A method, system, medium and program product for anti-condensation control of a CDU train

Abstract

A kind of anti-condensation control method, system, medium and program product of CDU unit, relate to the field of shell or structural components of electric equipment.In the method, when receiving cabinet unlocking request, pre-intercept physical door opening action, and obtain external dew point temperature, internal heat load and pipeline surface temperature.When there is condensation risk, the target throttling opening is calculated by comprehensively considering heat load and anti-condensation temperature difference, and the larger value of the minimum opening degree of safety heat dissipation is compared as the actual execution opening degree, to give consideration to anti-condensation and equipment heat dissipation safety.Subsequently, control liquid inlet proportional valve to reduce flow operation and continuously monitor pipe temperature, when pipeline surface temperature rises to safe anti-condensation threshold or reaches preset timeout threshold, physical door opening action is released again.The present application is used to reduce the risk of water vapor accumulation of electric equipment shell and internal structural components during rapid fluctuation of ambient temperature and humidity, and improve the stability of electronic equipment physical structure in complex operation and maintenance scenarios.

Description

Technical Field

[0001] This application belongs to the field of enclosures or structural components of electrical equipment, and particularly relates to a method, system, medium and procedure for preventing condensation control of a CDU unit. Background Technology

[0002] With the widespread use of high-density electronic equipment, cooling distribution units (CDUs) are extensively deployed in electronic equipment cabinets for liquid cooling. When the coolant delivered by the CDU into the cabinet is at a low temperature, moisture in the air inside the cabinet can easily condense on the surfaces of the cooler physical components, thus reducing the safety of electronic equipment operation.

[0003] To address the aforementioned condensation issue, Chinese invention patent publication number CN112702886A, entitled "A Liquid Cooling System, Heat Dissipation Control Method, and Control Chip," discloses a liquid cooling system comprising a control unit, a heat source housing, and a heat dissipation unit. The heat source housing houses a temperature sensing unit connected to the control unit. The heat dissipation unit includes an inlet and an outlet; coolant flows into the heat dissipation unit through the inlet and exits through the outlet. A flow regulating valve is installed on the inlet pipe, connected to the control unit. The control unit acquires temperature information from the temperature sensing unit and determines the minimum temperature T1 and the maximum temperature T2 based on this information. The control unit controls the opening of the flow regulating valve. This liquid cooling system does not require pre-control of the coolant temperature and is suitable for cooling equipment cabinets even when the coolant temperature is low or uncontrollable, offering greater compatibility.

[0004] During rack inspections or equipment plug-and-play maintenance in data centers, highly humid air from outside the racks rapidly enters. In this scenario, while existing technologies can quickly detect humidity changes and instruct the CDU (Controlled Duct Unit) to increase the water supply temperature, the surface temperature of the metal casings and internal structural components of electronic equipment lags behind the rate of increase in the air dew point temperature due to their significant thermophysical mass and thermal inertia. This results in the surface temperature of these components remaining below the rapidly rising dew point temperature during the transition period of water supply temperature adjustment. Consequently, transient condensation occurs on the surfaces of the equipment casings and structural components, increasing the risk of moisture accumulation inside the electronic equipment. Summary of the Invention

[0005] This application provides a method, system, medium, and program product for preventing condensation in CDU units, which reduces the risk of moisture accumulation in the casing and internal structural components of electrical equipment during rapid fluctuations in ambient temperature and humidity, and improves the stability of the physical structure of electronic equipment in complex operation and maintenance scenarios.

[0006] In one aspect, this application provides a method for preventing condensation in a CDU unit, which receives an access control unlocking request instruction from a target cabinet and intercepts the opening of the physical door lock; In response to the access control unlocking request command, obtain the current external dew point temperature of the target cabinet's external environment, as well as the real-time heat load power of the heat-generating devices inside the target cabinet; Obtain the current surface temperature of the coolant piping inside the target cabinet; When the current surface temperature is lower than the current external dew point temperature, calculate the anti-condensation temperature difference between the current external dew point temperature and the current surface temperature; Based on the real-time heat load power and the anti-condensation temperature difference, the flow reduction control coefficient of the liquid inlet proportional valve is calculated. The flow reduction control coefficient represents the reduction ratio of the current coolant flow. The flow reduction control coefficient is negatively correlated with the real-time heat load power and positively correlated with the anti-condensation temperature difference. Calculate the target throttling opening based on the flow reduction control coefficient and the current valve opening of the inlet proportional valve; The larger of the target throttling opening and the preset minimum opening for safe heat dissipation of the equipment is taken as the actual opening. The liquid inlet proportional valve is controlled to operate according to the actual opening degree, and the current surface temperature is continuously monitored; When the current surface temperature is detected to reach the safety anti-condensation threshold, or the actual continuous operation time of the opening reaches the preset waiting timeout threshold, an opening execution command is sent to the physical door lock.

[0007] By employing the above technical solution, the opening of the physical door lock is blocked upon receiving a cabinet access unlocking request, thus preventing cold piping from being directly exposed to external humid and hot air before treatment. By acquiring the external dew point temperature, piping surface temperature, and the real-time heat load power of internal heat-generating devices, a flow reduction control coefficient that is negatively correlated with the heat load and positively correlated with the anti-condensation temperature difference is calculated, thereby dynamically adjusting the target throttling opening of the inlet proportional valve. This flow reduction strategy based on real-time operating conditions can proactively increase the piping surface temperature before the door is opened. Simultaneously, the target throttling opening is compared with the minimum safe heat dissipation opening of the equipment, and the larger value is taken as the actual execution opening. This ensures the basic heat dissipation requirements of the heat-generating devices while increasing the piping temperature, reducing the risk of overheating due to excessive flow restriction. Finally, the door opening is only allowed after the piping surface temperature reaches a safe threshold, reducing the probability of condensation on the piping surface at the moment the cabinet door is opened and improving the operational reliability of the electronic components inside the cabinet.

[0008] In conjunction with some implementation methods of the first aspect, in some implementation methods, the flow reduction control coefficient of the liquid inlet proportional valve is calculated based on the real-time heat load power and the anti-condensation temperature difference, specifically including: Input the real-time heat load power and the anti-condensation temperature difference into the preset flow reduction matrix to determine the basic flow reduction coefficient. The basic flow reduction coefficient is negatively correlated with the real-time heat load power and positively correlated with the anti-condensation temperature difference. Obtain the current liquid supply temperature delivered by the CDU unit to the target cabinet; Calculate the temperature deviation between the preset standard design water temperature and the current supply temperature; The temperature compensation factor is calculated based on the temperature deviation value. When the current liquid supply temperature is lower than the preset standard design water temperature, the temperature compensation factor is greater than 1. Multiplying the basic flow reduction coefficient by the temperature compensation factor yields the flow reduction control coefficient.

[0009] By adopting the above technical solution, a calculation mechanism combining a preset flow reduction matrix and a temperature compensation factor is introduced when calculating the flow reduction control coefficient. The temperature compensation factor is dynamically generated by obtaining the current actual supply liquid temperature of the CDU unit and calculating its temperature deviation from the preset standard design water temperature. When the actual supply liquid temperature is lower than the standard design water temperature, it means that the pipeline faces a higher risk of condensation. In this case, a temperature compensation factor greater than 1 will amplify the base flow reduction coefficient. This multi-dimensional adjustment strategy, which combines real-time heat load, anti-condensation temperature difference, and the actual temperature state of the supply water side, makes the flow reduction control of the inlet proportional valve more closely match the actual operating conditions, enhances the system's adaptability to cold source water temperature fluctuations, improves the accuracy of anti-condensation control, and helps to more quickly increase the pipeline surface temperature under low-temperature water supply conditions, thereby reducing the waiting time for maintenance personnel to open the valves.

