A method and apparatus for regulating the distribution of cooling capacity for a liquid cooling system

Abstract

This application relates to a method and apparatus for regulating cooling capacity distribution in a liquid cooling system, relating to the field of computer technology. The method includes: acquiring electrical pulse test commands, internal flow channel volume, and real-time flow data; parsing and processing the electrical pulse test commands to obtain the predicted peak time of the heat load; generating a heat load time-series curve based on the predicted peak time of the heat load and a preset electrothermal conversion coefficient; performing time-delay correlation processing on the internal flow channel volume and real-time flow data to obtain the inherent heat transfer delay duration; generating a primary-side feedforward regulation command based on the heat load time-series curve and the inherent heat transfer delay duration; controlling the primary-side regulation mechanism based on the primary-side feedforward regulation command; generating a secondary-side transient drive command in response to the current moment reaching the predicted peak time of the heat load; and controlling the secondary-side circulating pump based on the secondary-side transient drive command. This application effectively eliminates temperature overshoot under pulsed heat loads and reduces redundant cooling energy consumption in the liquid cooling system.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method and apparatus for regulating the distribution of cooling capacity in a liquid cooling system. Background Technology

[0002] Liquid cooling systems are core equipment for thermal management in fields such as high-performance electronic device testing, aging of new energy power modules, and charging and discharging testing of power batteries.

[0003] In related technologies, the heat dissipation management of new energy electronic testing systems typically adopts a control method based on temperature sensor feedback: by real-time acquisition of the outlet water temperature on the secondary side of the liquid cooling circulation system and comparison with the set target process temperature, the proportional-integral-derivative algorithm is used to generate adjustment commands, thereby adjusting the opening degree of the chiller unit or the speed of the circulation pump to maintain the temperature stability of the testing environment.

[0004] Regarding the aforementioned technologies, traditional control systems exhibit water temperature control lag when facing instantaneous high-current pulse testing scenarios for new energy electronic devices (e.g., silicon carbide power modules). First, the rate of increase in system cooling supply lags behind the rate of increase in heat load, causing precision electronic devices to be in an overheated state during testing, resulting in irreversible thermal stress damage or even device burnout. Second, to forcibly offset this uncontrollable temperature fluctuation, some systems are forced to adopt a redundant cooling strategy of long-term high flow rate and low water temperature, resulting in serious energy waste and pump power loss, and also failing to eliminate the temperature spike problem under transient high pulses, indicating room for improvement. Summary of the Invention

[0005] In order to overcome the problems of water temperature control lag, temperature overshoot and energy waste caused by redundant cooling due to pulsed heat load in new energy electronic testing, this application provides a method for adjusting the cooling capacity distribution of a liquid cooling system.

[0006] In a first aspect, this application provides a method for adjusting the cooling capacity distribution of a liquid cooling system, employing the following technical solution: The system collects electrical pulse test commands from the testing equipment, as well as the internal flow channel volume and real-time flow rate data of the plate heat exchanger in the liquid cooling system. It then analyzes the electrical pulse test commands to obtain the predicted peak heat load time. Based on the predicted peak heat load time and a preset electrothermal conversion coefficient, it generates a heat load time-series curve. It performs time-delay correlation processing on the internal flow channel volume and real-time flow rate data to obtain the inherent heat transfer delay duration. Based on the heat load time-series curve and the inherent heat transfer delay duration, it generates a primary-side feedforward adjustment command for cold energy pre-storage. Based on the primary-side feedforward adjustment command, it controls the primary-side adjustment mechanism of the liquid cooling system. In response to the current moment reaching the predicted peak heat load time, it generates a secondary-side transient drive command to offset the transient heat load. Based on the secondary-side transient drive command, it controls the secondary-side circulation pump of the liquid cooling system.

[0007] Secondly, this application provides a cooling capacity distribution and regulation device for a liquid cooling system, employing the following technical solution: The acquisition module is used to acquire electrical pulse test commands, as well as the internal flow channel volume and real-time flow data of the plate heat exchanger in the liquid cooling system; the memory is used to store the program of the above-mentioned method for adjusting the cooling capacity of the liquid cooling system; the processor is used to load and execute the program in the memory and implement the above-mentioned method for adjusting the cooling capacity of the liquid cooling system.

[0008] The above-described embodiments of this disclosure have the following beneficial effects: the cooling capacity distribution and adjustment method for liquid cooling systems according to some embodiments of this disclosure effectively alleviates the temperature overshoot phenomenon under pulsed heat load and reduces the redundant cooling energy consumption of the liquid cooling system. Specifically, the reason for the water temperature control lag and temperature overshoot in liquid cooling systems in the related art when dealing with instantaneous heat pulses is that there is an inherent physical delay in the heat transfer process of the cooling medium in the pipeline and inside the plate heat exchanger. However, the traditional sensor feedback-based control mechanism does not consider this physical delay and only responds passively after the terminal temperature has deviated, resulting in the increase rate of system cooling capacity supply being slower than the rise rate of transient pulsed heat load. Based on this, the cooling capacity distribution and adjustment method for liquid cooling systems according to some embodiments of this disclosure first collects the electrical pulse test command issued by the test equipment, as well as the internal flow channel volume and real-time flow data of the plate heat exchanger in the liquid cooling system. Thus, by directly obtaining the underlying test command of the front-end test equipment and the physical flow channel parameters of the liquid cooling system, an objective data basis is provided for sensing the trend of transient heat load changes and quantifying the heat transfer delay of the system. Secondly, the aforementioned electrical pulse test commands are analyzed to obtain the predicted peak time of the heat load. This clarifies the specific timing of the impending transient high load (i.e., thermal shock), establishing a time reference for subsequent targeted processing (time axis offset and feedforward adjustment). Next, based on the predicted peak time of the heat load and the preset electrothermal conversion coefficient, a heat load time-series curve is generated. This realizes the conversion from electrical signal parameters to thermodynamic parameters, quantifying the dynamic change process of the heating power of the tested electronic device at different times, providing a calculation basis for subsequent matching of cooling demand. Furthermore, the aforementioned internal flow channel volume and real-time flow data are subjected to time delay correlation processing to obtain the inherent heat transfer delay time. This considers the objective influence of the heat exchanger body and fluid dynamic characteristics, quantifying the actual heat transfer lag time of the system, providing data support consistent with physical facts for setting a reasonable early intervention time. Then, based on the aforementioned heat load time-series curve and the aforementioned inherent heat transfer delay time, a primary-side feedforward adjustment command for cooling pre-storage is generated. Therefore, by integrating future heat load demand with the current physical delay of the system, a reasonable proactive intervention strategy can be derived, which helps to transform passive feedback response into proactive feedforward planning. Next, based on the aforementioned primary-side feedforward adjustment command, the primary-side adjustment mechanism of the liquid cooling system is controlled. This prompts the cold source side to act in advance, utilizing the cooling medium itself within the plate heat exchanger and its piping as a cold energy buffer for pre-cooling, thus building a cold energy reserve in physical space to resist thermal shock. Then, in response to the current moment reaching the predicted peak heat load, a secondary-side transient drive command is generated to offset the transient heat load. Thus, at the point where the heat pulse actually erupts in the test equipment, a drive signal is generated in a timely manner, ensuring that the timing of cold energy release and the timing of heat load generation are aligned on the time axis.Finally, based on the aforementioned transient drive commands on the secondary side, the secondary side circulation pump of the liquid cooling system is controlled. This ensures that the pre-stored cooling capacity is delivered promptly and directionally to the end of the heat-generating devices, achieving synchronous physical counterbalancing of transient heat loads. This effectively mitigates temperature overshoot in new energy electronic pulse testing scenarios, thereby ensuring the operational safety of precision testing devices. Simultaneously, by matching cooling capacity on demand, energy consumption caused by redundant cooling is reduced. Attached Figure Description

[0009] Figure 1 This is a flowchart of some embodiments of the cooling capacity distribution and adjustment method for a liquid cooling system according to the present disclosure. Detailed Implementation

[0010] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0011] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0012] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0013] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0014] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0015] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0016] refer to Figure 1 The diagram illustrates a flow 100 of some embodiments of a cooling capacity distribution and adjustment method for a liquid cooling system according to the present disclosure. The cooling capacity distribution and adjustment method for a liquid cooling system includes the following steps: Step 101: Collect the electrical pulse test command issued by the test equipment, as well as the internal flow channel volume and real-time flow data of the plate heat exchanger in the liquid cooling system.

