment value and a basic adjustment value. The basic adjustment value is determined using a basic temperature difference adjustment strategy, based at least on the temperature difference between the inlet temperature of the coolant input from the first cooling loop to the rack and the return temperature of the coolant output from the rack. That is, when the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object, a compensation adjustment strategy is used to add a compensation adjustment value to the basic adjustment value to obtain the final adjustment value. The working state of the flow rate adjustment device is adjusted using the final adjustment value to adjust the coolant delivery flow rate of the first cooling circuit per unit time. By acquiring the real-time temperature of a specified object on the server, a strong correlation is established between the real-time temperature of the specified object and the coolant flow rate per unit time of the first cooling loop. This breaks down the control silo between the immersion liquid cooling heat exchange system and the server, allowing the real-time temperature of the specified object on the server to be considered when determining the coolant flow rate per unit time of the first cooling loop. The real-time temperature of the specified object is used as the basis for adjusting the heat exchange system, ensuring that the coolant flow rate per unit time of the first cooling loop matches the actual heat dissipation requirements of the server. This accurately matches the heat load of the specified object on the server with the cooling capacity that the first cooling loop can provide, enabling timely heat dissipation of the server, reducing heat dissipation lag, meeting the timely heat dissipation requirements when the server experiences sudden power consumption changes, improving the accuracy of adjusting the coolant flow rate per unit time of the first cooling loop, avoiding server overheating, frequency throttling, computing power attenuation, and downtime, and improving the server's heat dissipation effect and operational stability.

[0020] Furthermore, by adjusting the coolant flow rate of the first cooling circuit per unit time based on the real-time temperature of the specified object on the server, the precision of the coolant flow rate adjustment of the first cooling circuit per unit time can be improved, thereby improving the effectiveness of the coolant flow rate adjustment of the first cooling circuit per unit time, avoiding overcooling and ineffective adjustment, significantly reducing the energy consumption of the immersion liquid cooling heat exchange system, saving energy, and meeting the development requirements of green data centers. Attached Figure Description

[0021] Figure 1 This is a flowchart of a control method for an immersion liquid cooling heat exchange system according to an embodiment of the present invention; Figure 2 This is a flowchart of another control method for an immersion liquid cooling heat exchange system in an embodiment of the present invention; Figure 3 This is a schematic diagram of the control device for another immersion liquid cooling heat exchange system in an embodiment of the present invention. Detailed Implementation

[0022] To make the above-mentioned objectives, features and beneficial effects of the embodiments of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] This application provides a control method for an immersion liquid cooling heat exchange system. The control method is used to control the immersion liquid cooling heat exchange system, specifically, to control the coolant flow rate per unit time in the first cooling loop of the system. By controlling the coolant flow rate per unit time in the first cooling loop, the cooling capacity of the first cooling loop can be adjusted. When the temperature of a specified object within the server is high, increasing the coolant flow rate per unit time in the first cooling loop can improve the cooling capacity of the first cooling loop, thereby quickly cooling the specified object within the server. When the temperature of the specified object within the server decreases, decreasing the coolant flow rate per unit time in the first cooling loop reduces the cooling capacity of the first cooling loop, thus saving energy while still meeting the cooling requirements of the specified object within the server.

[0024] The immersion liquid cooling heat exchange system includes: a cabinet, a first cooling circuit, and a flow rate adjustment device. The first cooling circuit is in fluid communication with the cabinet and is used to supply coolant to the cabinet to cool the server immersed in the coolant inside the cabinet.

[0025] The immersion liquid cooling heat exchange system also includes a heat exchanger and a second cooling circuit. The coolant in the first cooling circuit and the coolant in the second cooling circuit exchange heat through the heat exchanger. The coolant in the first cooling circuit is cooled by the coolant in the second cooling circuit, thus lowering its temperature. The second cooling circuit continuously cools the coolant in the first cooling circuit, thereby continuously supplying low-temperature coolant to the cabinet.

[0026] The first cooling circuit and the second cooling circuit can be matched one-to-one. Each first cooling circuit is configured with a second cooling circuit and heat is exchanged through a corresponding heat exchanger.

[0027] The first cooling circuit and the second cooling circuit can also have a many-to-one relationship, that is, one second cooling circuit can supply coolant to multiple first cooling circuits.

[0028] The immersion liquid-cooled heat exchange system also includes a main controller, which is used to execute control methods. The main controller can be a PLC main controller, a microcontroller (MCU), or a single-chip microcomputer, or other suitable controller.

[0029] Reference Figure 1 The control method for an immersion liquid-cooled heat exchange system includes the following steps: Step S11: Obtain the real-time temperature of a specified object within the server.

[0030] The specified object within the server can be a core chip, such as a central processing unit (CPU) or a graphics processing unit (GPU).

[0031] In step 11, the real-time temperature of a specified object can be periodically acquired according to a set temperature acquisition cycle. The temperature acquisition cycle can be a fixed period of duration. The real-time temperature of the specified object on the server is acquired periodically according to this fixed cycle. The duration of the fixed cycle can be configured based on factors such as the type of the specified object, its heat generation, and the impact of overheating. For example, the duration of the fixed cycle could be 1 second, 0.5 seconds, or other suitable values.

[0032] The temperature acquisition cycle can be dynamically configured. For example, the cycle length can be dynamically adjusted based on the real-time temperature of a specified object. When the acquired real-time temperature of the specified object is greater than or equal to a first temperature threshold, the temperature acquisition cycle length is reduced, and the real-time temperature sampling frequency of the specified object is increased to better monitor the real-time temperature of the specified object. This helps to detect overheating of the specified object in a timely manner and thus take cooling measures promptly. When the acquired real-time temperature of the specified object is less than the first temperature threshold, the temperature of the specified object is within a relatively safe temperature range, and the temperature acquisition cycle length and the real-time temperature sampling frequency of the specified object can be reduced, thereby saving energy consumption. The first temperature threshold is less than a preset safe temperature threshold for the specified object.

[0033] In practice, the real-time temperature of a specified object within the server can be obtained through various methods.

[0034] For example, a communication connection is established based on the Redfish protocol. A temperature acquisition command is sent to the server using an HTTP GET request within the Redfish protocol. This command is used to obtain the real-time temperature of a specified object within the server. Specifically, the temperature acquisition command is sent to the Baseboard Management Controller (BMC) within the server via an HTTP GET request. The Redfish protocol, based on a RESTful API design, is an industry-standard server management protocol that can directly read raw real-time temperature data from core computing components such as the CPU and GPU within the server.

[0035] For example, the IPMI protocol can be used to send SDR commands to the baseboard management controller (BMC) of each server in the rack. SDR commands are used to query and obtain the real-time temperature of a specified object within the server. Specifically, IPMI is an intelligent platform management interface, a hardware-level remote management interface. It accesses sensor data stored in the server's BMC. The SDR command can directly query and obtain the real-time temperature of chips such as the server's CPU and GPU, and can also obtain auxiliary hardware health information such as voltage, fan speed, and power status. The IPMI protocol can be used to achieve high-frequency temperature acquisition of specified objects such as chips at the hardware level, to adapt to servers that do not support the Redfish protocol.

[0036] In practical applications, it can simultaneously support both the Redfish and IPMI protocols to ensure compatibility with different server protocol types and adapt to different hardware environments.

[0037] In one embodiment, in step S11, the real-time temperature of a specified object within the server can be acquired through a hardware status acquisition module. The hardware status acquisition module has a device management protocol adapted to the server and communicates with the server to receive the real-time temperature of the specified object within the server from the baseboard management controller (BMC).

[0038] The hardware status acquisition module may include one or both of the IPMI protocol processing unit and the Redfish protocol processing unit.