[0010] In conjunction with some implementation methods of the first aspect, in some implementation methods, the larger of the target throttling opening and the preset minimum opening for safe heat dissipation of the equipment is used as the actual execution opening, specifically including: The highest real-time component temperature of the heat-generating devices inside the target cabinet is obtained through the Baseboard Management Controller (BMC) of the target cabinet. The difference between the preset hardware alarm temperature and the real-time highest component temperature is used as the real-time temperature margin. When the real-time temperature margin is greater than the preset safety margin threshold, the preset static minimum opening degree will be used as the preset minimum opening degree for safe heat dissipation of the equipment. When the real-time temperature margin is less than or equal to the preset safety margin threshold, the dynamic minimum opening degree that is inversely proportional to the real-time temperature margin is calculated, and the dynamic minimum opening degree is used as the preset minimum opening degree for safe heat dissipation of the equipment. The larger of the target throttling opening and the preset minimum opening for safe heat dissipation of the equipment is used as the actual opening.

[0011] By adopting the above technical solution, the real-time maximum component temperature of the internal heat-generating devices is obtained through the baseboard management controller of the target cabinet, and the real-time temperature margin between this temperature and the preset hardware alarm temperature is calculated, realizing a dynamic assessment of the equipment's heat dissipation safety boundary. When the temperature margin is sufficient, a static minimum opening is used to enhance the current reduction and anti-condensation effect; when the temperature margin is small or even faces the risk of overheating, a dynamic minimum opening that is inversely proportional to the temperature margin is introduced, so that the minimum safe heat dissipation opening can adaptively increase according to the equipment's heating state. This dynamic adjustment mechanism of the minimum opening based on real-time chip-level temperature feedback achieves a more refined dynamic balance between anti-condensation temperature rise control and safe heat dissipation of heat-generating devices, reducing the risk of hardware high-temperature alarms caused by local hot spots or sudden high loads during anti-condensation current reduction, and improving the unit's operational stability under complex operating conditions.

[0012] In conjunction with some implementations of the first aspect, in some implementations, after issuing an open command to the physical door lock, the method further includes: When the physical door lock is detected to switch from the open state to the closed state, the switch of the liquid inlet proportional valve to the normal heat dissipation control mode is executed. Obtain the current internal dew point temperature inside the target cabinet; Add the current internal dew point temperature to the preset anti-condensation safety margin to obtain the dynamic safety lower limit of pipe temperature; Calculate the upper limit of the throttling opening during the recovery period based on the lower limit of the dynamic safety pipe temperature and the real-time heat load power of the heat-generating devices inside the target cabinet; Control the liquid inlet proportional valve to operate within the range not exceeding the upper limit of the throttling opening during the recovery period, and continuously monitor the current internal dew point temperature; When the current internal dew point temperature is detected to drop to less than or equal to the preset standard liquid supply dew point threshold, the interception is lifted, and the liquid inlet proportional valve is controlled to return to the normal heat dissipation control mode.

[0013] By adopting the above technical solution, upon detecting that the physical door lock is closed, the immediate switch of the liquid inlet proportional valve to the conventional heat dissipation control mode is blocked, thus limiting the phenomenon of condensation caused by a sudden drop in pipe temperature due to a large influx of cold water and contact with the humid and hot air entering the cabinet. By acquiring the real-time dew point temperature inside the cabinet after the door is closed and adding a safety margin to generate a dynamic safe lower limit for the pipe temperature, and combining this with the real-time heat load power, the upper limit of the throttling opening during the recovery period is calculated. This mechanism reasonably constrains the upper limit of the coolant flow rate during the humidity recovery period after the door is closed, ensuring that the pipe surface temperature remains above the current internal dew point temperature. As the internal environment of the cabinet recovers, the blocking is only lifted when the internal dew point temperature gradually drops below the standard supply dew point threshold, reducing the risk of latent condensation in the initial period after maintenance and improving the safety of the cabinet's internal microenvironment recovery.

[0014] In conjunction with some implementation methods of the first aspect, in some implementation methods, the upper limit of the throttling opening during the recovery period is calculated based on the lower limit of the dynamic safety pipe temperature and the real-time heat load power of the heat-generating devices inside the target cabinet, specifically including: Obtain the current coolant supply temperature delivered from the CDU unit to the target cabinet; The difference between the dynamic safety lower limit of pipe temperature and the coolant supply temperature is used as the target heat exchange temperature difference; The maximum allowable coolant flow rate is calculated by dividing the real-time heat load power by the target heat exchange temperature difference and combining it with the specific heat capacity parameter of the coolant. Based on the inherent flow coefficient of the inlet proportional valve, the maximum allowable coolant flow rate is converted into the corresponding upper limit of the throttling opening during the recovery period.

[0015] By adopting the above technical solution, the target heat exchange temperature difference is obtained by subtracting the current coolant supply temperature to the target cabinet from the dynamic safety lower limit of the pipe temperature. Combined with real-time heat load power and the specific heat capacity of the coolant, the maximum allowable coolant flow rate is accurately calculated. Finally, the inherent flow coefficient of the inlet proportional valve is used to convert this into the upper limit of the throttling opening during the recovery period. This process transforms the abstract anti-condensation temperature boundary condition into specific valve control parameters, so that the setting of the flow upper limit no longer relies on rough empirical estimation, but is based on rigorous thermodynamic formulas for real-time quantitative calculation. This precise flow-limiting strategy based on actual heat exchange needs ensures that during the humidity recovery period after the cabinet is closed, the cooling capacity provided by the cooling system can meet the basic heat dissipation of the equipment while being strictly limited to a range that prevents the pipe temperature from falling below the safety lower limit. This improves the accuracy of the upper limit control of the inlet proportional valve opening during the cabinet's recovery period after the cabinet is closed and reduces the risk of condensation caused by excessively low pipe surface temperatures due to improper flow settings.

[0016] In conjunction with some implementations of the first aspect, in some implementations, the method further includes, prior to deblocking: The theoretical sensible heat rise rate of the heating device is calculated based on the real-time heat load power of the heating device inside the target cabinet, the actual coolant flow rate of the inlet proportional valve under the upper limit of the throttling opening during the recovery period, and the preset heat capacity parameters of the heating device. Obtain the current actual temperature rise rate of the heating element; The difference between the theoretical sensible heat rate and the actual temperature rate is taken as the phase change latent heat loss deviation. When the latent heat loss of phase change exceeds the preset microenvironment liquefaction residue threshold, it is determined that there is hidden condensation in the heat dissipation blind zone of the target cabinet in an evaporative heat absorption state. Maintain the interception state of the liquid inlet proportional valve and control the liquid inlet proportional valve to continue operating at the upper limit of the throttling opening during the recovery period until the phase change latent heat loss deviation is continuously monitored to decrease to less than or equal to the preset microenvironment liquefaction residue threshold, and then execute the step of releasing the interception.

[0017] By employing the above technical solution, the theoretical sensible heat rise rate is calculated based on real-time heat load power, actual coolant flow rate, and heat capacity parameters of the heating devices. This calculation is then compared with the obtained actual temperature rise rate to obtain the phase change latent heat loss deviation. When this deviation exceeds a preset microenvironment liquefaction residue threshold, the system can sensitively detect situations where the actual temperature rise is lower than expected, thus accurately determining the presence of latent condensation in the cabinet's heat dissipation blind zone, which is currently in an evaporative heat absorption state. Based on this determination, the system maintains the interception state of the inlet proportional valve and operates at the upper limit of the throttling opening during the recovery period, forcing the latent condensation to continue evaporating under controlled pipe temperature conditions until the phase change latent heat loss deviation drops to a safe range before releasing the interception. This improves the ability to identify and remove latent condensation that is difficult to directly observe in the complex internal structure of the cabinet, and reduces the risk of secondary condensation or short circuits in electronic components caused by residual moisture in the blind zone when the system prematurely resumes normal high-flow heat dissipation.

[0018] In conjunction with some implementations of the first aspect, in some implementations, the interception state of the liquid inlet proportional valve is maintained, and the liquid inlet proportional valve is controlled to continue operating at the upper limit of the throttling opening during the recovery period until the latent heat loss deviation of phase change is continuously monitored to decrease to less than or equal to the preset microenvironment liquefaction residue threshold, at which point the step of releasing the interception is executed, specifically including: While maintaining the interception state of the inlet proportional valve, based on the upper limit of the throttling opening during the recovery period, a micro-perturbation signal with a preset amplitude and preset frequency is superimposed to make the actual coolant flow rate exhibit periodic micro-oscillation. Calculate the theoretical temperature ripple amplitude of the heating device at a preset frequency based on the actual coolant flow rate of the periodic micro-oscillation; Extract the actual temperature ripple amplitude of the current actual temperature signal of the heating device at a preset frequency; Calculate the ripple attenuation rate of the actual temperature ripple amplitude relative to the theoretical temperature ripple amplitude; The net latent heat loss deviation is obtained by weighting the phase change latent heat loss deviation using the ripple attenuation rate. When the net latent heat deviation decreases to less than or equal to the preset microenvironment liquefaction residue threshold and the ripple attenuation rate approaches zero, the step of unblocking is executed.