[0017] In some embodiments, the executing entity (e.g., an electronic device) of the above-described method for adjusting the cooling capacity of a liquid cooling system can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is software, it can be installed in the hardware devices listed above. It can be implemented as multiple software programs or software modules to provide distributed services, or as a single software program or software module. No specific limitations are made here.

[0018] In some embodiments, the aforementioned executing entity can collect electrical pulse test commands issued by the test equipment, as well as the internal flow channel volume and real-time flow data of the plate heat exchanger in the liquid cooling system. The aforementioned electrical pulse test commands can be control signals characterizing the transient electrical load characteristics that the new energy electronic device (e.g., silicon carbide power module, insulated gate bipolar transistor, etc.) will withstand during testing, and may include, but are not limited to, test current amplitude, test voltage value, and pulse width data. The aforementioned liquid cooling system can be a cooling system deployed around the test bench to dissipate the instantaneous heat generated by high-power electrical testing, and may include a refrigeration unit (primary side equipment), a plate heat exchanger, a circulating water pump (secondary side equipment), and corresponding pipes and regulating valves. The aforementioned internal flow channel volume can be a parameter characterizing the physical space within the heat exchanger that allows the cooling medium to reside, and is related to the heat exchanger's model, number of plates, plate spacing, and corrugation depth, among other hardware structures. The aforementioned real-time flow data can be dynamic physical parameters characterizing the flow velocity and volume of the cooling medium within the liquid cooling system piping (flowing through the primary or secondary side piping of the plate heat exchanger) at the current moment.

[0019] In practice, a data communication link can be established between the aforementioned execution entity and the controller of the front-end testing equipment, the underlying sensors of the liquid cooling system, and the equipment configuration database to receive data information uploaded or sent by various data sources. The controller of the aforementioned testing equipment can encapsulate the test current, voltage, and pulse width parameters to be executed into data frames and send them to the aforementioned execution entity via an industrial communication bus (e.g., CAN bus, Profinet, or Modbus TCP protocol). The internal flow channel volume of the aforementioned plate heat exchanger can be used as a static equipment attribute parameter, which can be queried and read by calling a pre-established equipment ledger database or the API interface of the monitoring system, or entered as a fixed parameter by maintenance personnel during system initialization. The aforementioned real-time flow data can be measured in real time by flow monitoring devices (e.g., electromagnetic flow meters or ultrasonic flow meters) deployed on the inlet and outlet water pipes of the liquid cooling system. After being converted into standard electrical signals (e.g., 4-20mA analog signals or RS485 digital signals) by a transmitter, it is uploaded by a programmable logic controller (PLC). The aforementioned multi-source data can be received via wired industrial Ethernet or wireless network. After data parsing and integrity verification, target data that can be used for subsequent calculations can be obtained.

[0020] For example, firstly, an electrical pulse data packet can be obtained from the host computer of a silicon carbide power semiconductor test bench. After parsing, this data packet can display that the applied test current amplitude is 800A, the test voltage is 1200V, and the pulse width is 500ms, which is the electrical pulse test command. Secondly, by calling the local unit configuration database, the total internal flow channel volume of the primary and secondary sides of the currently operating plate heat exchanger can be found to be 15L, which is taken as the internal flow channel volume. Then, a high-precision electromagnetic flowmeter installed on the primary side inlet pipe of the heat exchanger can be used to collect the current pipe flow rate of the cooling medium in real time, which is 120. This refers to real-time traffic data.

[0021] Step 102: The electrical pulse test command is parsed and processed to obtain the predicted peak time of the heat load.

[0022] In some embodiments, the execution entity can parse and process the electrical pulse test command to obtain the predicted peak time of the thermal load. The predicted peak time of the thermal load can be a specific time point characterizing the maximum expected heat generation power (i.e., the highest point of thermal shock) of the new energy electronic device during electrical pulse testing.

[0023] In practice, the data packets of the collected electrical pulse test commands can be unpacked at the protocol level by calling the message parsing module. Specifically, firstly, the frame header, command type identifier, data segment, and timestamp in the data packet can be read according to the standard data frame format of the communication protocol (e.g., Profinet) that matches the front-end test equipment. Secondly, characteristic parameters directly related to the test timing are extracted from the data segment. For example, the absolute start time of the test trigger, the pulse delay time, the single pulse width, and the duty cycle of multiple pulses. Then, time-domain superposition is performed based on the above characteristic parameters. Typically, the peak heat power of electronic devices often occurs in the middle or late stage of electrical pulse loading or in a specific high-load phase. The executing entity can determine the relative delay deviation value of the heat load peak value compared to the trigger start time based on the system's preset correspondence between device types and standard test waveforms, and sum this relative delay deviation value with the trigger start time to obtain the predicted peak heat load time.

[0024] For example, consider the electrical pulse command data packet for a short-circuit withstand test (SC Test) of a certain type of Insulated Gate Bipolar Transistor (IGBT). Through protocol parsing, the planned trigger start time of this test is extracted as 14:30:00:000 milliseconds (i.e., system synchronization time), and the pulse width data is extracted as 10 microseconds. Then, based on preset empirical rules for short-circuit test heat generation, the peak transient heat generation power of this type of extreme test can be determined to occur at the instant before the pulse is turned off (e.g., set at the 8th microsecond of the pulse width). Finally, by superimposing the parsed parameters with the timing data (i.e., 14:30:00:000 milliseconds plus 8 microseconds), it can be concluded that this thermal shock will occur at 14:30:00:000 milliseconds 008 microseconds, which is the predicted peak heat load time.

[0025] Step 103: Generate a heat load time series curve based on the predicted peak heat load time and the preset electrothermal conversion coefficient.

[0026] In some embodiments, the aforementioned execution entity can generate a heat load time-series curve based on the predicted peak heat load time and a preset electrothermal conversion coefficient. The preset electrothermal conversion coefficient can be a physical parameter characterizing the proportion of input electrical energy converted into heat energy by the electronic device under test (DUT) under a specific operating state. This coefficient is typically determined by the physical material characteristics of the DUT (e.g., on-resistance, switching loss characteristics) and can be obtained by consulting the device's datasheet or a digital database. The aforementioned heat load time-series curve can be a dynamic waveform or data matrix depicting the continuous or discrete change of the heat generation power of the tested device over time throughout the entire pulse test cycle.

[0027] In practice, electrical commands can be converted into a thermodynamic energy distribution map based on the predicted peak heat load time (i.e., time base).

[0028] In some optional implementations of certain embodiments, the execution entity may generate a heat load time-series curve based on the predicted peak heat load time and a preset electrothermal conversion coefficient, which may include the following steps: The first step is to extract the test current amplitude, test voltage value, and pulse width data from the aforementioned electrical pulse test command. The test current amplitude and test voltage values ​​can be parameters characterizing the electrical intensity applied across the device during the pulse test. The pulse width data can be a timing parameter characterizing the duration of the high voltage or high current.