[0039] The IPMI protocol processing unit is electrically connected to both the baseboard management controller and the main controller of the server. The IPMI protocol processing unit includes a first network interface and a first protocol controller. The first network interface is connected to the baseboard management controller via a serial bus. The first protocol controller is electrically connected to both the first network interface and the main control unit. The first protocol controller manages the IPMI communication process, and when sending data (such as SDR commands), it encapsulates data according to IPMI protocol requirements and parses the real-time temperature of the received specified object. Specifically, IPMI is an intelligent platform management interface, a hardware-level remote management interface. The serial bus is an I2C bus or an SMBus bus, etc.

[0040] The IPMI protocol processing unit also includes a first buffer. The first buffer is used to buffer the Sensor Data Record (SDR) command to be sent to the substrate management controller and the real-time temperature of the specified object returned by the substrate management controller. The SDR command is used to obtain the real-time temperature of the specified object.

[0041] The Redfish protocol processing unit is electrically connected to both the baseboard management controller and the main controller. The Redfish protocol processing unit includes a second network interface and a second protocol controller. The second network interface is connected to the baseboard management controller via Ethernet, and the second protocol controller is electrically connected to both the second network interface and the main controller. The second network interface can be a suitable interface such as an RJ45 or SFP interface. The second protocol controller manages the Redfish communication process, and when sending data (such as HTTP GET requests), it encapsulates data according to Redfish protocol requirements and parses the real-time temperature of the received specified object. The Redfish protocol is designed based on a RESTful API and is an industry-standard server management protocol, capable of directly reading raw real-time temperature data from core computing components such as the CPU and GPU in the server.

[0042] The Redfish protocol processing unit also includes a second buffer. The second buffer is used to buffer the temperature acquisition command from the baseboard management controller to be sent to the server, as well as the real-time temperature of the specified object returned by the baseboard management controller. The temperature acquisition command is used to obtain the real-time temperature of the specified object.

[0043] The Redfish protocol processing unit establishes a connection with the server's BMC via HTTPS. For example, the main controller controls the Redfish protocol processing unit to periodically send HTTP GET requests to the server's BMC, carrying temperature acquisition instructions in the HTTP GET requests to send the temperature acquisition instructions to the server's BMC.

[0044] It should be noted that when the above hardware status acquisition module includes both the IPMI protocol processing unit and the Redfish protocol processing unit, the first buffer and the second buffer can be two independent buffers, or they can be a single buffer, or they can be two partitions integrated in the same buffer.

[0045] In other embodiments, in step S11, the real-time temperature of a specified object can be obtained from the BMC in the server via the BMC interface circuit. The BMC interface circuit can be connected to the BMC in the server via Ethernet. The BMC interface circuit includes: an Ethernet physical layer interface module, a protocol conversion module, a level matching module, and an auxiliary power supply module.

[0046] The input of the Ethernet physical layer interface module is electrically connected to the independent management port of the server's baseboard management controller (BMC). Real-time temperature of a specified object is received from the server's BMC via the input of the Ethernet physical layer interface module.

[0047] The protocol conversion module has a first terminal and a second terminal. The first terminal is electrically connected to the output terminal of the Ethernet physical layer interface module, and the second terminal is electrically connected to the input terminal of the level matching module. The protocol conversion module receives the real-time temperature of a specified object from the Ethernet physical layer interface module, converts the real-time temperature of the specified object into a standard format, and outputs it to the level matching module via the second terminal. This enables indirect communication between the server's BMC and the main controller via the BMC interface circuit. Furthermore, the protocol conversion module provides electrical isolation between the server's BMC and the main controller, and also enables transparent transmission of IPMI / Redfish protocol data.

[0048] The input terminal of the level matching module is electrically connected to the second terminal, and the output terminal of the level matching module is electrically connected to the first data input terminal of the main controller, completing signal matching between different level domains. By matching signals between different level domains through the level matching module, logic level compatibility can be achieved, preventing logical judgment errors or physical damage. In step S11 above, the real-time temperature is obtained by the main controller via the first data input terminal.

[0049] The auxiliary power supply modules provide power to the Ethernet physical layer interface module, the protocol conversion module, and the level matching module.

[0050] When using the BMC interface circuit to obtain the real-time temperature of a specified object from the BMC in the server, the main controller uses a microcontroller unit (MCU). The MCU includes an IPMI protocol interface and / or a Redfish protocol interface that are adapted to the server's device management protocol.

[0051] For example, an MCU includes a communication submodule, a logic processing circuit board, an analog signal acquisition submodule, and a drive output submodule.

[0052] The communication submodule has an IPMI protocol interface and / or a Redfish protocol interface adapted to the server's device management protocol. That is, the communication submodule can have one or both of the IPMI and Redfish protocol interfaces. The communication submodule can receive the real-time temperature of a specified object from the server's BMC via the BMC interface, and can also output commands or data (such as HTTP GET requests, SDR commands, etc.) to the server's BMC.

[0053] Furthermore, the communication submodule includes: RMII bus transceiver pins and a hardware-embedded protocol firmware storage unit. The RMII bus transceiver pins are electrically connected to the Ethernet physical layer interface module in the BMC interface circuit; the hardware-embedded protocol firmware storage unit contains built-in IPMI protocol stack firmware and / or Redfish protocol stack firmware, is electrically connected to the RMII bus transceiver pins, and is also electrically connected to the logic processing board. In this way, CPU power consumption, board temperature, and device fault alarm data reported by the server can be directly parsed without the need for an external protocol conversion chip.

[0054] The first data input terminal of the logic processing circuit board is electrically connected to the IPMI protocol interface and the Redfish protocol interface to receive the real-time temperature of the specified object output by the server's BLM. The logic processing circuit board is used to output control commands. Furthermore, the second data input terminal of the logic processing circuit board serves as the second data input terminal of the main controller, used to receive inlet and outlet liquid temperatures, etc.

[0055] The logic processing circuit board may include a main control core, a cache unit, and a threshold comparison unit. A preset temperature safety threshold is used to generate control commands. The cache unit temporarily stores BMC server data and pipeline temperature data; the threshold comparison unit pre-stores liquid cooling temperature control thresholds (e.g., upper limit of inlet temperature, upper limit of return temperature, preset temperature safety threshold, etc.), and generates corresponding speed regulation / start / stop control commands when measured parameters (actual speed, real-time temperature, etc.) exceed the limits.

[0056] The analog signal acquisition submodule is electrically connected to the logic processing circuit board. The analog signal acquisition submodule includes an analog-to-digital conversion (ADC) unit. The ADC unit receives the inlet and outlet temperatures, which are initially in analog signal format. These are converted into digital signals and output to the logic processing circuit board. In some embodiments, multiple ADC units can be configured to improve the analog-to-digital conversion efficiency. Furthermore, the analog signal acquisition submodule can incorporate a cold junction compensation circuit to eliminate temperature drift caused by environmental temperature drift during immersion in the liquid cooling system, achieving accurate temperature measurement. The cold junction compensation circuit can utilize existing circuit structures and will not be elaborated upon here.

[0057] The drive output submodule is electrically connected to the logic processing circuit board. The drive output submodule includes an optocoupler isolation circuit and a power drive chip. The optocoupler isolation circuit provides optocoupler isolation for control commands, while the power drive chip amplifies the control signals before outputting them to the actuator. The drive output submodule is configured with an optocoupler isolation circuit and a power drive chip, ensuring that control commands are optocoupled and then amplified by the power drive chip to control actuators such as flow rate adjustment devices (e.g., circulating pumps) and electric regulating valves. This isolates the weak current on the main controller side from the strong current on the actuator side, preventing reverse load surges that could damage the MCU. Existing structures can be used for the optocoupler isolation circuit and power drive chip, and will not be elaborated upon here.