[0019] By employing the above technical solution, while maintaining the interception state of the liquid inlet proportional valve, a micro-perturbation signal with specific amplitude and frequency is superimposed on the upper limit of the throttling opening during the recovery period, causing periodic micro-oscillations in the coolant flow rate. The actual temperature ripple amplitude of the heating element at a preset frequency is extracted and compared with the theoretical temperature ripple amplitude to calculate the ripple attenuation rate. This attenuation rate directly reflects the damping effect of the condensation evaporation phase change process on dynamic temperature fluctuations. The net latent heat deviation is obtained by weighting the latent heat loss deviation of the phase change using the ripple attenuation rate, effectively filtering out interference from sensor measurement errors or static thermal fluctuations in the environment. Only when the net latent heat deviation drops to a safe threshold and the ripple attenuation rate approaches zero is it confirmed that the latent condensation has completely evaporated, and the interception is lifted. This composite verification mechanism, which introduces active micro-perturbation and combines it with frequency domain feature analysis, improves the anti-interference capability and accuracy of determining the completion state of latent condensation evaporation, and reduces the risk of premature or delayed interception due to inaccurate determination of a single time-domain temperature deviation.

[0020] Secondly, embodiments of this application provide an anti-condensation control system for a CDU unit, the anti-condensation control system for the CDU unit comprising: one or more processors and a memory; the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and one or more processors call the computer instructions to cause the system to execute the anti-condensation control method for the CDU unit as described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a system, cause the system to perform the anti-condensation control method for a CDU unit as described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer program product that, when run on a system, causes the system to execute the anti-condensation control method for a CDU unit as described in any possible implementation of the first aspect.

[0023] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. This application provides an anti-condensation control method for CDU units. By intercepting the opening of the physical door lock upon receiving a cabinet door unlocking request, it restricts the cold piping from being directly exposed to external humid and hot air before treatment. By acquiring the external dew point temperature, pipe surface temperature, and real-time heat load power of internal heating devices, a flow reduction control coefficient that is negatively correlated with the heat load and positively correlated with the anti-condensation temperature difference is calculated, thereby dynamically adjusting the target throttling opening of the liquid inlet proportional valve. This flow reduction strategy based on real-time operating conditions can proactively increase the pipe surface temperature before the door is opened. At the same time, the target throttling opening is compared with the minimum safe heat dissipation opening of the equipment, and the larger value is taken as the actual execution opening. While increasing the pipe temperature, the basic heat dissipation requirements of the heating devices are ensured, reducing the risk of equipment overheating due to excessive flow restriction. Finally, the door opening is allowed only after the pipe surface temperature reaches the safe threshold, reducing the probability of condensation on the pipe surface at the moment the cabinet door is opened and improving the operational reliability of the electronic components inside the cabinet.

[0024] 2. This application provides a method for preventing condensation in CDU units. Upon detecting that the physical door is closed, the method intercepts the immediate switch of the inlet proportional valve to the conventional cooling control mode, limiting condensation caused by a sudden drop in pipe temperature due to a large influx of cold water and its contact with the intruding humid air. By acquiring the real-time dew point temperature inside the cabinet after the door is closed and adding a safety margin to generate a dynamic safety lower limit for the pipe temperature, and combining this with the real-time heat load power, the upper limit of the throttling opening during the recovery period is calculated. This mechanism provides a reasonable upper limit constraint on the coolant flow rate during the humidity recovery period after the door is closed, ensuring that the pipe surface temperature remains above the current internal dew point temperature. As the internal environment of the cabinet recovers, the interception is only lifted when the internal dew point temperature gradually drops below the standard supply dew point threshold, reducing the risk of latent condensation in the initial period after maintenance and improving the safety of the cabinet's internal microenvironment recovery.

[0025] 3. This application provides a method for preventing condensation in CDU units. The method calculates the theoretical sensible heat rise rate based on real-time heat load power, actual coolant flow rate, and heat capacity parameters of heating devices. This theoretical rate is then compared with the actual temperature rise rate to obtain the phase change latent heat loss deviation. When this deviation exceeds a preset microenvironment liquefaction residue threshold, the system can sensitively detect when the actual temperature rise is lower than expected, thus accurately determining the presence of latent condensation in the cabinet's heat dissipation blind zone, where evaporation and heat absorption are occurring. Based on this determination, the system maintains the inlet proportional valve in an intercepting state and operates at the upper limit of the throttling opening during the recovery period, forcing the latent condensation to continue evaporating under controlled pipe temperature conditions until the phase change latent heat loss deviation drops to a safe range before releasing the interception. This improves the ability to identify and remove latent condensation that is difficult to directly observe in the complex internal structure of the cabinet, reducing the risk of secondary condensation or short circuits in electronic components caused by residual moisture in the blind zone prematurely resuming normal high-flow heat dissipation. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating a method for preventing condensation in a CDU unit according to an embodiment of this application.

[0027] Figure 2 This is another schematic flowchart of an anti-condensation control method for a CDU unit in an embodiment of this application.

[0028] Figure 3 This is another flowchart illustrating a method for preventing condensation in a CDU unit as described in this application.

[0029] Figure 4 This is a schematic diagram of the physical device structure of an anti-condensation control system for a CDU unit provided in an embodiment of this application. Detailed Implementation

[0030] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0031] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0032] The following example is used in conjunction with Figure 1 The present application describes a method for preventing condensation in a CDU unit: Please see Figure 1 This is a flowchart illustrating an anti-condensation control method for a CDU unit in an embodiment of this application.

[0033] S101. Receive the access control unlocking request command from the target cabinet and intercept the opening of the physical door lock; The target rack refers to a physical cabinet in a data center or server room used to house electronic equipment, and is not limited to server racks, network communication racks, or storage array racks. An access control unlock request command refers to a trigger signal input by an external entity indicating an intention to open the rack door, and is not limited to RFID card swipe signals, keypad input confirmation signals, fingerprint recognition signals, or authorized unlock commands issued by a remote control center. A physical door lock refers to an electromechanical actuator installed on the target rack door to control the mechanical opening and closing of the door, and is not limited to electromagnetic locks, electronic mortise locks, or electromechanical integrated smart handle locks. When the system receives an opening request, it does not immediately execute the opening action, but instead intercepts the request and postpones the physical door lock action through control logic. This interception mechanism delays the actual opening time of the rack door, reducing the probability that the cold piping inside the rack is directly and suddenly exposed to potentially hot and humid external air, thereby reducing the risk of instantaneous condensation on the piping surface and improving the stability of the internal microenvironment of the rack.

[0034] The system can implement this step by combining hardware interrupts and relay control. Specifically, the general-purpose input / output pin of the system's main microprocessor is connected to the signal output terminal of the access control card reader or keypad, and this pin is configured to external interrupt trigger mode. When the access control device verifies the user's identity, it outputs a level transition signal as an access control unlock request command. After the main microprocessor captures this interrupt signal, it enters the interrupt service routine, in which a predefined access control request status flag is set to valid and the request is recorded. At the same time, the main microprocessor controls the control pin of the relay connected to the physical door lock's power supply circuit, forcing the output of a level signal that maintains the locked state. For example, a continuous low level output keeps the normally open relay open, thereby cutting off the drive power to the physical door lock, preventing the physical door lock from performing the opening mechanical action, thus completing the physical layer interception of the physical door lock's opening execution.