[0029] In practice, these three electrical parameters can be retrieved directly from the instruction buffer after protocol parsing by calling the data extraction interface.

[0030] The second step is to multiply the above test current amplitude and the above test voltage value to obtain the original value of electric power.

[0031] In practice, based on the built-in arithmetic logic unit (ALU), a multiplication operation can be performed on the extracted test current amplitude and test voltage value according to the basic electrical formula (i.e., power = voltage × current). The above raw power value can represent the total transient power input by the test bench to the device under test at the moment of the pulse.

[0032] The third step is to convert the original value of the electric power and the preset electrothermal conversion coefficient to obtain the heating power distribution value corresponding to each sampling point.

[0033] In practice, since not all the input electrical energy is converted into heat energy during the test, the original value of the electrical power can be multiplied by the preset electrothermal conversion coefficient according to a preset sampling frequency (e.g., sampling once every 1ms) to filter out the non-heat dissipated energy. Furthermore, the instantaneous heating power value actually converted into heat load at each discrete sampling point during the pulse duration can be determined.

[0034] The fourth step is to perform time-axis mapping processing on the heat power distribution values ​​corresponding to each sampling point according to the pulse width data mentioned above to obtain the heat load time-series curve.

[0035] In practice, a two-dimensional coordinate system can be established with time as the horizontal axis and heat generation power as the vertical axis. Then, using the predicted peak heat load time as the time anchor point in this coordinate system, the start and end points of the curve are determined based on the pulse width data. Finally, the heat generation power distribution values ​​corresponding to each sampling point calculated in the third step are filled into the coordinate system in chronological order. Through numerical interpolation or smoothing fitting algorithms, a complete heat load time-series curve representing the change of heat energy over time (e.g., increase or decrease) is generated.

[0036] For example, in conducting a power cycling test on a silicon carbide (SiC) module, the command parameters are extracted in the first step: the test current amplitude is 600A, the test voltage is 800V, and the pulse width is 50ms. In the second step, 600A and 800V can be multiplied to obtain the original power value of 480kW. In the third step, the database can be consulted to find that the preset electrothermal conversion coefficient for this type of SiC module in this test mode is 0.85 (meaning 85% of the power will be dissipated as heat), and the sampling interval is set to 1ms. Multiplying 480kW by 0.85 yields a plateau period heat distribution value of approximately 408kW. In the fourth step, the peak heat load time (e.g., 10:05:00.050) can be predicted as an anchor point, and these 50 discrete heat load points can be mapped onto a 50ms time axis window, fitting and generating a heat load time series curve exhibiting rectangular or trapezoidal wave characteristics. The above heat load time-series curves show that the system will withstand a transient thermal shock of up to 408kW within 50ms of pulse duration.

[0037] Step 104: Perform time delay correlation processing on the internal flow channel volume and real-time flow data to obtain the inherent heat transfer delay duration.

[0038] In some embodiments, the aforementioned execution entity may perform time-delay correlation processing on the aforementioned internal flow channel volume and the aforementioned real-time flow data to obtain the inherent heat transfer delay duration. The aforementioned inherent heat transfer delay duration may characterize the time required for cold energy to actually transfer from the cooling source side (e.g., the primary cooling medium input end of the heat exchanger), through physical space migration and heat conduction via the metal insulating material, and then act on the heat load side (e.g., the secondary water outlet end).

[0039] In some optional implementations of certain embodiments, the execution entity may perform time-delay correlation processing on the internal flow channel volume and the real-time flow data to obtain the inherent heat transfer delay duration, which may include the following steps: The first step is to perform a division operation between the aforementioned internal flow channel volume and the aforementioned real-time flow rate data to obtain the basic fluid migration time. This basic fluid migration time characterizes the time required for a specific volume of cooling fluid to completely flow through a designated internal space of the heat exchanger at the current flow rate.

[0040] In practice, to standardize physical dimensions, the collected internal flow channel volume and real-time flow rate data can first be standardized and converted. Then, the internal flow channel volume is divided by the real-time flow rate data using an arithmetic logic unit, and the quotient can be used as the basic fluid migration time. The basic fluid migration time can quantify the time delay caused by the mechanical propulsion and transport of cooling capacity using coolant as a carrier.

[0041] The second step involves collecting the plate thickness parameters and the thermal diffusivity of the plate material for the aforementioned plate heat exchanger. The plate thickness parameter can be a spatial geometric quantity characterizing the physical thickness of the metal plates separating the hot and cold sides inside the heat exchanger. The thermal diffusivity of the plate material (usually equal to the thermal conductivity divided by the product of density and specific heat capacity at constant pressure) can be a thermophysical property parameter characterizing the internal temperature homogenization capability and the rate of heat transfer and diffusion within the metal material.

[0042] In practice, these two static physical property parameters (plate thickness parameter and plate material thermal diffusivity) can be directly retrieved from the pre-configured system hardware asset database (equipment nameplate parameter table) through hardware communication interface or table lookup program.

[0043] The third step involves performing heat transfer characteristic mapping on the aforementioned plate thickness parameters and the thermal diffusivity of the plate material to obtain the thermal response time of the metal medium. This thermal response time of the metal medium can characterize the conduction time required for cold energy to penetrate the solid metal insulating layer.

[0044] In practice, according to the principle of unsteady heat conduction in a one-dimensional flat plate in heat transfer, the time it takes for heat to penetrate the plate is directly proportional to the square of the thickness and inversely proportional to the thermal diffusivity. The square of the extracted plate thickness parameter can be divided by the thermal diffusivity of the plate material and multiplied by a preset shape correction factor determined by the corrugated structure of the heat exchanger. This yields the time required for the cooling capacity to overcome the metal's heat capacity and establish a stable temperature gradient across the plate, i.e., the thermal response time of the metal medium.

[0045] The fourth step is to sum and accumulate the above-mentioned basic fluid migration time and the above-mentioned metal medium thermal response time to obtain the inherent heat transfer delay time.

[0046] In practice, the executing entity linearly sums the basic fluid migration time, representing macroscopic convection hysteresis, with the thermal response time of the metallic medium, representing microscopic thermal conduction hysteresis. The result is used as a sum physical quantity characterizing the overall hysteresis characteristics of the heat exchange process in the liquid cooling system, namely the inherent heat transfer delay, and is stored in a local register or memory matrix for subsequent steps.

[0047] For example, in the scenario of a computing power testing platform, when the liquid cooling system of a high-computing-power testing platform is in operation, firstly, the internal flow channel volume on one side of the plate heat exchanger can be obtained as 0.015 in the first step. The real-time flow data of the current pipeline was collected by the flow meter and found to be 0.005. Then, through division, the basic fluid migration time was determined to be 3 seconds. Subsequently, in the second step, by querying the equipment database, it was found that the heat exchanger uses 316L stainless steel plates with a thickness of 0.0006 μm and a thermal diffusivity of approximately 4.0 × 10⁻⁶ μm. -6 Next, in the third step, based on the unsteady-state heat conduction law, the square of the thickness (3.6 × 10⁻⁶) is... -7 Dividing by the thermal diffusivity, the thermal response time of the metal medium penetrating the plate is calculated to be approximately 0.09 s. Then, in the fourth step, the two are summed (3 s + 0.09 s) to obtain the inherent heat transfer delay time under this operating condition as 3.09 s. Through this step, the system defect of time lag can be transformed into a time parameter of 3.09 s, which can overcome the blindness of control lag in conventional control algorithms.