[0058] This main controller supports both the IPMI and Redfish protocols to ensure compatibility with mainstream servers and cross-brand servers. IPMI is the intelligent platform management interface, a hardware-level remote management interface. The Redfish protocol, based on a RESTful API design, is an industry-standard server management protocol that can directly read raw, real-time data such as the temperature of core computing components like the CPU and GPU.

[0059] When the main controller uses the IPMI protocol interface, it can output a Sensor Data Record (SDR) command and transmit it to the BMC in the server via the IPMI protocol interface and the BMC interface circuit. When the main controller uses the Redfish protocol interface, it can output an HTTP GET request, carrying a temperature acquisition command in the HTTP GET request, and transmit the temperature acquisition command to the BMC in the server via the Redfish protocol interface and the BMC interface circuit.

[0060] In some other embodiments, in step S11, the real-time temperature of a specified object can be obtained from the BMC in the server via the gateway module. The gateway module has at least a first data interface supporting the Redfish protocol and / or a second data interface supporting the IPMI protocol. The first and / or second data interfaces are communicatively connected to the BMC in the server to receive the real-time temperature of the specified object within the server. In this case, the main controller used to execute the control method is a Programmable Logic Controller (PLC) main controller.

[0061] The gateway module may also include a protocol conversion module and a data output unit. The protocol conversion module converts the input real-time temperature into a standard format. Specific conversion schemes for the protocol conversion module are described below. The data output unit outputs the real-time temperature in a standardized format under a specified protocol. The specified protocol is an industrial protocol supported by the main controller, such as the Modbus serial communication protocol or the Transmission Control Protocol (TCP). The protocol conversion module connects to the first data interface and the second data interface to receive the real-time temperature of the specified object. The data output unit is adapted to and electrically connected to the first data input terminal of the main controller. The data output unit can be an Ethernet interface, a wireless communication interface, an industrial bus interface, etc.

[0062] The PLC main controller includes a communication module, an analog input module, a central processing unit, an analog output module, and a power supply module. Using the PLC main controller in conjunction with other components of the immersion liquid cooling heat exchange system allows for the construction of an industrial-grade architecture, offering high reliability and simple maintenance.

[0063] The communication module is configured as the first data input terminal of the PLC main controller. The communication module can be a wireless communication module or a wired communication module, and it only needs to be compatible with the communication interface of the gateway module.

[0064] The analog input module includes a second data input terminal, a data input terminal for electrically connecting to a flow sensor, or a data input terminal for a speed sensor, etc. The analog input module can be electrically connected to a data acquisition device through one or more of its data input terminals, such as electrically connecting to a first temperature sensor and a second temperature sensor, to receive the inlet and outlet temperatures as described later. The analog input module can also be electrically connected to a speed sensor to obtain the rotational speed of the flow rate adjustment device, etc.

[0065] The central processing unit (CPU) has a logic processing circuit board. The input ports of the logic processing circuit board are electrically connected to the communication module and the analog input module to receive the real-time temperature of a specified object via the communication module, and to receive the inlet and outlet liquid temperatures via the analog input modules. The logic processing circuit board outputs control commands. It is used to execute control programs, perform data calculations, and make logical judgments. The control programs, data calculations, and logical judgments executed by the logic processing circuit board can be pre-configured and stored or programmed.

[0066] The analog output module is electrically connected to the central processing unit to receive control commands, and the output terminal of the analog output module is electrically connected to the actuator to output control commands.

[0067] The power supply module is electrically connected to the communication module, analog input module, central processing unit, and analog output module to provide power to these modules.

[0068] The control instructions mentioned later can be obtained from a logic processing circuit board.

[0069] In practical implementation, since the real-time temperature data acquired by different protocols has different formats, the acquired temperature data can be converted into a specified format according to preset format conversion rules. These conversion rules between different data formats are pre-configured. In actual use, the corresponding conversion rules can be obtained based on the acquired real-time temperature data format, and the conversion can be performed according to the preset rules. The real-time temperature data format conversion can be performed by the protocol conversion module mentioned above.

[0070] Based on the protocol identifier field of the real-time temperature data packet (such as the HTTP / JSON characteristics of the Redfish protocol and the RMCP session identifier of the IPMI protocol), the system can automatically identify the data source and data format type of the current real-time temperature and match the corresponding conversion rules. According to the conversion rules, the received real-time temperature data is converted into a standard format. The specific format type of the standard format can be preset according to actual needs. By converting to a standard format, the data unit and precision of real-time temperature can be standardized. For example, all temperature data can be uniformly converted to a Celsius (°C) format with two decimal places.

[0071] Furthermore, data from different protocols is mapped to a unified data structure that the heat exchange system can directly read. For example, the nested Temperatures[O].ReadingCelsius field in Redfish and the flat sensor data in IPM1 are uniformly converted into a structured data format containing sensor ID, temperature value, data acquisition timestamp, and server ID. This eliminates compatibility issues caused by protocol differences.

[0072] The following examples illustrate the conversion process: The real-time temperature of the CPU in a certain server is collected based on the Redfish protocol. The raw Redfish data collected is as follows: {"Temperatures":[{"Name":"CPUo_Temp","ReadingCelsius":66,"Status":"OK"}]}.

[0073] The real-time temperature of the CPU in this server is collected via the IPMI protocol. The raw IPMI data collected is as follows: Sensor ID: CPUe (Oxθ1), Reading: Ox3D (61), Manufacturer offset: +5°℃.

[0074] After the Redfish raw data and IPMI raw data are converted using the above protocol, both are output as standardized temperature data that can be directly used by the heat exchange system. json ^ "sensor_id":"cpuo", Temperature: 66.00 "unit":"℃", "timestamp": "2026-05-27T10:30:00Z", "server_id":"Server1" Step S12: Based on the relationship between the acquired real-time temperature and the preset temperature safety threshold of the specified object, and in conjunction with the compensation conditions, determine the adjustment strategy for the current adjustment cycle.

[0075] When the compensation conditions are met, the adjustment strategy is determined to be a compensation adjustment strategy. The compensation conditions include a real-time temperature greater than or equal to a preset temperature safety threshold for the specified object. The compensation adjustment strategy is configured as follows: determine the compensation adjustment value, and obtain the final adjustment value based on the compensation adjustment value and the base adjustment value. The base adjustment value is determined using a base temperature difference adjustment strategy, based at least on the temperature difference between the inlet and outlet temperatures of the first cooling circuit. The inlet temperature is the temperature of the coolant input to the cabinet from the first cooling circuit, and the outlet temperature is the temperature of the coolant output from the cabinet to the first cooling circuit.

[0076] The first cooling circuit includes an inlet pipe and a return pipe. The cabinet has an inlet and an outlet. The inlet pipe is connected to the outlet, supplying coolant to the cabinet via the inlet. The return pipe is connected to the outlet, receiving coolant discharged via the outlet. A coolant loop is formed between the first cooling circuit and the cabinet. A flow rate adjustment device is installed on the first cooling circuit.

[0077] The inlet liquid temperature can be collected by setting a first temperature sensor and the return liquid temperature by setting a second temperature sensor. The first temperature sensor is installed inside the inlet pipe or at the inlet of the cabinet. The second temperature sensor is installed inside the return liquid pipe or at the outlet of the cabinet. It is understood that thermocouples can also be used to collect the inlet or return liquid temperatures.