[0035] The system can also implement this step via a software gateway and application programming interface (API) calls. In this method, the access control system of the target rack is connected to the local area network of the computer room as an independent network node, and the system runs on the management server as a central control node. When a user initiates an unlocking operation at the front end of the target rack, the access control node sends an access unlocking request instruction data packet containing the rack number and unlocking timestamp to the system via the network. After receiving the data packet, the network monitoring service inside the system parses it and extracts the identification information of the target rack. Subsequently, the system's business logic processing module marks the rack's opening authorization status as pending in the memory database and deliberately does not return a confirmation data packet allowing opening to the access control node. Since the underlying lock control daemon process of the access control node is set to remain locked until it receives an explicit authorization instruction from the system, the physical lock's drive circuit will not be activated, thus intercepting the physical lock opening execution at the software protocol level.

[0036] S102. In response to the access control unlocking request command, obtain the current external dew point temperature of the target cabinet's external environment and the real-time heat load power of the heat-generating devices inside the target cabinet. The current external dew point temperature refers to the temperature at which water vapor in the ambient air outside the target cabinet condenses into dew, characterizing the humidity and condensation threshold of the external air. Internal heat-generating devices refer to electronic components within the target cabinet that generate heat during operation, not limited to the central processing unit, graphics processor, memory modules, or power supply modules. Real-time thermal load power refers to the total power consumed by these internal heat-generating devices and converted into heat energy at the current moment. After intercepting the door opening action, the system immediately collects the humidity status of the external environment and the heat generation status of the internal equipment. By acquiring this real-time data, an accurate data foundation is provided for subsequent anti-condensation control strategies, improving the system's ability to perceive the current environment and equipment operating conditions, and enhancing the targeting and accuracy of subsequent flow regulation strategies.

[0037] The system can achieve this step through multi-sensor fusion and out-of-band management polling. The system utilizes an integrated temperature and humidity sensor deployed outside the target rack, acquiring analog temperature and humidity signals from the external environment via an internal analog-to-digital converter and converting them into digital signals. After reading this digital signal, the system calls its built-in dew point temperature calculation algorithm to calculate the current external dew point temperature based on the current ambient temperature and relative humidity. Simultaneously, the system sends status query commands to the baseboard management controllers of each server within the target rack via its internal management network. The baseboard management controllers read the real-time output power sensor data from each power module on the server motherboard, aggregate it, and return it to the system via the network. Upon receiving the power data from all servers, the system performs a summation calculation to obtain the real-time thermal load power of the heat-generating devices inside the target rack.

[0038] The system can also implement this step by interfacing with the data center infrastructure management platform through an edge computing gateway. An intelligent environmental monitor outside the target rack collects environmental data in real time and directly calculates the current external dew point temperature in its internal edge computing module. This temperature value is then pushed to the data center infrastructure management platform via a message queue telemetry transmission protocol. Simultaneously, the intelligent power distribution sockets inside the rack monitor the output current and voltage of each socket in real time, calculate the real-time power consumption of each device, and also push this data to the management platform. The system sends a Hypertext Transfer Protocol request containing the target rack identifier by calling the Representational State Transition Application Programming Interface provided by the data center infrastructure management platform. Upon receiving the request, the management platform retrieves the latest current external dew point temperature and total power consumption data from its time-series database and returns it to the system in Extensible Markup Language (XML) format. The system can then parse this message to obtain the required real-time heat load power and dew point temperature.

[0039] S103. Obtain the current surface temperature of the coolant piping inside the target cabinet; Coolant piping refers to the piping system within the cooling capacity distribution unit or connected to the target cabinet for transporting coolant, and is not limited to the main inlet line, main return line, or distribution branches within the cabinet. Current surface temperature refers to the actual temperature value of the outer wall of the coolant piping at the current moment. After acquiring external environmental and internal heat load data, the system further acquires the actual temperature of the physical carriers where condensation may occur. By acquiring the actual surface temperature of the piping, the system can accurately determine the current hot and cold state of the piping, providing direct physical evidence for determining whether condensation will occur and improving the reliability of condensation risk assessment.

[0040] The system can implement this step using a contact temperature sensor and a serial communication bus. A platinum resistance temperature sensor is tightly mounted on the surface of a critical node in the coolant piping within the target cabinet using thermal grease and metal clamps. This platinum resistance temperature sensor is connected to a high-precision analog-to-digital converter (ADC) chip, which converts the minute resistance changes of the platinum resistance due to temperature variations into a digital temperature signal. The system's main control unit periodically sends read commands to the ADC chip via its internal integrated circuit bus. Upon receiving the command, the ADC chip transmits the converted digital temperature data back to the main control unit via the bus. The main control unit then linearizes the digital signal using a lookup table or polynomial fitting algorithm to obtain the accurate current surface temperature of the coolant piping.

[0041] The system can also achieve this step using a non-contact infrared temperature measurement module and a wireless sensor network. A miniature infrared thermopile sensor module is installed inside the target cabinet, directly opposite the surface of the coolant piping. This module senses temperature by receiving infrared radiation energy emitted from the piping surface and performs signal amplification and temperature compensation internally. This infrared temperature measurement module is connected to a Bluetooth Low Energy (BLE) microcontroller. The microcontroller periodically reads the temperature data output by the infrared module and encapsulates it into BLE broadcast data packets. The system's Bluetooth gateway continuously listens for these broadcast data packets. Once it receives a data packet from the infrared temperature measurement module, it extracts the temperature field and sends it to the system's central processing unit (CPU) via the local area network. The CPU parses this field to obtain the current surface temperature of the coolant piping.

[0042] S104. When the current surface temperature is lower than the current external dew point temperature, calculate the anti-condensation temperature difference between the current external dew point temperature and the current surface temperature. The anti-condensation temperature difference refers to the numerical difference between the current external dew point temperature and the current surface temperature. It is used to quantitatively represent how close the current pipe surface temperature is to the critical state where condensation will occur. After acquiring the pipe surface temperature, the system compares it with the external dew point temperature. The difference is only calculated when the pipe surface temperature is lower than the dew point temperature, i.e., when the physical conditions for condensation exist. By calculating this temperature difference, the system can intuitively assess the severity of the current condensation risk. The larger the temperature difference, the higher the risk. This provides a quantitative input parameter for subsequent dynamic adjustment of coolant flow rate, improving the sensitivity of the control strategy.

[0043] The system can implement this step through the microcontroller's internal arithmetic logic unit and register operations. The system's main microcontroller allocates two floating-point variable storage areas in memory to store the previously acquired current external dew point temperature and current surface temperature. The microcontroller executes a conditional judgment instruction according to a preset control cycle, comparing the current surface temperature with the current external dew point temperature. When the comparison result flag indicates that the current surface temperature is lower, the microcontroller calls the internal arithmetic logic unit to perform a subtraction operation, subtracting the current external dew point temperature as the minuend and the current surface temperature as the subtrahend. After the arithmetic logic unit completes the calculation, it stores the resulting difference in a specially named anti-condensation temperature difference register for subsequent control algorithms.

[0044] The system can also implement this step through programmable logic controller (PLC) and function block programming. The system uses a PLC as the core control device, calling standard comparison and mathematical operation function blocks within the controller's programming environment. The controller converts the analog values ​​of the current external dew point temperature and current surface temperature into engineering values, which are then input to the two input pins of the comparison function block. The comparison function block is configured with "less than" logic; when the pin value of the current surface temperature is less than the pin value of the current external dew point temperature, the output pin of the comparison function block generates a Boolean high-level signal. This high-level signal is connected to the enable pin of the subtraction mathematical operation function block, activating the subtraction function block. The subtraction function block then subtracts the current surface temperature from the input current external dew point temperature and outputs the calculated anti-condensation temperature difference value to the designated data storage block.

[0045] S105. Calculate the flow reduction control coefficient of the liquid inlet proportional valve based on the real-time heat load power and the anti-condensation temperature difference. The system calculates the flow reduction control coefficient of the inlet proportional valve based on the real-time heat load power and the anti-condensation temperature difference. The flow reduction control coefficient represents the reduction ratio of the current coolant flow. The flow reduction control coefficient is negatively correlated with the real-time heat load power and positively correlated with the anti-condensation temperature difference. This step can be achieved as follows: The system inputs the real-time heat load power and the anti-condensation temperature difference into a preset flow reduction base matrix to determine the base flow reduction coefficient. The base flow reduction coefficient is negatively correlated with the real-time heat load power and positively correlated with the anti-condensation temperature difference. The system obtains the current supply coolant temperature delivered by the CDU unit to the target cabinet. The system calculates the temperature deviation between the preset standard design water temperature and the current supply coolant temperature. The system calculates the temperature compensation factor based on the temperature deviation. When the current supply coolant temperature is lower than the preset standard design water temperature, the temperature compensation factor is greater than 1. The system multiplies the base flow reduction coefficient by the temperature compensation factor to obtain the flow reduction control coefficient.