[0048] Step 105: Based on the heat load time sequence curve and the inherent heat transfer delay time, generate a primary side feedforward adjustment command for cold energy pre-storage.

[0049] In some embodiments, the aforementioned execution entity may generate a primary-side feedforward adjustment command for cold capacity pre-storage based on the aforementioned heat load timing curve and the aforementioned inherent heat transfer delay duration. This primary-side feedforward adjustment command for cold capacity pre-storage may be a machine-readable control message issued in advance to the cold source-side equipment of the liquid cooling system (e.g., a chiller compressor or a primary-side regulating valve) to increase the cooling capacity before the pulse heat load.

[0050] In some optional implementations of certain embodiments, the execution entity may generate a primary-side feedforward adjustment command for cold energy pre-storage based on the aforementioned heat load timeline and the inherent heat transfer delay duration, which may include the following steps: The first step involves performing integral calculations on the aforementioned heat load time-series curve based on a preset time window to obtain the total cooling demand for the transient thermal shock. The preset time window can be the start and end time period of a complete test pulse cycle. The total cooling demand can be the total energy consumed to offset the pulsed electrothermal heating, and the unit can be kJ.

[0051] In practice, numerical integration algorithms (such as the trapezoidal rule or Simpson's rule) can be used to calculate the area enclosed by the heat load time series curve and the time axis within a preset time window. This area can represent the accumulated heat energy, thus yielding the total cooling demand.

[0052] For example, with a preset time window size of 100ms, the heat load time series curve shows a trapezoidal wave with a peak power of 400kW during this time period. By performing a definite integral calculation on the time dimension of this time series curve, the cumulative heat generated within this time window is found to be 30kJ. Therefore, the total cooling demand for the transient thermal shock can be calculated as 30kJ.

[0053] The second step involves shifting the predicted peak heat load time forward on the time axis, based on the inherent heat transfer delay, to obtain the initial trigger time for primary-side pre-cooling. This time axis shifting is a mathematical operation involving subtraction over time. The initial trigger time can be the absolute time point at which the control system must initiate a cooling action to the primary-side equipment.

[0054] In practice, the predicted peak heat load time can be directly subtracted from the inherent heat transfer delay time to shift the time axis forward and obtain the initial trigger time.

[0055] For example, the predicted peak heat load time could be 14:00:05.000 (i.e., 2:00:05 PM). The inherent heat transfer delay could be 3.500 seconds. Subtracting 3.500 from 14:00:05.000 yields the initial trigger time for primary-side pre-cooling storage as 14:00:01.500. This indicates that a cooling command is issued 3.5 seconds before the thermal shock occurs, and the cooling capacity can reach the end of the test device at the peak time.

[0056] The third step involves calculating the cooling capacity gap based on the total cooling demand and the current heat capacity status data of the liquid cooling system, yielding the target cooling power on the primary side. The aforementioned heat capacity status data can be a set of parameters characterizing the cooling potential already stored within the system and readily available for use. This cooling capacity gap calculation process involves deducting the system's inherent buffer capacity to calculate the net cooling demand. The target cooling power on the primary side is the benchmark for the actual cooling capacity output of the primary side chiller unit during the pre-storage phase.

[0057] In practice, the total cooling demand can be compared with and subtracted from the system's available cooling capacity to obtain the basic cooling capacity gap. Then, by dividing the above basic cooling capacity gap by the allowable pre-storage time, the energy and power dimensions are converted to obtain the primary side target cooling power.

[0058] For example, the total cooling demand could be 30 kJ, but the current residual heat capacity data within the system indicates that the chilled water in the system's piping can only neutralize 10 kJ of heat. Therefore, through cooling capacity deficit calculation, it is determined that 20 kJ of heat remains undistributed. If the required cooling capacity is to be utilized within the remaining 2 seconds of pre-storage time, the calculated target cooling power on the primary side is 10 kW.

[0059] The fourth step involves encapsulating the aforementioned initial trigger time and primary-side target cooling power into data to generate a primary-side feedforward adjustment command. This data encapsulation process converts theoretical physical parameters into a communication protocol frame sequence that the underlying hardware can recognize and execute.

[0060] In practice, the calculated timestamp tag and cooling power setpoint can be packaged according to the message format of the Industrial Ethernet protocol to generate a hexadecimal machine code. This is the primary-side feedforward adjustment command.

[0061] In some optional implementations of certain embodiments, the execution entity may perform cooling capacity gap calculation on the total cooling capacity requirement and the current heat capacity state data of the liquid cooling system to obtain the primary side target cooling power, which may include the following steps: The first step is to collect the current supply and return water temperatures of the primary-side refrigerant, as well as the total water volume of the primary-side system piping. The current supply and return water temperatures can be the real-time temperatures of the refrigerant entering and exiting the plate heat exchanger. The total water volume can be the total physical volume of the liquid in the entire primary-side circulation loop, including the heat exchanger, pump, piping, and storage tank.

[0062] In practice, firstly, temperature data can be read in real time using platinum resistance (PT100) temperature sensors deployed on the supply and return water pipes to obtain the current supply and return water temperatures. Simultaneously, the total water capacity parameter can be read from the static asset database or configuration parameters of the liquid cooling system to obtain the total water capacity.

[0063] For example, the sensor detects that the current supply water temperature is 15℃ and the current return water temperature is 22℃. Checking the equipment's factory BOM (Bill of Materials), the total water capacity of the primary side system piping is 150L.

[0064] The second step involves converting the current supply water temperature, current return water temperature, and total water volume using specific heat capacity to obtain the baseline of the available cooling capacity of the liquid cooling system. This specific heat capacity conversion is performed using the thermodynamic formula (…). The process of calculating the sensible heat capacity difference of a fluid. The above-mentioned available cooling capacity baseline can be the maximum heat that the system can absorb under its current operating conditions, without relying on additional cooling from the compressor, solely from the chilled water inside the pipes.

[0065] In practice, you can first calculate the difference between the current return water temperature and the current supply water temperature, and then multiply it by the isobaric specific heat capacity parameter of the cooling medium (e.g., ethylene glycol aqueous solution) and the fluid mass (obtained from the total volume and density of the water) to get the base amount of available cooling capacity.

[0066] For example, the supply and return water temperature difference is 7℃ (22℃-15℃). Assuming the specific heat capacity of the cooling medium at constant pressure is 4.0 kJ / (kg·℃), the total water volume of 150L corresponds to a mass of approximately 150 kg. After multiplication and conversion (4.0 × 150 × 7), the current usable cooling capacity of the above liquid cooling system is obtained as 4200 kJ.

[0067] The third step is to calculate the difference between the total cooling demand and the available cooling capacity to obtain the basic cooling capacity gap. This basic cooling capacity gap can be assessed from an ideal thermodynamic perspective, representing excess heat that the system cannot absorb using its own water heat capacity.

[0068] In practice, the total cooling demand for transient thermal shock can be directly calculated by subtracting the current available cooling capacity of the liquid cooling system. If the result is less than or equal to 0, it indicates that the system has sufficient cooling capacity, with a deficit of 0. If the result is greater than 0, the difference is retained.

[0069] For example, the total cooling demand could be 15,000 kJ, while the available cooling capacity is 4,200 kJ. Subtracting 4,200 from 15,000 yields a basic cooling capacity deficit of 10,800 kJ.

[0070] The fourth step is to collect the heat loss coefficient corresponding to the current ambient temperature. This heat loss coefficient can be considered as a compensation constant that accounts for the portion of the cooling effect caused by the pipe insulation layer absorbing heat from the outside when the ambient temperature is higher than the chilled water temperature.