[0078] In some embodiments, when the compensation condition is that the real-time temperature is greater than or equal to a preset temperature safety threshold of the specified object, the compensation adjustment value can be determined as follows: Specifically, the adjustment increment is determined according to the first adjustment step size, and the compensation adjustment value is obtained by summing the compensation adjustment value of the previous adjustment cycle and the adjustment increment. It should be noted that if the current adjustment cycle is the first cycle, the compensation adjustment value of the previous adjustment cycle is zero. That is, in the first adjustment cycle, the adjustment increment is the compensation adjustment value.

[0079] In some non-limiting embodiments, the first adjustment step size is a preset fixed step size.

[0080] In some other non-limiting embodiments, the first adjustment step size is determined as follows: based on a first mapping relationship between the difference and the adjustment step size, the first adjustment step size is determined according to the difference between the real-time temperature and the preset temperature safety threshold of the specified object, where the difference is positively correlated with the first adjustment step size. That is, the larger the difference between the real-time temperature and the preset temperature safety threshold of the specified object, the larger the first adjustment step size, and consequently, the larger the compensation adjustment value obtained by summing the compensation adjustment value and the adjustment increase value of the previous adjustment cycle. The compensation adjustment value based on the basic adjustment value is also larger, resulting in a larger final adjustment value. Ultimately, by increasing the coolant flow rate of the first cooling return per unit time, the cooling capacity of the first cooling circuit can be improved, thereby rapidly reducing the temperature of the specified object within the server.

[0081] The initial mapping relationship between the difference and the adjustment step size can be pre-defined. This initial mapping relationship can be a linear positive correlation, a non-linear positive correlation, or a mapping data table composed of the difference and the adjustment step size, configured based on experimental data. The representation method of the initial mapping relationship between the difference and the adjustment step size is not limited here.

[0082] In practice, the type of parameters used for adjusting the step size varies depending on the type of flow rate adjustment device.

[0083] In some embodiments, when the flow rate adjustment device is an electronic pump, the parameter type used for the adjustment step size can be duty cycle. There is a linear relationship between the duty cycle and the speed of the electronic pump. By adjusting the duty cycle, the speed of the electronic pump can be adjusted, thereby adjusting the coolant delivery flow rate per unit time in the first cooling circuit. For example, the first adjustment step size can be a fixed step size, such as 5% duty cycle, 7% duty cycle, or 10% duty cycle, etc. Alternatively, the first adjustment step size can be obtained according to a first mapping relationship. The duty cycle size corresponding to the first adjustment step size is positively correlated with the difference value; different differences result in different duty cycles for the corresponding first adjustment step size. As a non-limiting example, the duty cycle value corresponding to the final adjustment value ranges from 0-100%, with a maximum value of 100% duty cycle. When the final adjustment value is 100% duty cycle, the speed of the electronic pump is at its maximum speed.

[0084] In other embodiments, when the flow rate adjustment device is a frequency pump, the parameter type used for the adjustment step size can be a frequency value. There is a linear relationship between the frequency value and the speed ratio of the frequency pump; by adjusting the frequency value, the speed of the frequency pump can be adjusted, thereby adjusting the coolant delivery flow rate per unit time in the first cooling circuit. For example, the first adjustment step size can be a fixed step size, such as 5% of the frequency value, 7% of the frequency value, or 8% of the frequency value, etc. Alternatively, the first adjustment step size can be obtained based on a first mapping relationship, and the magnitude of the frequency value corresponding to the first adjustment step size is positively correlated with the magnitude of the difference.

[0085] It should be noted that when there are multiple specified objects, if the real-time temperature of any one of the specified objects meets the compensation condition, then the compensation condition is considered met. If the real-time temperature of at least two of the specified objects meets the compensation condition, then the specified object with the highest priority is selected as the basis for judgment in subsequent steps.

[0086] For example, the priority of a specified object can be preset. This can be configured based on the impact of an object overheating on server performance. Objects with a higher impact on server performance due to overheating can be configured with a higher priority.

[0087] For example, the priority of a specified object can also be configured based on how much its real-time temperature exceeds a preset temperature safety threshold. The more its real-time temperature exceeds the preset temperature safety threshold, the higher its priority. In this way, specified objects with a higher degree of overheating can be cooled down first.

[0088] Step S13: Adjust the working state of the flow rate adjustment device by using the final adjustment value obtained by the adjustment strategy determined by the current adjustment cycle, so as to adjust the coolant delivery flow rate of the first cooling circuit per unit time.

[0089] In some non-limiting embodiments, the duration of the adjustment cycle can be longer than the duration of the temperature acquisition cycle. For example, the duration of the temperature acquisition cycle is 1 second, and the duration of the adjustment cycle can be 30 seconds. In this way, since the server needs a certain amount of time to be cooled by exchanging heat with the coolant, after adjusting the flow rate adjustment device, a relatively sufficient time can be reserved for cooling the server, avoiding frequent adjustments.

[0090] The above scheme involves acquiring the real-time temperature of a specified object within the server and determining the adjustment strategy for the current adjustment cycle based on the relationship between the acquired real-time temperature and the preset temperature safety threshold of the specified object, combined with compensation conditions. When the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object, i.e., when the compensation conditions are met, a compensation adjustment strategy is used to determine the compensation adjustment value. The final adjustment value is obtained based on the compensation adjustment value and the basic adjustment value. The basic adjustment value is determined using a basic temperature difference adjustment strategy, based at least on the temperature difference between the inlet temperature of the coolant input to the cabinet from the first cooling circuit and the return temperature of the coolant output from the cabinet. In other words, when the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object, the compensation adjustment strategy adds the compensation adjustment value to the basic adjustment value to obtain the final adjustment value. The final adjustment value is then used to adjust the operating state of the flow rate adjustment device to regulate the coolant delivery flow rate of the first cooling circuit per unit time. By acquiring the real-time temperature of a specified object within the server, a strong correlation is established between the real-time temperature of the specified object and the coolant flow rate per unit time in the first cooling loop. This breaks down the control silo between the immersion liquid cooling heat exchange system and the server, allowing the real-time temperature of the specified object to be considered when determining the coolant flow rate per unit time in the first cooling loop. The real-time temperature of the specified object serves as the basis for adjusting the heat exchange system, ensuring that the coolant flow rate per unit time in the first cooling loop matches the actual heat dissipation requirements of the server. This precise matching of the heat load of the specified object within the server and the cooling capacity provided by the first cooling loop enables timely heat dissipation of the server, reducing heat dissipation lag and meeting the timely heat dissipation needs during sudden changes in server power consumption. It also improves the accuracy of adjusting the coolant flow rate per unit time in the first cooling loop, preventing server overheating, frequency throttling, computing power degradation, and downtime, thus improving the server's heat dissipation effect and operational stability. This is particularly suitable for meeting the heat dissipation control requirements of next-generation high-density, high-computing-power servers.

[0091] By adjusting the coolant flow rate of the first cooling circuit per unit time based on the real-time temperature of a specified object within the server, the precision of the coolant flow rate adjustment can be improved, thereby enhancing the effectiveness of the adjustment. This avoids overcooling and ineffective adjustment, significantly reducing the energy consumption of the immersion liquid cooling heat exchange system, saving energy, and meeting the development requirements of green data centers.

[0092] After step S13 is executed, that is, after adjusting the working state of the flow rate adjustment device using the final adjustment value obtained by the adjustment strategy determined in the current adjustment cycle, the actual rotational speed of the flow rate adjustment device can be obtained. If the actual rotational speed is less than the first rotational speed threshold of the flow rate adjustment device, the real-time temperature of the specified object is obtained again. In the next adjustment cycle, based on the relationship between the real-time temperature obtained in the next adjustment cycle and the preset temperature safety threshold of the specified object, the adjustment strategy for the next adjustment cycle is determined, that is, step S11 is continued. The actual rotational speed of the flow rate adjustment device can be obtained through a rotational speed sensor.