[0046] First, let's explain the terms used in this step: A proportional inlet valve is a fluid control device installed on the coolant pipeline that continuously adjusts the valve's flow cross-sectional area based on a control signal. It is not limited to stepper motor-driven regulating valves or electromagnetic proportional regulating valves. The flow reduction control coefficient is a numerical value used to characterize the proportion of coolant flow reduction. The preset flow reduction baseline matrix is ​​a pre-built two-dimensional data table that maps the relationship between heat load, temperature difference, and the baseline flow reduction ratio. The current supply coolant temperature refers to the actual temperature of the coolant output from the cooling distribution unit to the target cabinet. The preset standard design water temperature refers to the rated supply water temperature specified in the design phase of the cooling system. The temperature deviation value is the difference between the standard water temperature and the actual supply water temperature. The temperature compensation factor is a multiplier calculated based on the temperature deviation value to correct the baseline flow reduction coefficient. The system comprehensively considers the equipment's heat generation, condensation risk, and actual supply water temperature, and derives the final flow reduction control coefficient through matrix lookup and compensation calculations. This multivariate comprehensive calculation improves the scientific nature of the flow reduction strategy, simultaneously increasing pipeline temperature and meeting the equipment's heat dissipation requirements.

[0047] The system can implement this step by combining a two-dimensional array lookup table with a piecewise linear interpolation algorithm. First, the system pre-stores a two-dimensional array in memory as a preset flow reduction matrix. The row indices of the array correspond to the discretized real-time heat load power range, the column indices correspond to the discretized anti-condensation temperature difference range, and the array elements are the basic flow reduction coefficients. The system uses the real-time heat load power and anti-condensation temperature difference as input coordinates, searches for four adjacent data points in the two-dimensional array, and uses a bilinear interpolation algorithm to calculate the accurate basic flow reduction coefficient, ensuring that this coefficient is negatively correlated with the heat load and positively correlated with the temperature difference. Next, the system obtains the current supply liquid temperature through a temperature sensor on the water supply pipeline and calls the subtraction operation module to subtract the current supply liquid temperature from the preset standard design water temperature to obtain the temperature deviation value. Subsequently, the system inputs this temperature deviation value into a preset piecewise linear function. When the temperature deviation value is greater than zero (i.e., the current supply liquid temperature is lower than the preset standard design water temperature), the function outputs a value greater than 1 as a temperature compensation factor. Finally, the system calls the multiplication module to multiply the basic flow reduction coefficient by the temperature compensation factor to obtain the final flow reduction control coefficient.

[0048] The system can also implement this step through relational database queries and proportional-integral (PI) calculations. The system stores a preset flow reduction matrix in a local relational database, constructing a data table containing upper and lower limits for heat load, upper and lower limits for temperature difference, and a basic flow reduction coefficient. The system constructs a structured query language statement, inputting real-time heat load power and anti-condensation temperature difference as query conditions into the database. The database engine matches records that meet the conditions and returns the corresponding basic flow reduction coefficient. The data patterns within this table ensure that the coefficient is negatively correlated with heat load and positively correlated with temperature difference. Simultaneously, the system obtains the current supply liquid temperature by reading the status register of the cooling capacity distribution unit controller and uses the microprocessor's arithmetic unit to calculate the temperature deviation between the preset standard design water temperature and the current supply liquid temperature. The system inputs this temperature deviation value into a proportional calculation logic module, sets a fixed proportional gain constant, multiplies the temperature deviation value by this gain constant, and adds a base value of 1 to calculate the temperature compensation factor. Logically, this ensures that the compensation factor is greater than 1 when the supply liquid temperature is low. Finally, the system executes a floating-point multiplication instruction to multiply the basic reduction coefficient by the temperature compensation factor to obtain the reduction control coefficient.

[0049] S106. Calculate the target throttling opening based on the flow reduction control coefficient and the current valve opening of the inlet proportional valve; The current valve opening refers to the open state of the inlet proportional valve before this adjustment, usually expressed as a percentage or the number of steps by the stepper motor. The target throttling opening refers to the new open state value that the inlet proportional valve is expected to reach after this adjustment. After calculating the flow reduction control coefficient, the system needs to convert it into an opening command that the inlet proportional valve can directly execute. By combining the current valve opening and the flow reduction control coefficient to calculate the target throttling opening, the system can achieve relative adjustment of the coolant flow rate, improve the smoothness of flow control, and reduce the risk of water hammer in the piping system caused by sudden changes in valve opening.

[0050] The system can implement this step through direct multiplication mapping and pulse width modulation duty cycle conversion. The system's main control chip reads the current valve opening percentage value stored in the inlet proportional valve driver. Subsequently, the main control chip calls its internal floating-point multiplication unit to directly multiply the current valve opening percentage by the flow reduction control coefficient calculated in step S105. Since the flow reduction control coefficient is a value representing the reduction ratio, the result after multiplication is the desired target throttling opening percentage. In order to control the valve, the main control chip inputs the target throttling opening percentage into a linear mapping function, converts it into the corresponding pulse width modulation signal duty cycle value, and stores this duty cycle value in the timer's compare register, ready to drive the motor of the inlet proportional valve.

[0051] The system can also achieve this step through flow characteristic curve inversion and stepper motor step count calculation. Since the opening degree of the proportional valve and the actual flow rate are often not strictly linear, the system pre-stores a flow characteristic curve data table for this type of proportional valve in memory. First, the system looks up the current relative flow rate value in the table based on the current valve opening (represented by the current stepper motor step count). Next, the system multiplies this relative flow rate value by the calculated flow reduction control coefficient to obtain the target relative flow rate value. Subsequently, the system uses this target relative flow rate value to perform a reverse lookup and interpolation calculation in the flow characteristic curve data table to obtain the target stepper motor step count corresponding to the target flow rate. This calculated target step count is used as the target throttling opening for subsequent precise control of the valve's mechanical position.

[0052] S107. The larger of the target throttling opening and the preset minimum opening for safe heat dissipation of the equipment shall be used as the actual opening. The system uses the larger of the target throttling opening and the preset minimum safe heat dissipation opening as the actual execution opening. This step can be achieved as follows: The system obtains the real-time maximum component temperature of the heat-generating devices inside the target cabinet via the Baseboard Management Controller (BMC); the difference between the preset hardware alarm temperature and the real-time maximum component temperature is used as the real-time temperature margin; when the real-time temperature margin is greater than the preset safety margin threshold, the preset static minimum safe heat dissipation opening is used as the preset minimum safe heat dissipation opening; when the real-time temperature margin is less than or equal to the preset safety margin threshold, a dynamic minimum safe heat dissipation opening that is inversely proportional to the real-time temperature margin is calculated, and this dynamic minimum safe heat dissipation opening is used as the preset minimum safe heat dissipation opening; the larger of the target throttling opening and the preset minimum safe heat dissipation opening is used as the actual execution opening.

[0053] First, let's explain the terms used in this step: Actual execution opening degree refers to the opening command value ultimately received and executed by the proportional valve. Preset minimum safe heat dissipation opening degree refers to the minimum degree of valve opening that must be maintained to ensure that heat-generating components within the rack are not damaged or have reduced frequency due to overheating. The Baseboard Management Controller (BMC) is an independent microcontroller on the server motherboard used to monitor hardware status. Real-time maximum component temperature refers to the highest current temperature value among all monitored components within the rack. Preset hardware alarm temperature is the upper limit of the safe operating temperature specified by the chip manufacturer. Real-time temperature margin is the difference between the alarm temperature and the maximum component temperature. Preset safety margin threshold is a system-set boundary value used to determine whether the temperature is within an absolutely safe range. Static minimum opening degree is a fixed minimum opening degree empirical value. Dynamic minimum opening degree is the minimum opening value calculated in real-time based on the temperature margin. The system must ensure the basic heat dissipation requirements of the equipment while limiting flow to prevent condensation. By using a strategy of comparing and selecting the larger value, system safety is improved, and the risk of thermal failure of core electronic components due to excessive pursuit of condensation prevention is reduced.