[0071] In practice, the ambient temperature of the outdoor environment or the area around the test bench can be obtained by deploying temperature and humidity sensors in the workshop, and a pre-stored environment and heat leakage mapping lookup table can be called to match and obtain the heat leakage loss coefficient corresponding to the ambient temperature.

[0072] For example, the sensor detects that the current ambient temperature in the test workshop is high, reaching 35°C. By using the environmental and heat leakage mapping lookup table, it can be found that under this high-temperature environment, the heat leakage loss coefficient of the primary side pipeline is 1.15 (that is, an additional 15% of cooling capacity is needed to offset the ambient heat leakage).

[0073] The fifth step involves amplifying and correcting the aforementioned basic cooling capacity gap based on the heat leakage loss coefficient, thereby obtaining the target cooling power on the primary side. This amplification and correction process involves calculating a safety margin compensation by combining the cooling capacity gap under ideal conditions with actual physical losses.

[0074] In practice, the basic cooling capacity shortfall can be multiplied by the heat loss coefficient to obtain the total cooling capacity that actually needs to be compensated, and then divided by the execution time allocated by the system to the pre-storage action (usually the heat exchanger response constant) to convert it into the target cooling power per unit time.

[0075] For example, if the basic cooling capacity deficit is 10800 kJ, multiplying it by the heat loss coefficient of 1.15 yields a total cooling capacity of 12420 kJ that actually needs to be compensated. If the system requires this cooling capacity to be replenished within 5 seconds, then 12420 kJ divided by 5 seconds gives the required cooling rate. This results in a primary-side target cooling power of 2484 kW.

[0076] In some optional implementations of certain embodiments, the execution entity may encapsulate the initial triggering time and the primary-side target cooling power into data to generate a primary-side feedforward adjustment command, which may include the following steps: The first step involves converting the primary-side target cooling power based on a pre-defined equipment performance mapping relationship to obtain the initial setpoint for the primary-side water supply temperature and the initial frequency of the primary-side water pump. The pre-defined equipment performance mapping relationship can be bench data of multi-dimensional energy efficiency curves relating the chiller compressor cooling capacity, water pump flow rate, and water supply temperature.

[0077] In practice, the calculated primary-side target cooling power can be input into the system's built-in hardware mapping lookup table or polynomial fitting function, and the combination of control variables that should be executed by the underlying hardware can be derived in reverse.

[0078] For example, the system needs to output a primary-side target cooling power of 2484kW. Consulting the equipment performance mapping table for the chiller unit, it is found that to achieve this instantaneous high cooling capacity output, the system must accelerate the water flow and significantly reduce the outlet water temperature. After conversion, the initial setpoint for the primary-side supply water temperature is 8℃, and the initial frequency of the primary-side water pump is 48Hz.

[0079] The second step involves comparing the initial setpoint of the primary water supply temperature with the real-time dew point temperature of the test environment to obtain the condensation risk assessment result. The real-time dew point temperature is the critical temperature parameter at which moisture in the air begins to condense into liquid water on the pipe surface under the current temperature and humidity conditions of the test workshop. This condensation risk assessment result serves as a logical indicator to evaluate whether the current low-temperature setting will cause a potential short-circuit hazard due to dripping water around the equipment.

[0080] In practice, the dew point temperature is calculated using ambient temperature and humidity data. Then, the real-time dew point temperature is subtracted from the initial set value of the primary water supply temperature. If the difference is less than the preset safety margin against condensation (e.g., 2°C), it is considered a risk. Otherwise, it is considered a no-risk condition.

[0081] For example, the initial setting of the primary water supply temperature is 8℃. The calculated dew point temperature in the workshop is 12℃. 8℃ is significantly lower than 12℃. After comparison, the condensation risk assessment result is "high risk (condensation risk exists)".

[0082] The third step involves comparing the initial frequency of the primary pump with the preset upper limit frequency for safe operation of the pump to obtain an overload determination result. The upper limit frequency for safe operation of the pump can be the maximum operating speed frequency set by the motor at the factory or permitted to prevent burnout (e.g., the inverter's upper limit is 50Hz). The overload determination result serves as a logical indicator to assess whether the speed control command exceeds physical limits.

[0083] For example, the initial frequency of the primary side water pump mentioned above could be 48Hz. Checking the inverter parameters, the upper limit frequency for safe operation of the water pump is 50Hz. 48Hz is less than 50Hz. After comparison, the overload judgment result is "no overload risk".

[0084] Fourthly, in response to the aforementioned condensation risk assessment result indicating the existence of condensation risk or the aforementioned overload assessment result indicating the existence of overload risk, a safety limit calculation is performed on the aforementioned initial setpoint of the primary water supply temperature and the aforementioned initial frequency of the primary water pump to obtain the corrected setpoint of the water supply temperature and the corrected pump frequency. The aforementioned safety limit calculation is a method of depriving the user of the authority to issue unreasonable commands and replacing them with safety thresholds.

[0085] In practice, when a risk of condensation is identified, the water supply temperature is forcibly increased to the value of "dew point temperature + safety margin". When a risk of overfrequency is identified, the frequency is forcibly truncated to the maximum upper limit frequency.

[0086] For example, if the water temperature of 8°C is deemed to pose a risk of condensation (dew point 12°C), the executing unit performs a safety limit calculation, forcibly adjusting the water temperature above the dew point (setting a safety margin of 12°C plus 1°C). This results in a corrected water supply temperature setting of 13°C. Since there is no risk of overload, the pump frequency does not need to be limited, and the corrected pump frequency remains unchanged at 48Hz.

[0087] The fifth step involves encapsulating the revised water supply temperature setpoint, the revised pump frequency, and the initial trigger time into a primary-side feedforward adjustment command. This data encapsulation process packages compliant service-level parameters into transmittable segments for the physical communication link.

[0088] In practice, temperature and frequency values ​​subject to safety constraints can be appended with execution time stamps, written into the data bits of a specific protocol instruction frame, and a checksum can be added to obtain a first-side feedforward adjustment instruction.

[0089] For example, the safe, corrected water supply temperature setpoint of 13°C, the corrected water pump frequency of 48Hz, and the predetermined start trigger time "14:00:01.500" can be packaged into a hexadecimal data frame stream (e.g., 0106 00 01 00 0D ...) using a preset protocol combination. This is the primary side feedforward adjustment command.

[0090] Step 106: Control the primary side adjustment mechanism of the liquid cooling system based on the primary side feedforward adjustment command.

[0091] In some embodiments, the aforementioned execution entity can control the primary-side adjustment mechanism of the liquid cooling system based on the aforementioned primary-side feedforward adjustment command. The primary-side adjustment mechanism can be physical execution hardware responsible for distributing the cooling medium on the primary side (i.e., the cold source side) in the liquid cooling system, and may include, but is not limited to: the variable frequency compressor of the chiller unit, the primary-side variable frequency circulating water pump, and the primary-side electric proportional regulating valve.

[0092] In practice, the timer inside the aforementioned execution unit compares the current system clock with the start trigger time in the primary-side feedforward regulation command in real time. Once the current system clock reaches the start trigger time, it sends the various control parameters in the command (e.g., the corrected supply water temperature setpoint and the corrected pump frequency) to the underlying controller of the primary-side regulation mechanism. After receiving and verifying the command, the underlying controller can convert it into specific electrical drive signals. Specifically, if the command requires a reduction in the supply water temperature, the controller will drive the variable frequency compressor of the chiller unit to increase its speed or adjust the opening of the electronic expansion valve to increase the instantaneous cooling capacity of the system. If the command requires an increase in flow rate, the controller will adjust the output frequency of the primary-side circulating water pump or increase the opening of the proportional control valve to accelerate the flow of the low-temperature cooling medium into the primary-side flow channel of the plate heat exchanger.