[0093] When the compensation conditions are still met in the next adjustment cycle, that is, when the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object, an additional adjustment increase value can be added on the basis of the compensation adjustment value of the previous adjustment cycle, so as to gradually increase the final adjustment value, thereby gradually increasing the coolant delivery flow rate of the first cooling circuit per unit time, until the actual speed of the flow rate adjustment device is controlled by the final adjustment value to reach the upper limit of the speed, for example, the speed of the flow rate adjustment device reaches the second speed threshold.

[0094] In some embodiments, if the actual rotational speed is greater than or equal to a first rotational speed threshold of the flow rate adjustment device, an alarm is output. The first rotational speed threshold is less than or equal to the maximum rotational speed of the flow rate adjustment device. The alarm may include an audible and visual alarm or a text alarm. The audible and visual alarm can be configured as a local alarm, while the text alarm can be configured as a cloud alarm. Audible and visual alarms and text alarms can be used in combination. Setting an alarm when the actual rotational speed is greater than or equal to the first rotational speed threshold of the flow rate adjustment device helps to provide early warning before a thermal failure occurs in a specified object on the server. This allows for timely reminders to maintenance personnel to troubleshoot thermal failures, ensuring the normal operation of the server, reducing maintenance difficulty, and improving the intelligent management level of the system.

[0095] Furthermore, if the actual rotational speed is greater than or equal to the second rotational speed threshold, the final adjustment value is locked to the maximum adjustment value. This maximum adjustment value is used to adjust the flow rate adjustment device to rotate at the maximum speed; that is, the rotational speed of the flow rate adjustment device is locked to the maximum speed. The second rotational speed threshold is greater than or equal to the first rotational speed threshold. The maximum speed can be the maximum rated speed of the flow rate adjustment device, or it can be a speed lower than the maximum rated speed. This allows the flow rate adjustment device to operate at the maximum permissible speed under safe working conditions, ensuring that the first cooling circuit can provide maximum cooling capacity and ensuring effective cooling of the server.

[0096] When the second speed threshold is equal to the first speed threshold, when the actual speed of the flow rate adjustment device reaches the first speed threshold, an alarm can be triggered simultaneously and the final adjustment value can be locked to the maximum adjustment value.

[0097] In practice, the real-time temperature of a specified object within a server changes dynamically. To improve the cooling effect on the server and maintain the real-time temperature of the specified object within a suitable range, reducing temperature fluctuations, in some embodiments of this invention, the compensation condition may further include: the real-time temperature is less than a preset temperature safety threshold for the specified object, and the compensation adjustment value of the previous adjustment cycle is greater than zero. In this case, the compensation adjustment value can be determined as follows: determine the adjustment reduction value according to the second adjustment step size, and obtain the compensation adjustment value based on the difference between the compensation adjustment value and the adjustment reduction value of the previous adjustment cycle. In other words, the compensation adjustment value of the current adjustment cycle is the difference between the compensation adjustment value and the adjustment reduction value of the previous adjustment cycle. Thus, if the real-time temperature of the specified object is greater than or equal to the preset temperature safety threshold, that is, a compensation adjustment strategy has been implemented in a certain cycle. During subsequent adjustment cycles, even if the real-time temperature of the specified object is lower than the preset temperature safety threshold, but the compensation adjustment value of the previous adjustment cycle is not zero, the compensation adjustment value is gradually reduced by a determined reduction value. This gradually reduces the coolant flow rate of the first cooling circuit per unit time, i.e., gradually reduces the cooling capacity of the first cooling circuit, to better cope with temperature fluctuations of the specified object. This allows for smooth stabilization of the real-time temperature of the specified object within the server within a suitable temperature range, avoiding drastic temperature fluctuations and improving the operational stability of the immersion liquid cooling heat exchange system. The duration of the adjustment cycle when gradually reducing the compensation adjustment value can be shorter than the duration of the adjustment cycle when increasing the compensation adjustment value, to better cope with temperature fluctuations of the specified object, avoid drastic temperature fluctuations, and quickly reduce the compensation adjustment value to zero to save energy. For example, as a non-limiting example, the duration of the adjustment cycle when gradually reducing the compensation adjustment value is 10 seconds, and the duration of the adjustment cycle when gradually reducing the compensation adjustment value is 30 seconds. It is understandable that the duration of the adjustment period when the compensation adjustment value is gradually decreased can be equal to the duration of the adjustment period when the compensation adjustment value is increased.

[0098] In some embodiments, the second adjustment step size is a preset fixed step size.

[0099] In other embodiments, the second adjustment step size is determined as follows: based on a second mapping relationship between the absolute value difference and the adjustment step size, the second adjustment step size is determined according to the absolute value of the difference between the real-time temperature and the preset temperature safety threshold of the specified object. The absolute value difference and the second adjustment step size can be positively correlated. The larger the absolute value difference, the lower the real-time temperature of the specified object is compared to the preset temperature safety threshold. In this case, the cooling effect of the specified object is relatively good, and a relatively large second adjustment step size can be used to reduce the compensation adjustment value more quickly, so that the compensation adjustment value decreases to zero more quickly and smoothly. While taking into account the cooling effect on the server, energy consumption can be further saved. The smaller the absolute value difference, the closer the real-time temperature of the specified object is to the preset temperature safety threshold. In this case, the probability that the real-time temperature of the specified object exceeds the preset temperature safety threshold is greater. A relatively small second adjustment step size can be used to reduce the compensation adjustment value while also taking into account the cooling capacity of the first cooling loop, reducing the probability of the specified object's temperature rising and overheating, and ensuring the cooling effect on the server.

[0100] In some embodiments, when the relationship between the real-time temperature obtained in the current adjustment cycle and the preset temperature safety threshold of the specified object is used to determine that the compensation condition is not met (i.e., the real-time temperature is less than the preset temperature safety threshold of the specified object, and the compensation adjustment value of the previous adjustment cycle is zero), the adjustment strategy for the current adjustment cycle is determined to be a basic temperature difference adjustment strategy. The basic temperature difference adjustment strategy is configured as follows: based on the temperature difference, the temperature difference setpoint, and the setpoint deviation, a proportional-integral-derivative (PID) algorithm is used to determine the basic adjustment value, which is then used as the final adjustment value determined by the adjustment strategy for the current adjustment cycle. The setpoint deviation value is the allowable temperature difference to fluctuate above and below the temperature difference setpoint. For example, if the temperature difference setpoint is ΔT_set and the setpoint deviation value is ΔT_hys, then the allowable fluctuation range of the temperature difference is (ΔT_set - ΔT_hys, ΔT_set + ΔT_hys). The temperature difference is the temperature difference between the inlet temperature of the first cooling circuit input to the cabinet and the return temperature of the coolant output from the cabinet. In other words, if the real-time temperature of the specified object is within a suitable temperature range, the coolant delivery flow rate of the first cooling circuit per unit time can be adjusted based on the temperature difference, thus balancing the cooling effect on the server and the optimal energy efficiency requirements.

[0101] In some non-limiting embodiments, the basic adjustment value can be determined using a proportional-integral-derivative (PID) algorithm based on the temperature difference, temperature difference setpoint, and set deviation value as follows: In each adjustment cycle, the deviation value is calculated based on the temperature difference and temperature difference setpoint of the current adjustment cycle. The control increment for the current adjustment cycle is calculated based on the proportional coefficient, integral coefficient, derivative coefficient, the deviation value of the current adjustment cycle, and the deviation value of the previous adjustment cycle. The basic adjustment value for the current adjustment cycle is obtained based on the basic adjustment value of the previous adjustment cycle and the control increment for the current adjustment cycle.