[0054] The system can implement this step using the intelligent platform management interface protocol and conditional branching logic. The system sends sensor read commands via the intelligent platform management interface protocol to each baseboard management controller (BMC) in the target cabinet through the local area network. The BMC returns temperature data for each heat-generating device, and the system iterates through this data to find the highest real-time component temperature. The system performs a subtraction operation, subtracting the highest real-time component temperature from the preset hardware alarm temperature to obtain the real-time temperature margin. Next, the system executes conditional judgment logic, comparing the real-time temperature margin with a preset safety margin threshold. When the real-time temperature margin is greater than the preset safety margin threshold, the system reads the pre-set static safety margin opening value from the configuration file and assigns it to the preset minimum safe heat dissipation opening variable. When the real-time temperature margin is less than or equal to the preset safety margin threshold, the system calls an inverse proportional function algorithm module, using the real-time temperature margin as the denominator input, to calculate a dynamic safety margin opening that increases as the margin decreases, and assigns it to the preset minimum safe heat dissipation opening variable. Finally, the system calls the maximum value comparison function to compare the target throttling opening with the preset minimum opening for safe heat dissipation of the device, and outputs the larger of the two values ​​as the actual opening.

[0055] The system can also perform this step via the Redfish application programming interface (API) and polynomial curve fitting calculation. As a client, the system sends a secure retrieval request (HTTP) to the Redfish API of the server in the target rack to obtain a universal data format file containing the status of all thermal sensors. The system parses this file and extracts the real-time maximum component temperature. The system uses an arithmetic logic unit (ALU) to calculate the difference between the preset hardware alarm temperature and the real-time maximum component temperature, using this as the real-time temperature margin. The system's internal comparator module compares this real-time temperature margin with a preset safety margin threshold. If the margin is greater than the threshold, the system directly calls a static minimum operating degree constant stored in memory as the preset minimum operating degree for safe heat dissipation. If the margin is less than or equal to the threshold, the system inputs the real-time temperature margin into a pre-trained polynomial curve fitting model. This model outputs a dynamic minimum operating degree that changes inversely to the margin, which the system uses as the preset minimum operating degree for safe heat dissipation. Ultimately, the system inputs the target throttling opening and the preset minimum opening for safe heat dissipation of the device into a hardware comparator circuit or software comparison logic, selects the one with the larger value, and registers it as the actual execution opening.

[0056] S108. Control the liquid inlet proportional valve to operate according to the actual opening degree and continuously monitor the current surface temperature; Controlling the proportional valve to operate at the actual opening degree means that the system sends the calculated final opening command to the valve actuator, causing it to reach the specified physical position. Continuous monitoring of the current surface temperature means that the system continuously and cyclically acquires the temperature data of the outer wall of the coolant pipeline after the valve actuates. After determining the optimal valve opening degree that both prevents condensation and ensures heat dissipation, the system physically executes the operation and establishes a closed-loop monitoring mechanism. Through continuous monitoring, the system can monitor the temperature rise in the pipeline after flow reduction in real time, improving its control over the anti-condensation process and enhancing the system's dynamic response characteristics.

[0057] The system can achieve this step by outputting an analog voltage signal and sampling via a timer interrupt. The analog-to-digital converter inside the system's main microcontroller generates an analog voltage signal with a corresponding amplitude, such as a control voltage of 0 to 10 volts, based on the calculated actual opening value. This analog voltage signal is transmitted via wires to the servo driver of the liquid inlet proportional valve. The driver controls the internal motor to rotate according to the voltage amplitude, moving the valve core to the physical position corresponding to the actual opening. Simultaneously, the main microcontroller is equipped with a hardware timer that triggers an interrupt at a fixed time period. Each time a timer interrupt occurs, the microcontroller pauses the current main loop and executes an analog-to-digital conversion instruction to read the analog voltage from the temperature sensor connected to the coolant line, converting it into a digital temperature value, thereby achieving continuous monitoring of the current surface temperature.

[0058] The system can also implement this step through industrial fieldbus communication and an event-driven subscription mechanism. The system encapsulates the actual opening value into a standard industrial fieldbus protocol data frame, such as a Controller Area Network (CLAN) bus data frame. The system sends this data frame to the communication network connecting various underlying devices via its built-in bus transceiver. Upon receiving this data frame, the intelligent node built into the proportional valve parses the opening command and autonomously controls the stepper motor to operate in a closed loop to the specified opening. To continuously monitor temperature, the system sends a data subscription command to the intelligent temperature transmitter on the pipeline. After receiving the subscription, the temperature transmitter autonomously collects the current surface temperature at a preset high-frequency sampling rate and actively pushes the temperature data frame to the fieldbus after each collection. The system's main control program adopts an event-driven architecture; once the bus interface receives a temperature data frame, it triggers the corresponding callback function to update and record the data.

[0059] S109. When the current surface temperature is detected to reach the safety anti-condensation threshold, or the actual continuous running time of the opening reaches the preset waiting timeout threshold, an opening execution command is sent to the physical door lock.

[0060] The safety anti-condensation threshold refers to the temperature limit obtained by adding a safety margin to the current external dew point temperature. When the pipeline temperature reaches this value, it is considered that there is no longer a risk of condensation when the door is opened. Continuous operating time refers to the cumulative time the inlet proportional valve remains in its actual open state. The preset waiting timeout threshold refers to the maximum allowed waiting time limit for the system to intercept door opening actions. The system is configured with dual release conditions: temperature meets the standard or waiting timeout. This dual condition judgment ensures timely release when the pipeline temperature is safe, while avoiding the problem of users being unable to open the cabinet door for a long time due to the pipeline temperature failing to rise for an extended period under extreme operating conditions. This improves system availability and reduces the waiting costs for maintenance personnel.

[0061] The system can implement this step by polling the main loop and toggling the level of the general-purpose input / output pins. The system's main control program runs continuously in an infinite loop. In each loop, the system first compares the continuously monitored current surface temperature with a pre-calculated safety anti-condensation threshold. Simultaneously, the system reads the current time from the internal hardware real-time clock, subtracts the recorded time when the actual opening was initiated, calculates the continuous running time, and compares it with a preset waiting timeout threshold. These two comparison results are connected via a logical OR operation. When either comparison result is true, the system's main control program exits the polling loop and writes a high-level command to the general-purpose input / output pin controlling the physical door lock relay. After the pin level toggles, the relay closes, the physical door lock's drive circuit is activated, and the electromagnet or motor inside the lock performs the unlocking action, thus completing the process of issuing an opening command to the physical door lock.

[0062] The system can also implement this step through multi-threaded concurrent monitoring and network application interface calls. The system starts two independent monitoring threads in the operating system. The first thread is dedicated to temperature monitoring; it blocks and waits for data updates from the temperature sensor. Once a new current surface temperature is obtained, it is compared with the safety anti-condensation threshold. If the threshold is reached, a temperature compliance event signal is sent to the main thread. The second thread is a timer thread that starts timing when the opening is actually issued. When the timer value reaches a preset timeout threshold, a timeout event signal is sent to the main thread. The main thread uses an event listening mechanism. Once it receives any of the above event signals, it immediately constructs a Hypertext Transfer Protocol (HTTP) request data packet containing an unlocking command and the target cabinet identifier. The main thread sends this request to the access control subsystem's network application interface via the data center LAN. Upon receiving the network command, the access control subsystem controls the underlying lock hardware to perform the unlocking action.