[0093] For example, when the system clock precisely reaches "14:00:01.500" (i.e., the initial trigger time), a primary-side feedforward adjustment command can be sent to the PLC control cabinet of the liquid cooling station. After parsing the command, the PLC outputs an analog control signal to drive the primary-side variable frequency water pump to accelerate to the command-set "48Hz", while simultaneously linking the chiller unit to set the primary-side water supply target temperature to "13℃". The 13℃ low-temperature chilled water is rushed into the plate heat exchanger at a high flow rate 3.5 seconds in advance (i.e., the inherent heat transfer delay time). In these 3.5 seconds before the actual arrival of the pulsed heat load, the internal flow channels and stainless steel plates of the heat exchanger are fully pre-cooled, completing the physical pre-storage of cold energy. When the test bench experiences a transient thermal shock at 14:00:05.000 (i.e., the predicted peak heat load time), the primary side of the liquid cooling system has already prepared the cold energy reserve.

[0094] Step 107: In response to the current time reaching the predicted peak heat load, generate a secondary transient drive command to offset the transient heat load.

[0095] In some embodiments, the aforementioned executing entity may generate a secondary-side transient drive command to offset the transient heat load in response to the current time reaching the predicted peak heat load. The current time may be the local real-time high-precision clock tick of the liquid cooling control system. The secondary-side transient drive command may be a device control message specifically generated for secondary-side equipment of the liquid cooling system (e.g., a secondary-side process circulation pump directly responsible for supplying liquid to the cold plates of the test bench). Unlike the primary-side feedforward regulation command, which differs in physical function, the secondary-side transient drive command, at the moment of heat burst in the heating device, changes the low-speed, stable state of the normal cycle, rapidly delivering the pre-stored cold energy within the plate heat exchanger to the terminal heating surface.

[0096] In practice, the current system clock and the predicted peak heat load can be compared in real time. When the comparison result indicates that the current time has reached the peak (or the time difference between the two falls within the system communication hysteresis compensation range), the hardware interrupt service routine inside the execution entity will be triggered immediately. The conventional closed-loop calculation logic based on the outlet water temperature deviation can be temporarily bypassed and replaced with peak data in the heat load time series curve. Using a preset heat balance mapping relationship (e.g., a linear correspondence function between power and flow rate), the instantaneous high flow rate parameter required to offset the thermal shock can be calculated. Subsequently, the above instantaneous high flow rate parameter is mapped to the target execution frequency of the secondary side circulation pump and packaged according to the corresponding industrial communication protocol (e.g., Profinet) to generate a secondary side transient drive instruction that can be directly parsed by the secondary side drive hardware.

[0097] For example, when the system clock precisely reaches "14:00:05.000" (i.e., the predicted peak heat load time), a low-level hardware interrupt is triggered instantaneously. This means the tested new energy electronic device is currently under electrical pulse loading, resulting in a high-amplitude transient thermal shock. To physically address the heat load during this transient thermal shock, the execution entity generates a dedicated control data frame for the secondary-side process circulation pump. This control data frame embeds control word parameters requiring the secondary-side pump to immediately exit its current 20Hz low-speed standby state and initiate a step switch to a high-frequency operating state, generating a secondary-side transient drive command to counteract the transient heat load.

[0098] Step 108: Control the secondary side circulation pump of the liquid cooling system based on the transient drive command on the secondary side.

[0099] In some embodiments, the aforementioned execution entity can control the secondary-side circulation pump of the liquid cooling system based on the aforementioned secondary-side transient drive command. The secondary-side circulation pump can be a physical power device in the liquid cooling system directly responsible for pumping the cryogenic working fluid cooled by the plate heat exchanger to the terminal heat-generating device (e.g., the water-cooled plate of the test bench).

[0100] In practice, when the current time reaches the predicted peak heat load, a transient drive command for the secondary side, equipped with a transient acceleration mechanism, can be sent to the frequency converter of the secondary side circulating pump, causing the cooling medium to accelerate its flow instantaneously in the secondary side piping. This action propels the fully cooled working medium, which has been trapped inside the heat exchanger, to the surface of the heat-generating device in a short time, thereby physically offsetting the thermal shock and effectively counteracting the transient temperature rise.

[0101] In some optional implementations of certain embodiments, the execution entity may control the secondary-side circulation pump of the liquid cooling system based on the aforementioned secondary-side transient drive command, which may include the following steps: The first step is to collect the current steady-state operating frequency value of the secondary circulation pump. This current steady-state operating frequency value can be the low pump operating frequency maintained by the liquid cooling system to maintain basic heat dissipation or standby state before receiving a pulsed thermal load impact.

[0102] In practice, the real-time operating status register inside the secondary circulating pump frequency converter can be read through the industrial bus to obtain the current frequency feedback parameters.

[0103] For example, before the pulse trigger, the liquid cooling system is in a basic standby cycle state. The actuator can read the current output frequency of the frequency converter through Modbus communication and obtain the current steady-state operating frequency value of the secondary side circulating pump as 20Hz.

[0104] The second step involves calculating the peak heat generation power in the aforementioned heat load time-series curve to obtain the target step frequency value used to offset the thermal shock. This target step frequency value can be a target parameter that the secondary-side circulating pump must instantly reach to meet the transient high-power heat dissipation requirements.

[0105] In practice, based on the thermodynamic heat exchange formula, combined with the specific heat capacity and maximum allowable temperature rise of the cooling working fluid, the peak heat generation power can be converted into the target flow rate requirement, and then the corresponding target frequency can be mapped through the flow rate frequency characteristic curve of the water pump.

[0106] For example, the heat load time-series curve shows that the peak heat output is about to reach 400 kW. Based on the thermodynamic heat exchange formula, it is found that 50 kW of heat is needed instantaneously to offset this 400 kW heat shock. The flow rate. Refer to the characteristic curve of this pump model, 50 The flow rate corresponds to an operating frequency of 45Hz. Therefore, the target step frequency value for offsetting thermal shock is determined to be 45Hz.

[0107] The third step involves reconstructing the target step frequency value and the current steady-state operating frequency value to generate a feedforward step frequency command, which serves as the secondary-side transient drive command. This command reconstruction process involves changing a conventional smooth speed control command into a sudden change command package with specific control identifiers.

[0108] In practice, the current steady-state operating frequency value (starting point) can be combined with the target step frequency value (end point), and a control word indicating skipping the normal acceleration slope can be added to the data frame to obtain the secondary transient drive command.

[0109] For example, the action logic can be "jump directly from 20Hz to 45Hz", and a control word representing the highest priority in the inverter protocol can be added to generate a hexadecimal communication message, which generates a feedforward step frequency command as a secondary transient drive command.

[0110] The fourth step involves controlling the secondary-side circulating pump based on the aforementioned feedforward step frequency command to disable the conventional proportional-integral (PI) limitation and to perform a step acceleration to the target step frequency value. Disabling the conventional PI limitation refers to the preset soft-start acceleration time within the bypass inverter (e.g., a default 10-second ramp-up time) to prevent delayed cooling capacity delivery due to slow pump acceleration. The step acceleration refers to the physical response process by which the pump motor reaches the target speed within a short time (e.g., 0.1 seconds).