[0102] For example, the deviation value of the kth period can be calculated using the following formula (1).

[0103] e(k) =ΔT_set-ΔT(k); (1)

[0104] Where k is the current adjustment cycle number, and k is a natural number greater than 1; e(k) is the deviation value of the k-th cycle; ΔT(k) is the temperature difference between the inlet and outlet temperatures collected in the k-th cycle; and ΔT_set is the temperature difference setpoint.

[0105] The incremental control quantity increment is calculated using the incremental PID algorithm and the following formula (2).

[0106] Δu(k)=K p ·[e(k)-e(k-1)]+K i ·e(k)+Kd·[e(k)-2e(k-1)+e(k-2)]; (2)

[0107] Where Δu(k) is the control increment; K p K is the proportionality coefficient. i Kd is the integral coefficient; Kd is the differential coefficient; e(k-1) is the deviation value of the (k-1)th period; e(k-2) is the deviation value of the (k-2)th period; for the first period, e(k-1) and e(k-2) take fixed values, such as zero. For the second period, e(k-1) is denoted as the deviation value of the first period, and e(k-2) can take fixed values, such as zero.

[0108] The basic adjustment value is calculated using the following formula (3).

[0109] u_base(k)=u_base(k-1)+Δu(k); (3)

[0110] Where u_base(k) is the base adjustment value for the k-th cycle; u_base(k-1) is the base adjustment value for the (k-1)-th cycle; and Δu(k) is the control increment. It should be noted that the base adjustment value for the first adjustment cycle can be set based on empirical values.

[0111] Furthermore, the basic adjustment value can be limited to keep it within a safe speed range. Simultaneously, combined with the set deviation value ΔT_hys, when the temperature difference is within the allowable fluctuation range of [ΔT_set-ΔT_hys, ΔT_set+ΔT_hys], the value of the control increment Δu(k) can be reduced to decrease the adjustment intensity of the PID algorithm, maintain the stability of the basic adjustment value, and avoid frequent pump adjustments. For example, K can be dynamically adjusted... p K i The value of the Kd coefficient makes the absolute value of Δu(k) output by formula (2) very small and tends to 0.

[0112] In practice, the determination of the basic adjustment value using the PID algorithm based on the temperature difference, temperature difference setpoint, and setpoint deviation can take many forms, each requiring a different method of determination. The following explains the methods for determining the basic adjustment value under different scenarios.

[0113] Scenario 1: If the temperature difference is greater than or equal to the first value, the basic adjustment value is increased based on the current basic adjustment value according to the PID algorithm. The first value is the sum of the temperature difference setpoint and the set deviation value. For example, if the temperature difference setpoint is ΔT_set and the set deviation value is ΔT_hys, then the first value is ΔT_set + ΔT_hys.

[0114] Furthermore, the increased baseline adjustment value is used as the final adjustment value determined by the adjustment strategy for the current adjustment cycle. The actual rotational speed of the flow rate adjustment device, determined based on the final adjustment value, is calculated. If the actual rotational speed is greater than or equal to the second rotational speed threshold, the flow rate adjustment device is forced to rotate at the maximum rotational speed corresponding to the maximum adjustment value to prevent overload of the flow rate adjustment device.

[0115] Scenario 2: If the temperature difference is less than or equal to the second value, the PID algorithm reduces the base adjustment value based on the current base adjustment value. The second value is the difference between the temperature difference setpoint and the set deviation value. For example, if the temperature difference setpoint is ΔT_set and the set deviation is ΔT_hys, then the second value is ΔT_set - ΔT_hys. For instance, PID control can be either forward or reverse, so the increment Δu(k) will be calculated as either a positive or negative number. When the calculated Δu(k) is negative, it achieves the effect of reducing the base adjustment value based on the current base adjustment value.

[0116] Furthermore, the reduced baseline adjustment value is used as the final adjustment value determined by the adjustment strategy for the current adjustment cycle. The actual rotational speed of the flow rate adjustment device, determined based on the final adjustment value, is calculated. If the actual rotational speed is less than or equal to the minimum rotational speed threshold, the flow rate adjustment device is forced to rotate at the speed corresponding to the minimum rotational speed threshold to ensure the minimum circulating flow rate of coolant in the first cooling circuit and maintain the cooling capacity of the first cooling circuit to cope with possible temperature rises in the specified object.

[0117] Scenario 3: If the temperature difference is greater than the second value but less than the first value, maintain the current base adjustment value, that is, maintain the current final adjustment value, to avoid frequent adjustments. For example, if the temperature difference setpoint is ΔT_set and the set deviation is ΔT_hys, then the temperature difference ΔT satisfies: ΔT_set - ΔT_hys < ΔT < ΔT_set + ΔT_hys. For example, PID control has forward and reverse regulation, so the increment Δu(k) will be calculated as a positive or negative number. When the calculated Δu(k) is negative, it can achieve the effect of reducing the base adjustment value based on the current base adjustment value.

[0118] To facilitate a better understanding and implementation of the solution in this application by those skilled in the art, the following is combined with... Figure 2 The control methods for immersion liquid cooling heat exchange systems are explained in detail for specific application scenarios: Step S201, system initialization.

[0119] After the immersion liquid-cooled heat exchange system unit starts normally, system initialization is performed. During system initialization, some control parameters can be loaded and fixed.

[0120] Control parameters may include one or more of the following: preset temperature safety thresholds for each specified object, such as the preset temperature safety threshold for the CPU and the preset temperature safety threshold for the GPU; the temperature acquisition cycle (i.e., the acquisition frequency) for the specified object; setting deviation values, temperature difference setpoints, maximum rotation speed of the flow rate adjustment device, minimum rotation speed of the flow rate adjustment device, first rotation speed threshold, second rotation speed threshold, first adjustment step size, second adjustment step size, duration of the adjustment cycle, control parameters of the basic temperature difference adjustment strategy, and alarm triggering conditions, etc.

[0121] Step S202: Collect the real-time temperature of the CPU, GPU, etc. in the server.

[0122] Among them, the real-time temperature T_chip of high-power chips such as CPU and GPU can be collected from the Baseboard Management Controller (BMC) of each server installed in the immersion liquid cooling system through a standardized industrial control protocol, at a high-frequency sampling frequency with a preset temperature acquisition cycle (such as ≤1 second).

[0123] For example, the real-time temperature of the CPU, GPU, etc., can be obtained from each server based on the Redfish protocol or IPMI protocol.

[0124] Step S203: Determine whether the real-time temperature is greater than or equal to the preset temperature safety threshold.

[0125] If the judgment result is yes, that is, the compensation condition is met, step S205 is executed; if the judgment result is no, step S204 is executed.

[0126] Step S205: Determine the adjustment increment value according to the first adjustment step size, and obtain the compensation adjustment value based on the sum of the compensation adjustment value and the adjustment increment value of the previous adjustment cycle.

[0127] Step S206: Obtain the final adjustment value based on the compensation adjustment value and the basic adjustment value.

[0128] Step S207: Adjust the working state of the flow rate adjustment device using the final adjustment value obtained by the adjustment strategy determined in the current adjustment cycle.

[0129] Step S208: Determine whether the actual rotation speed of the flow rate adjustment device is greater than or equal to the maximum rotation speed.

[0130] When the judgment result is yes, that is, the actual speed of the flow rate adjustment device is greater than or equal to the maximum speed, step S209 is executed; when the judgment result is no, step S202 is executed.

[0131] Step S209: Output an alarm notification.