[0063] In the above embodiments, by intercepting the opening of the physical door lock upon receiving a cabinet access unlocking request, the cold piping is prevented from being directly exposed to the external humid and hot air before treatment. By acquiring the external dew point temperature, the piping surface temperature, and the real-time heat load power of the internal heat-generating devices, a flow reduction control coefficient that is negatively correlated with the heat load and positively correlated with the anti-condensation temperature difference is calculated, thereby dynamically adjusting the target throttling opening of the liquid inlet proportional valve. This flow reduction strategy based on real-time operating conditions can proactively increase the piping surface temperature before the door is opened. At the same time, the target throttling opening is compared with the minimum safe heat dissipation opening of the equipment, and the larger value is taken as the actual execution opening. While increasing the piping temperature, the basic heat dissipation requirements of the heat-generating devices are ensured, reducing the risk of equipment overheating due to excessive flow restriction. Finally, the door opening is allowed only after the piping surface temperature reaches the safe threshold, reducing the probability of condensation on the piping surface at the moment the cabinet door is opened and improving the operational reliability of the electronic devices inside the cabinet.

[0064] The above embodiments detail the anti-condensation control strategies before and at the moment the cabinet door is opened, effectively ensuring equipment safety during the door opening phase through proactive temperature control. However, after maintenance operations are completed and the physical door lock is re-closed, some external humid and hot air inevitably intrudes into the cabinet. If the normal high-flow cooling mode is immediately restored at this time, the pipe temperature will drop sharply, easily causing secondary condensation due to contact with the humid and hot air that has not yet been dissipated from the cabinet. To address the condensation risk during the recovery of microenvironment humidity after the door is closed and to achieve closed-loop control throughout the entire process, this application embodiment further provides another anti-condensation control method for CDU units. The following is in conjunction with... Figure 2 Another method for preventing condensation in a CDU unit, as described in the embodiments of this application, is as follows: Please see Figure 2 This is another flowchart illustrating a method for preventing condensation in a CDU unit according to an embodiment of this application.

[0065] S201. When the physical door lock is detected to switch from the open state to the closed state, the switching of the liquid inlet proportional valve to the normal heat dissipation control mode is intercepted. The "open" state refers to the physical door lock being released from its mechanical lock, allowing the cabinet door to be physically opened. The "closed" state refers to the physical door lock being fully engaged, with the cabinet door physically sealed. The standard cooling control mode refers to the system automatically adjusting the coolant flow rate based solely on the real-time heat load power or temperature feedback of the heat-generating components inside the target cabinet, not limited to a temperature-closed-loop control mode based on proportional-integral-derivative algorithms or an open-loop control mode based on power feedforward. Upon sensing that the cabinet door has been closed again, the system proactively intervenes and prevents it from immediately reverting to the standard cooling mode that only considers temperature or power. Because hot, humid air has already entered the cabinet during the opening period, the humidity inside the cabinet is high after the door is closed. If a large flow of coolant is immediately restored, the surface temperature of the pipes will drop rapidly, causing secondary condensation upon contact with the high-humidity air inside the cabinet. By switching to the interception mode, the entry of a large flow of coolant is delayed, reducing the probability of condensation on the pipe surface during the initial closing phase and improving the stability of the cabinet's internal microenvironment during humidity recovery.

[0066] This step can be implemented by triggering an interrupt using a hardware door magnetic sensor and locking the state machine logic. The system's main control board's general-purpose input / output pins are connected to a dry-contact door magnetic sensor mounted on the target cabinet door frame. When the cabinet door changes from open to closed, the door magnetic sensor's reed switch engages, causing a high-to-low transition in the level of the general-purpose input / output pins. After capturing this falling-edge interrupt signal, the main control board's microprocessor enters the interrupt service routine, updating the door lock status flag maintained in memory to the closed state. Subsequently, in the state machine scheduling logic of the main loop, the microprocessor detects this state transition event and immediately forces a Boolean global variable named "Normal Mode Enable" to false. When the underlying valve control task attempts to update the opening of the liquid inlet proportional valve based on the result of the conventional temperature proportional-integral-differential algorithm, it first checks this enable variable. Since it is false, the control task skips the issuance of the opening update instruction, thus intercepting the switch to the conventional heat dissipation control mode at the software logic level.

[0067] This step can also be achieved through a smart electronic lock status polling and control command filtering gateway. The system, acting as a host computer, continuously sends status query messages to the smart electronic lock in the target cabinet at fixed time intervals via a serial communication bus. The microcontroller inside the smart electronic lock reads the status of the latch position sensor and encapsulates it in a response message, returning it to the system. The system parses the response message, compares it with the previous latch status, and when it detects that the latch has changed from a retracted state to an extended locked state, it determines that the physical lock has switched to the closed state. At this time, the command filtering gateway module inside the system is activated. This module is located between the heat dissipation algorithm calculation layer and the underlying hardware driver layer. The heat dissipation algorithm calculation layer continues to output target valve opening commands according to the conventional heat dissipation control mode, but the command filtering gateway module intercepts these control commands with conventional mode identifiers, discards them, and prevents them from entering the sending queue of the underlying hardware driver layer. This keeps the liquid inlet proportional valve in the limited opening state before closing, thus intercepting the mode switching execution.

[0068] S202. Obtain the current internal dew point temperature inside the target cabinet; The current internal dew point temperature refers to the critical temperature at which water vapor in the enclosed space inside the target server rack reaches saturation and begins to condense into water droplets. After confirming the rack door is closed and the normal operating mode is blocked, the system needs to immediately obtain the actual micro-environmental humidity state inside the rack. Due to the intrusion of warm and humid air from outside during the door opening period, the temperature and humidity field inside the rack has changed. Obtaining this real-time data can accurately reflect the current condensation critical point of the air inside the rack. By knowing the current internal dew point temperature, the system can provide a direct physical calculation basis for setting a safe lower limit for pipeline temperature, improving the pertinence and accuracy of the anti-condensation control strategy during the door-close recovery period, and reducing the risk of secondary condensation caused by blind control.

[0069] The system can evenly distribute multiple digital temperature and humidity sensor chips in key locations prone to humid and hot air accumulation, such as the inside of the front door, the exhaust side of the back panel, and the cable management channels on the sides of the target cabinet. These sensor chips are connected in parallel to a local data acquisition board inside the cabinet via an internal integrated circuit bus. The microcontroller on the data acquisition board sends read commands to each sensor chip sequentially according to a set high-frequency sampling period to acquire real-time digital values ​​of air temperature and relative humidity at different locations inside the cabinet. Next, the microcontroller calls the built-in Magnus empirical formula algorithm module, substitutes the acquired temperature and humidity data into the formula, performs floating-point calculations, and calculates the local dew point temperature at each measuring point. Finally, the microcontroller performs a weighted average calculation of these local dew point temperatures, or directly selects the highest value, as the current internal dew point temperature inside the target cabinet, and sends it to the system main control unit via the data center LAN.

[0070] S203. Add the current internal dew point temperature to the preset anti-condensation safety margin to obtain the dynamic safety lower limit of pipe temperature; The preset anti-condensation safety margin refers to an additional temperature compensation value set in advance to compensate for factors such as sensor measurement errors, uneven temperature distribution on the pipe surface, and response delays in the control system. This value is derived from multiple condensation limit tests conducted by the system in a laboratory environment. The dynamic safe lower limit of pipe temperature refers to the lowest safe temperature limit that the coolant pipe surface can reach under the current microenvironmental conditions inside the cabinet. The system adds a safety margin to the acquired internal dew point temperature to form a dynamically changing temperature baseline. Because the dew point temperature inside the cabinet changes continuously during the dehumidification process after the door is closed, a fixed pipe temperature limit cannot simultaneously ensure both safety and heat dissipation efficiency. By calculating the dynamic safe lower limit of pipe temperature, the system can gradually relax the restriction on coolant temperature as the humidity inside the cabinet decreases, improving the system's heat dissipation capacity during the recovery period and reducing the risk of equipment overheating due to overly conservative pipe temperature control.

[0071] This step can be implemented through parameter calls from the microprocessor's arithmetic logic unit (ALU) and read-only memory (ROM). During the initialization phase, the system's main microprocessor reads the pre-programmed preset anti-condensation safety margin value from the electrically erasable programmable read-only memory (EPROM) on the motherboard and loads it into a dedicated variable in the microprocessor's internal static random access memory (SRAM). After the system obtains the current internal dew point temperature through the preprocessing steps and stores it in a register, the microprocessor's control unit schedules the ALU to execute an addition instruction. The ALU performs a floating-point addition operation using the current internal dew point temperature value in the register and the preset anti-condensation safety margin value in the SRAM as operands. After the operation is completed, the ALU outputs the sum to another designated register. The value stored in this register is then marked by the system and used as the dynamic safety lower limit of the pipe temperature for direct reading and use by the subsequent flow control algorithm module.