[0111] In practice, after the instruction is issued, the secondary-side frequency converter will forcibly ignore the original slow acceleration smooth curve and directly drive the water pump impeller to accelerate instantly according to the maximum torque output current allowed by the motor.

[0112] In some optional implementations of certain embodiments, after the execution entity controls the secondary-side circulation pump of the liquid cooling system based on the aforementioned secondary-side transient drive command, the following steps may be included: The first step is to collect the actual return water temperature at the secondary end of the liquid cooling system. This actual return water temperature can be the fluid temperature of the cooling medium after it absorbs heat from the heat-generating device and returns to the liquid cooling circulation system.

[0113] In practice, a high-precision temperature sensor (e.g., a PT1000 RTD) installed on the return water main of the test bench can be used to continuously acquire physical temperature data after transient thermal shock at a high frequency sampling rate (e.g., once every 100ms).

[0114] For example, within a few seconds after the aforementioned feedforward counter-current action is completed, the sensor can collect the actual value of the current return water temperature, which is 32°C.

[0115] The second step involves performing deviation analysis on the actual return water temperature value based on a preset process operating temperature range to obtain the dynamic temperature deviation. The preset process operating temperature range can be a reference range for return water temperature specified to ensure the safety of the test equipment and the continuity of subsequent testing.

[0116] In practice, the actual value of the collected return water temperature can be compared with the target median value of the process range by subtraction to quantify the specific degree that exceeds or falls below the standard, thus obtaining the dynamic temperature deviation.

[0117] For example, the system's preset process operating temperature range is 28℃ to 30℃ (the target median is 29℃). Subtracting 29℃ from 32℃ and performing deviation analysis yields a dynamic temperature deviation of +3℃.

[0118] The third step involves performing proportional-integral (PI) control calculations on the aforementioned dynamic temperature deviation to obtain the transient recovery compensation coefficient. This PI calculation is a classic control algorithm that uses the proportional term (P) of the current deviation and the integral term (I) of the historical deviation to calculate the correction amplitude. The transient recovery compensation coefficient is used for fine-tuning the pump speed.

[0119] In practice, the dynamic temperature deviation can be input into the built-in PI algorithm module, which calculates the compensation force to eliminate this residual deviation based on preset gain parameters. The aforementioned PI algorithm module is a closed-loop control unit based on classical proportional-integral theory. Specifically, the PI algorithm module uses the real-time acquired dynamic temperature deviation as the input variable and synchronously introduces it into both the proportional and integral control loops: In the proportional loop, the current deviation is multiplied by a preset proportional gain constant to output an instantaneous adjustment force proportional to the current error, achieving a rapid response to transient temperature rises. In the integral loop, the deviations from each historical sampling period are accumulated or integrated and multiplied by a preset integral gain constant to output a continuous compensation force to eliminate the system's steady-state static error.

[0120] For example, regarding the aforementioned dynamic temperature deviation of +3℃, the PI algorithm module calculates that to smoothly eliminate this residual heat, the output needs to be increased by 15% from the current base speed. This results in a transient recovery compensation coefficient of +0.15. Here, the aforementioned output refers to the work capacity output by the motor of the secondary circulating pump.

[0121] The fourth step involves implementing closed-loop speed control of the secondary circulating pump's current operating frequency based on the aforementioned transient recovery compensation coefficient. This closed-loop speed control characterizes the system's smooth transition from feedforward open-loop control mode to feedback closed-loop fine-tuning mode after successfully handling extreme thermal pulses, thus maintaining long-term process stability.

[0122] In practice, the compensation coefficient can be applied to the frequency setting of the water pump to drive the water pump to gradually reduce its speed, thereby avoiding overcooling and bringing the temperature back to the process range.

[0123] For example, after the thermal shock subsides, the high-speed output of 45Hz is no longer needed. By incorporating a transient recovery compensation coefficient, the frequency of the secondary circulation pump is smoothly reduced from 45Hz and dynamically locked at 23Hz (i.e., a 15% increase based on the 20Hz fundamental frequency). This closed-loop speed control eliminates residual heat in the pipeline, allowing the water temperature to return to 29°C.

[0124] The above-described embodiments of this disclosure have the following beneficial effects: the cooling capacity distribution and adjustment method for liquid cooling systems according to some embodiments of this disclosure effectively alleviates the temperature overshoot phenomenon under pulsed heat load and reduces the redundant cooling energy consumption of the liquid cooling system. Specifically, the reason for the water temperature control lag and temperature overshoot in liquid cooling systems in the related art when dealing with instantaneous heat pulses is that there is an inherent physical delay in the heat transfer process of the cooling medium in the pipeline and inside the plate heat exchanger. However, the traditional sensor feedback-based control mechanism does not consider this physical delay and only responds passively after the terminal temperature has deviated, resulting in the increase rate of system cooling capacity supply being slower than the rise rate of transient pulsed heat load. Based on this, the cooling capacity distribution and adjustment method for liquid cooling systems according to some embodiments of this disclosure first collects the electrical pulse test command issued by the test equipment, as well as the internal flow channel volume and real-time flow data of the plate heat exchanger in the liquid cooling system. Thus, by directly obtaining the underlying test command of the front-end test equipment and the physical flow channel parameters of the liquid cooling system, an objective data basis is provided for sensing the trend of transient heat load changes and quantifying the heat transfer delay of the system. Secondly, the aforementioned electrical pulse test commands are analyzed to obtain the predicted peak time of the heat load. This clarifies the specific timing of the impending transient high load (i.e., thermal shock), establishing a time reference for subsequent targeted processing (time axis offset and feedforward adjustment). Next, based on the predicted peak time of the heat load and the preset electrothermal conversion coefficient, a heat load time-series curve is generated. This realizes the conversion from electrical signal parameters to thermodynamic parameters, quantifying the dynamic change process of the heating power of the tested electronic device at different times, providing a calculation basis for subsequent matching of cooling demand. Furthermore, the aforementioned internal flow channel volume and real-time flow data are subjected to time delay correlation processing to obtain the inherent heat transfer delay time. This considers the objective influence of the heat exchanger body and fluid dynamic characteristics, quantifying the actual heat transfer lag time of the system, providing data support consistent with physical facts for setting a reasonable early intervention time. Then, based on the aforementioned heat load time-series curve and the aforementioned inherent heat transfer delay time, a primary-side feedforward adjustment command for cooling pre-storage is generated. Therefore, by integrating future heat load demand with the current physical delay of the system, a reasonable proactive intervention strategy can be derived, which helps to transform passive feedback response into proactive feedforward planning. Next, based on the aforementioned primary-side feedforward adjustment command, the primary-side adjustment mechanism of the liquid cooling system is controlled. This prompts the cold source side to act in advance, utilizing the cooling medium itself within the plate heat exchanger and its piping as a cold energy buffer for pre-cooling, thus building a cold energy reserve in physical space to resist thermal shock. Then, in response to the current moment reaching the predicted peak heat load, a secondary-side transient drive command is generated to offset the transient heat load. Thus, at the point where the heat pulse actually erupts in the test equipment, a drive signal is generated in a timely manner, ensuring that the timing of cold energy release and the timing of heat load generation are aligned on the time axis.Finally, based on the aforementioned transient drive commands on the secondary side, the secondary side circulation pump of the liquid cooling system is controlled. This ensures that the pre-stored cooling capacity is delivered promptly and directionally to the end of the heat-generating devices, achieving synchronous physical counterbalancing of transient heat loads. This effectively mitigates temperature overshoot in new energy electronic pulse testing scenarios, thereby ensuring the operational safety of precision testing devices. Simultaneously, by matching cooling capacity on demand, energy consumption caused by redundant cooling is reduced.