[0132] Furthermore, when the determination result is yes, the actual rotational speed of the flow rate adjustment device can be locked to the maximum rotational speed. In this embodiment, both the first rotational speed threshold and the second rotational speed threshold are the maximum rotational speed.

[0133] Step S204: Determine whether the real-time temperature is lower than the preset temperature safety threshold and whether the compensation adjustment value is greater than zero.

[0134] When the judgment result is yes, that is, the real-time temperature is less than the preset temperature safety threshold and the compensation adjustment value is greater than zero, the compensation condition is also met, and step S218 is executed. When the judgment result is no, step S210 is executed.

[0135] Step S218: Determine the adjustment reduction value according to the second adjustment step size, and obtain the compensation adjustment value based on the difference between the compensation adjustment value and the adjustment reduction value of the previous adjustment cycle, until the compensation adjustment value is reduced to zero.

[0136] After step S218 is completed, step S206 is executed.

[0137] Step S210: Obtain the inlet and outlet temperatures of the first cooling circuit and execute the basic temperature difference adjustment strategy.

[0138] Step S211: Determine whether the temperature difference is greater than or equal to the first value.

[0139] If the judgment result is yes, proceed to step S212; if the judgment result is no, proceed to step S215.

[0140] The temperature difference value is the difference between the return temperature and the inlet temperature of the first cooling circuit. The first value is the sum of the temperature difference setpoint and the set deviation value.

[0141] Step S212: Increase the basic adjustment value according to the PID algorithm.

[0142] The final adjustment value is obtained based on the basic adjustment value, and a control command is generated based on the final adjustment value. The control command is used to control the rotation of the flow rate adjustment device.

[0143] Step S213: Determine whether the real-time rotation speed of the flow rate adjustment device is greater than or equal to the maximum rotation speed.

[0144] If the judgment result is yes, proceed to step S214. If the judgment result is no, continue to proceed to step S210.

[0145] Step S214: Maintain rotation speed.

[0146] Step S215: Determine whether the temperature difference value is less than or equal to the second value.

[0147] If the judgment result is yes, proceed to step S216; if the judgment result is no, proceed to step S214.

[0148] Step S216: Reduce the basic adjustment value according to the PID algorithm.

[0149] Step S217: Determine whether the real-time rotational speed of the flow rate adjustment device is less than or equal to the minimum rotational speed.

[0150] If the judgment result is yes, proceed to step S214; if the judgment result is no, proceed to step S210.

[0151] It should be noted that the specific rotational speed maintained in step S214 varies depending on the conditions that trigger its execution. When step S214 is executed after step S213, the flow rate adjustment device is maintained at its maximum rotational speed. When step S214 is executed after step S215, the flow rate adjustment device is maintained at its existing rotational speed. When step S214 is executed after step S217, the flow rate adjustment device is maintained at its minimum rotational speed.

[0152] Steps S210 and S202 can be executed independently. The basic adjustment value in step S206 can be obtained through step S210.

[0153] This application also provides a control device for an immersion liquid cooling heat exchange system. The immersion liquid cooling heat exchange system includes: a cabinet, a first cooling circuit, and a flow rate adjustment device. The first cooling circuit is in fluid communication with the cabinet and is used to supply coolant to the cabinet to cool the server immersed in the coolant in the cabinet.

[0154] Reference Figure 3 The control device 300 of the immersion liquid cooling heat exchange system includes: an acquisition unit 301, used to acquire the real-time temperature of a specified object within the server according to a set cycle; an adjustment strategy determination unit 302, used to determine the adjustment strategy for the current adjustment cycle based on the relationship between the acquired real-time temperature and a preset temperature safety threshold of the specified object, combined with compensation conditions; wherein, when the compensation conditions are met, the adjustment strategy is determined to be a compensation adjustment strategy, the compensation conditions including that the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object, the compensation adjustment strategy is configured to: determine a compensation adjustment value, and obtain a final adjustment value based on the compensation adjustment value and a basic adjustment value, the basic adjustment value being determined using a basic temperature difference adjustment strategy based at least on the temperature difference between the inlet temperature and the return temperature of the first cooling circuit, the inlet temperature being the temperature of the coolant input from the first cooling circuit to the cabinet, and the return temperature being the temperature of the coolant output from the cabinet to the first cooling circuit; and a processing unit 303, used to adjust the working state of the flow rate adjustment device using the final adjustment value obtained by the adjustment strategy determined in the current adjustment cycle, so as to adjust the coolant delivery flow rate of the first cooling circuit per unit time.

[0155] The control device 300 for the immersion liquid cooling heat exchange system can be used to implement the control method for the immersion liquid cooling heat exchange system described above, and has units or modules for implementing each step of the control method described above. For the specific working principle and workflow of the control device 300 for the immersion liquid cooling heat exchange system, please refer to the description in the above embodiments, which will not be repeated here.

[0156] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of any of the above-described control methods for an immersion liquid cooling heat exchange system.

[0157] This application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, performs the steps of any of the above-described control methods for an immersion liquid cooling heat exchange system.

[0158] The computer-readable storage medium may include non-volatile or non-transitory memory, and may also include optical discs, hard disk drives, solid-state drives, etc.

[0159] Specifically, in this embodiment of the invention, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0160] It should also be understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DR RAM).

[0161] This application also provides a terminal, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the steps of any of the above-described control methods for an immersion liquid cooling heat exchange system when running the computer program.

[0162] The memory and the processor are coupled, and the memory can be located inside or outside the terminal. The memory and the processor can be connected via a communication bus.

[0163] Terminals can include, but are not limited to, mobile phones, computers, tablets, and other terminal devices, as well as servers, cloud platforms, etc.

[0164] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means.

[0165] In the several embodiments provided in this application, it should be understood that the disclosed methods, apparatus, and systems can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for example, the division of units is merely a logical functional division, and other division methods may exist in actual implementation; for example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0166] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can be physically included separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or in a combination of hardware and software functional units. For example, for various devices or products applied to or integrated into a chip, each module / unit can be implemented using hardware such as circuits, or at least some modules / units can be implemented using software programs running on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware such as circuits; for various devices or products applied to or integrated into a chip module, each module / unit can be implemented using hardware such as circuits, and different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using hardware such as circuits. The components can be implemented using software programs that run on the processor integrated within the chip module. The remaining (if any) modules / units can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into the terminal, each of its components / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or in different components within the terminal. Alternatively, at least some modules / units can be implemented using software programs that run on the processor integrated within the terminal, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits.

[0167] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article indicates that the preceding and following related objects have an "or" relationship.

[0168] In the embodiments of this application, "multiple" refers to two or more.

[0169] The descriptions of "first," "second," etc., appearing in the embodiments of this application are for illustrative purposes and to distinguish the objects being described. They have no order and do not indicate any special limitation on the number of devices in the embodiments of this application, nor do they constitute any limitation on the embodiments of this application.

[0170] It should be noted that the sequence number of each step in this embodiment does not represent a limitation on the execution order of each step.

[0171] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in any computer-readable storage medium, which may include ROM, RAM, disk, or optical disk, etc.

[0172] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A control method for an immersion liquid-cooled heat exchange system, characterized in that, The immersion liquid cooling heat exchange system includes: a server rack, a first cooling circuit, and a flow rate adjustment device. The first cooling circuit is in fluid communication with the server rack and is used to supply coolant to the server rack to cool the server immersed in the coolant within the server rack. The control method of the immersion liquid cooling heat exchange system includes: Obtain the real-time temperature of a specified object within the server; Based on the relationship between the acquired real-time temperature and the preset temperature safety threshold of the specified object, and in conjunction with compensation conditions, the adjustment strategy for the current adjustment cycle is determined. Specifically, when the compensation conditions are met, the adjustment strategy is determined to be a compensation adjustment strategy. The compensation conditions include that the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object. The compensation adjustment strategy is configured to: determine a compensation adjustment value, and obtain a final adjustment value based on the compensation adjustment value and a basic adjustment value. The basic adjustment value is determined using a basic temperature difference adjustment strategy, based at least on the temperature difference between the inlet and outlet temperatures of the first cooling circuit. The inlet temperature is the temperature of the coolant input from the first cooling circuit to the cabinet, and the outlet temperature is the temperature of the coolant output from the cabinet to the first cooling circuit. The operating state of the flow rate adjustment device is adjusted by using the final adjustment value obtained by the adjustment strategy determined by the current adjustment cycle, so as to adjust the coolant delivery flow rate of the first cooling circuit per unit time.

2. The control method for the immersion liquid cooling heat exchange system as described in claim 1, characterized in that, The determination of the compensation adjustment value includes: The adjustment increment is determined according to the first adjustment step size, and the compensation adjustment value is obtained by summing the compensation adjustment value of the previous adjustment cycle and the adjustment increment.

3. The control method for the immersion liquid cooling heat exchange system as described in claim 2, characterized in that, The first adjustment step size is a preset fixed step size; Alternatively, the first adjustment step size can be determined as follows: based on the first mapping relationship between the difference and the adjustment step size, the first adjustment step size is determined according to the difference between the real-time temperature and the preset temperature safety threshold of the specified object, wherein the difference is positively correlated with the first adjustment step size.

4. The control method for the immersion liquid cooling heat exchange system as described in claim 1, characterized in that, After adjusting the operating state of the flow rate adjustment device using the final adjustment value obtained by the adjustment strategy determined by the current adjustment cycle, the process further includes: Obtain the actual rotational speed of the flow rate adjustment device; If the actual rotational speed is less than the first rotational speed threshold of the flow rate adjustment device, the real-time temperature of the specified object will continue to be acquired, and in the next adjustment cycle, the adjustment strategy for the next adjustment cycle will be determined based on the relationship between the acquired real-time temperature and the preset temperature safety threshold of the specified object.

5. The control method for the immersion liquid cooling heat exchange system as described in claim 4, characterized in that, Also includes: If the actual rotational speed is greater than or equal to the first rotational speed threshold of the flow rate adjustment device, an alarm reminder will be output. And / or, If the actual rotational speed is greater than or equal to the second rotational speed threshold, the final adjustment value is locked as the maximum adjustment value. The maximum adjustment value is used to control the flow rate adjustment device to rotate at the maximum rotational speed, and the second rotational speed threshold is greater than or equal to the first rotational speed threshold.

6. The control method for the immersion liquid cooling heat exchange system as described in claim 1, characterized in that, The compensation conditions further include: the real-time temperature is less than the preset temperature safety threshold of the specified object, and the compensation adjustment value of the previous adjustment cycle is greater than zero; determining the compensation adjustment value includes: determining the adjustment reduction value according to the second adjustment step size, and obtaining the compensation adjustment value based on the difference between the compensation adjustment value of the previous adjustment cycle and the adjustment reduction value.

7. The control method for the immersion liquid cooling heat exchange system as described in claim 6, characterized in that, The second adjustment step size is a preset fixed step size; Alternatively, the second adjustment step size can be determined as follows: based on the second mapping relationship between the absolute difference and the adjustment step size, the second adjustment step size is determined according to the absolute difference between the real-time temperature and the preset temperature safety threshold of the specified object.

8. The control method for the immersion liquid cooling heat exchange system as described in claim 1, characterized in that, The method for determining the adjustment strategy for the current adjustment cycle based on the relationship between the acquired real-time temperature and the preset temperature safety threshold of the specified object includes: If the real-time temperature is less than the preset temperature safety threshold of the specified object, and the compensation adjustment value of the previous adjustment cycle is zero, then the adjustment strategy for the current adjustment cycle is determined to be a basic temperature difference adjustment strategy. The basic temperature difference adjustment strategy is configured as follows: based on the temperature difference, the temperature difference setpoint, and the set deviation value, the basic adjustment value is determined using a PID algorithm, and the basic adjustment value is used as the final adjustment value determined by the adjustment strategy for the current adjustment cycle. The set deviation value is a value that allows the temperature difference to fluctuate around the temperature difference setpoint.

9. The control method for the immersion liquid cooling heat exchange system as described in claim 8, characterized in that, The step of determining the basic adjustment value using a PID algorithm based on the temperature difference, the temperature difference setpoint, and the set deviation value includes: If the temperature difference is greater than or equal to the first value, the basic adjustment value is increased based on the current basic adjustment value according to the PID algorithm, where the first value is the sum of the temperature difference setpoint and the set deviation value. If the temperature difference is less than or equal to the second value, the basic adjustment value is reduced based on the current basic adjustment value according to the PID algorithm, where the second value is the difference between the temperature difference setpoint and the set deviation value. If the temperature difference is greater than the second value and less than the first value, maintain the current base adjustment value.

10. The control method for the immersion liquid cooling heat exchange system as described in claim 1, characterized in that, The step of obtaining the real-time temperature of a specified object within the server includes any of the following: Based on the Redfish protocol, an HTTP GET request is sent to the baseboard management controller of the server. The HTTP GET request carries a temperature acquisition instruction, which is used to read the real-time temperature of a specified object in the server from the baseboard management controller of each server. The IPMI protocol is used to send an SDR command to the baseboard management controller of each server in the rack. The SDR command is used to query and obtain the real-time temperature data of a specified object in the server from the baseboard management controller.

11. A control device for an immersion liquid-cooled heat exchange system, characterized in that, The immersion liquid cooling heat exchange system includes: a server rack, a first cooling circuit, and a flow rate adjustment device. The first cooling circuit is in fluid communication with the server rack and is used to supply coolant to the server rack to cool the server immersed in the coolant within the server rack. The control device for the immersion liquid cooling heat exchange system includes: The acquisition unit is used to acquire the real-time temperature of a specified object within the server. An adjustment strategy determination unit is used to determine the adjustment strategy for the current adjustment cycle based on the relationship between the acquired real-time temperature and the preset temperature safety threshold of the specified object, combined with compensation conditions. When the compensation conditions are met, the adjustment strategy is determined to be a compensation adjustment strategy. The compensation conditions include that the real-time temperature is greater than or equal to the preset temperature safety threshold of the specified object. The compensation adjustment strategy is configured to: determine a compensation adjustment value, and obtain a final adjustment value based on the compensation adjustment value and a basic adjustment value. The basic adjustment value is determined using a basic temperature difference adjustment strategy, based at least on the temperature difference between the inlet and outlet temperatures of the first cooling circuit. The inlet temperature is the temperature of the coolant input from the first cooling circuit to the cabinet, and the outlet temperature is the temperature of the coolant output from the cabinet to the first cooling circuit. The processing unit is used to adjust the working state of the flow rate adjustment device by using the final adjustment value obtained by the adjustment strategy determined in the current adjustment cycle, so as to adjust the coolant delivery flow rate of the first cooling circuit per unit time.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program, when executed by the processor, performs the steps of the control method for the immersion liquid-cooled heat exchange system according to any one of claims 1 to 10.

13. A terminal comprising a memory and a processor, wherein the memory stores a computer program capable of running on the processor, characterized in that, When the processor runs the computer program, it performs the steps of the control method for the immersion liquid-cooled heat exchange system according to any one of claims 1 to 10.

14. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the control method for the immersion liquid-cooled heat exchange system according to any one of claims 1 to 10.

Images