[0125] Based on the same inventive concept, embodiments of this application provide a cooling capacity distribution and adjustment device for a liquid cooling system, comprising: The data acquisition module is used for electrical pulse test commands, as well as the internal flow channel volume and real-time flow data of the plate heat exchanger in the liquid cooling system; A memory for storing a program for adjusting the cooling capacity distribution of a liquid cooling system; The processor can load and execute programs in memory to implement a method for adjusting the distribution of cooling capacity in a liquid cooling system.

[0126] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0127] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as a method for adjusting the cooling capacity distribution of a liquid cooling system.

[0128] Computer storage media include, for example, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media that can store program code.

[0129] Based on the same inventive concept, embodiments of this application provide a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded and executed by the processor to provide a method for adjusting the cooling capacity distribution of a liquid cooling system.

[0130] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0131] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for adjusting the cooling capacity distribution in a liquid cooling system, characterized in that, include: The system collects electrical pulse test commands issued by the testing equipment, as well as the internal flow channel volume and real-time flow data of the plate heat exchanger in the liquid cooling system. The electrical pulse test command is parsed and processed to obtain the predicted peak time of the heat load; Based on the predicted peak heat load time and the preset electrothermal conversion coefficient, a heat load time series curve is generated; The inherent heat transfer delay time is obtained by performing time-delay correlation processing on the internal flow channel volume and the real-time flow data. Based on the heat load time series curve and the inherent heat transfer delay duration, a primary side feedforward adjustment command for cold energy pre-storage is generated. Based on the primary-side feedforward adjustment command, control the primary-side adjustment mechanism of the liquid cooling system; In response to the arrival of the predicted peak heat load at the current time, a secondary transient drive command is generated to offset the transient heat load; Based on the transient drive command on the secondary side, the secondary side circulation pump of the liquid cooling system is controlled.

2. The method for adjusting the cooling capacity distribution of a liquid cooling system according to claim 1, characterized in that, The step of generating a heat load time-series curve based on the predicted peak heat load time and the preset electrothermal conversion coefficient includes: Extract the test current amplitude, test voltage value, and pulse width data from the electrical pulse test command; The original value of electric power is obtained by multiplying the test current amplitude and the test voltage value. The original value of the electric power and the preset electrothermal conversion coefficient are converted to obtain the heating power distribution value corresponding to each sampling point; The heat load time-series curve is obtained by mapping the heat power distribution values ​​corresponding to each sampling point to the pulse width data.

3. The method for adjusting the cooling capacity distribution of a liquid cooling system according to claim 1, characterized in that, The step of performing time-delay correlation processing on the internal flow channel volume and the real-time flow data to obtain the inherent heat transfer delay time includes: The basic fluid migration time is obtained by performing a division operation between the internal flow channel volume and the real-time flow data. Collect the plate thickness parameters and the thermal diffusivity of the plate material of the plate heat exchanger; The thermal response time of the metal medium is obtained by performing heat transfer characteristic mapping on the plate thickness parameter and the thermal diffusivity of the plate material; The inherent heat transfer delay time is obtained by summing and accumulating the basic fluid migration time and the thermal response time of the metal medium.

4. The method for adjusting the cooling capacity distribution of a liquid cooling system according to claim 1, characterized in that, The step of generating a primary-side feedforward control command for cold energy pre-storage based on the heat load time-series curve and the inherent heat transfer delay time includes: Based on a preset time window, the heat load time series curve is integrally calculated to obtain the total cooling demand of the transient thermal shock. Based on the inherent heat transfer delay time, the predicted peak heat load time is shifted forward by the time axis to obtain the initial trigger time of primary side pre-storage. The total cooling demand and the current heat capacity status data of the liquid cooling system are processed to calculate the cooling gap, and the target cooling power on the primary side is obtained. The initial trigger time and the primary side target cooling power are encapsulated and processed to generate a primary side feedforward adjustment command.

5. The method for adjusting the cooling capacity distribution of a liquid cooling system according to claim 4, characterized in that, The step of performing cooling capacity gap calculation on the total cooling capacity demand and the current heat capacity state data of the liquid cooling system to obtain the target cooling power on the primary side includes: Collect the current supply water temperature and current return water temperature of the primary side refrigerant, as well as the total water volume of the primary side system pipeline; The specific heat capacity of the liquid cooling system is calculated by converting the current supply water temperature, the current return water temperature and the total water volume. The basic cooling capacity gap is obtained by subtracting the total cooling capacity requirement from the available cooling capacity baseline. Collect the heat loss coefficient corresponding to the current ambient temperature; Based on the heat loss coefficient, the basic cooling capacity gap is amplified and corrected to obtain the target cooling power on the primary side.

6. The method for adjusting the cooling capacity distribution of a liquid cooling system according to claim 4, characterized in that, The step of encapsulating and processing the initial triggering time and the primary side target cooling power to generate a primary side feedforward adjustment command includes: Based on the preset equipment performance mapping relationship, the target cooling power of the primary side is converted to obtain the initial set value of the primary side water supply temperature and the initial frequency of the primary side water pump. The initial set value of the primary water supply temperature is compared with the real-time dew point temperature of the test environment to obtain the condensation risk assessment result. The initial frequency of the primary water pump is compared with the preset upper limit frequency for safe operation of the water pump to obtain the overload determination result. In response to the condensation risk assessment result indicating the presence of condensation risk or the overload assessment result indicating the presence of overload risk, a safety limit calculation is performed on the initial set value of the primary side water supply temperature and the initial frequency of the primary side water pump to obtain the corrected water supply temperature set value and the corrected water pump frequency. The corrected water supply temperature setpoint, the corrected water pump frequency, and the start trigger time are encapsulated and processed to generate a primary side feedforward adjustment command.

7. The method for adjusting the cooling capacity distribution of a liquid cooling system according to claim 1, characterized in that, The step of controlling the secondary-side circulation pump of the liquid cooling system based on the secondary-side transient drive command includes: Collect the current steady-state operating frequency value of the secondary side circulation pump; The peak value of the heating power in the heat load time series curve is calculated to obtain the target step frequency value used to offset the thermal shock; The target step frequency value and the current steady-state operating frequency value are subjected to instruction reconstruction processing to generate a feedforward step frequency instruction, which is used as a secondary-side transient drive instruction. Based on the feedforward step frequency command, the secondary side circulation pump is controlled to shield the conventional proportional-integral limitation and to perform a step acceleration action to the target step frequency value.

8. The method for adjusting the cooling capacity distribution of a liquid cooling system according to claim 1, characterized in that, After controlling the secondary-side circulation pump of the liquid cooling system based on the secondary-side transient drive command, the method further includes: Collect the actual return water temperature at the secondary side terminal of the liquid cooling system; Based on the preset process operating temperature range, the actual value of the return water temperature is analyzed to obtain the dynamic temperature deviation. The dynamic temperature deviation is subjected to proportional-integral control calculation to obtain the transient recovery compensation coefficient; Based on the transient recovery compensation coefficient, closed-loop speed control is performed on the current operating frequency of the secondary side circulation pump.

9. A cooling capacity distribution and adjustment device for a liquid cooling system, characterized in that, include: The acquisition module is used to acquire electrical pulse test commands, as well as the internal flow channel volume and real-time flow data of the plate heat exchanger in the liquid cooling system; A memory for storing a program for adjusting the cooling capacity distribution of a liquid cooling system as described in any one of claims 1 to 8; The processor and the program in the memory can be loaded and executed by the processor to implement the cooling capacity distribution and adjustment method for a liquid cooling system as described in any one of claims 1 to 8.

Images