[0072] This step can also be achieved through programmable logic controller (PLC) function block diagram programming and parameter distribution via a human-machine interface. The system uses a PLC as its core hardware, allocating two specific data block addresses in the controller's memory to store the current internal dew point temperature and the preset anti-condensation safety margin, respectively. The preset anti-condensation safety margin is set by maintenance personnel via a touchscreen human-machine interface connected to the controller. The interface program writes the set value into the corresponding data block address on the controller via a communication protocol. In the controller's function block diagram program, a standard mathematical addition function block is called. The data block address storing the current internal dew point temperature is linked to the first input pin of the addition function block, and the data block address storing the preset anti-condensation safety margin is linked to the second input pin. When the program execution cycle reaches this function block, the controller automatically adds the values ​​of the two pins and outputs the result to the third data block address linked to the output pin of the addition function block. The data at this address is the dynamic safety lower limit of the pipe temperature.

[0073] S204. Calculate the upper limit of the throttling opening during the recovery period based on the lower limit of the dynamic safety pipe temperature and the real-time heat load power of the heat-generating devices inside the target cabinet. Based on the dynamic safety lower limit of pipe temperature and the real-time heat load power of the heat-generating devices inside the target cabinet, the upper limit of the throttling opening during the recovery period is calculated. This step can be achieved as follows: obtain the current coolant supply temperature delivered by the CDU unit to the target cabinet; use the difference between the dynamic safety lower limit of pipe temperature and the coolant supply temperature as the target heat exchange temperature difference; divide the real-time heat load power by the target heat exchange temperature difference and, in conjunction with the specific heat capacity parameter of the coolant, calculate the maximum allowable coolant flow rate; and convert the maximum allowable coolant flow rate into the corresponding upper limit of the throttling opening during the recovery period based on the inherent flow coefficient of the inlet proportional valve.

[0074] First, let's explain the terms used in this step: The upper limit of the throttling opening during the recovery period refers to the maximum physical limit to which the inlet proportional valve is allowed to open during the humidity recovery phase after the cabinet door is closed. The coolant supply temperature refers to the temperature of the coolant currently being output by the unit and delivered to the target cabinet. The target heat exchange temperature difference refers to the difference between the desired minimum safe temperature of the pipe surface and the actual supply temperature. The specific heat capacity parameter of the coolant refers to the value of the amount of heat absorbed or released per unit mass of coolant when its temperature increases or decreases by a unit temperature. The maximum allowable coolant flow rate refers to the maximum volume or mass of coolant that can flow through the pipe per unit time without causing the pipe temperature to fall below the dynamic safe lower limit of the pipe temperature. The inherent flow coefficient is a physical constant representing the flow capacity of the inlet proportional valve at a specific opening. Based on the principle of thermodynamic energy conservation, the system uses the dynamic safe lower limit of the pipe temperature, the supply temperature, and the heat output of the equipment to deduce the maximum allowable valve opening. Through this rigorous physical modeling and calculation, the maximum cooling flow is provided while meeting the minimum anti-condensation temperature requirement, thereby improving heat dissipation efficiency during the recovery period and reducing the likelihood of equipment frequency reduction due to insufficient heat dissipation.

[0075] This step can be achieved by using the microcontroller's internal floating-point unit and a polynomial fitting curve for reverse lookup. The system obtains the coolant supply temperature through a high-precision temperature transmitter installed on the unit's main outlet pipe. The microcontroller's floating-point unit executes a subtraction instruction, subtracting the coolant supply temperature from the dynamic safety pipe temperature lower limit to obtain the target heat exchange temperature difference. Next, the floating-point unit executes a division instruction, dividing the real-time heat load power by the target heat exchange temperature difference, and then dividing the quotient by the coolant's specific heat capacity parameter and density constant pre-stored in memory to calculate the maximum allowable coolant flow rate. To convert the flow rate into the opening degree, the system pre-stores discrete data points in memory showing the relationship between the flow rate and opening degree determined by the inlet proportional valve based on its inherent flow coefficient, and generates a polynomial fitting curve function using the least squares method. The microcontroller substitutes the calculated maximum allowable coolant flow rate as the independent variable into the polynomial fitting curve function for algebraic operations, and the resulting dependent variable is the upper limit of the throttling opening during the recovery period.

[0076] This step can also be achieved by combining the script parsing engine of the edge computing gateway with a lookup table interpolation algorithm. The system subscribes to and obtains the coolant supply temperature from the unit's controller via the industrial IoT platform in the data center using a message queue telemetry transmission protocol. Upon receiving this data, the script parsing engine running inside the edge computing gateway executes the subtraction logic in the script, subtracting the coolant supply temperature from the dynamic safety lower limit of the pipe temperature to calculate the target heat exchange temperature difference. The script engine then executes the division function in the mathematical operation library, dividing the real-time heat load power by the target heat exchange temperature difference, and combines this with the specific heat capacity parameter of the coolant read from the configuration file to obtain the maximum allowable coolant flow rate. Subsequently, the script engine reads a lookup table file containing the inherent flow coefficient of the inlet proportional valve and the correspondence between flow rate and opening degree. The script engine locates the interval containing the maximum allowable coolant flow rate in the lookup table and calls the linear interpolation algorithm function to accurately calculate the corresponding upper limit value of the throttling opening during the recovery period based on the flow rate and opening degree data at both ends of the interval.

[0077] S205. Control the liquid inlet proportional valve to operate within the range not exceeding the upper limit of the throttling opening during the recovery period, and continuously monitor the current internal dew point temperature; The range not exceeding the upper limit of the throttling opening during the recovery period means that the actual operating opening of the inlet proportional valve is strictly limited to the range between zero and the upper limit of the throttling opening during the recovery period. After calculating the safe upper limit of the opening, the system implements amplitude limiting control on the valve while maintaining continuous monitoring of the humidity inside the cabinet. Since the air conditioning system or dehumidification equipment inside the cabinet is continuously operating, the internal dew point temperature is in a dynamic process of decreasing. Continuous monitoring allows the system to grasp the progress of humidity recovery in real time. By limiting the valve opening within the upper limit range, the system ensures that no matter how the internal heat load fluctuates, the coolant flow rate will not exceed the critical value that causes condensation, improving the absolute safety of anti-condensation control during the recovery period and reducing the risk of pipeline condensation caused by flow runaway.

[0078] This step can be implemented using a proportional-integral (PI) control algorithm combined with output limiting logic and timer interrupt polling. The system's main control program runs a PI closed-loop control algorithm module targeting the temperature of the heating components in the target cabinet. This module calculates a preliminary desired valve opening based on the deviation between the real-time temperature and the target temperature. Before outputting this desired opening to the underlying hardware driver, the system calls a software limiting function. This function compares the preliminary desired opening with the upper limit of the recovery period throttling opening calculated in the previous step. If the desired opening is greater than the upper limit, the output value is forcibly truncated and assigned the upper limit of the recovery period throttling opening; if the desired opening is less than or equal to the upper limit, the original value is maintained. The underlying driver outputs the corresponding pulse width modulation signal to control the liquid inlet proportional valve based on the limited opening value. Simultaneously, the system is configured with a hardware timer that triggers an interrupt every fixed millisecond interval. In the interrupt service routine, the system sends read commands to the temperature and humidity sensors inside the cabinet via the internal integrated circuit bus to obtain the latest temperature and humidity data, calculate the current internal dew point temperature, and update the global variables in memory for continuous monitoring.

[0079] This step can also be achieved through hardware comparator instructions of the programmable logic controller (PLC) and a background asynchronous data acquisition task. In the ladder diagram program of the PLC, a conventional temperature control loop uses a proportional-integral-derivative (PID) instruction block to calculate the required analog output value for valve control. This output value is then sent to a limit control instruction block. The upper limit pin of the limit control instruction block is connected to the register address of the upper limit of the throttling opening during the storage recovery period. The instruction block automatically compares the input value with the upper limit value at the hardware level, ensuring that the valve opening corresponding to the voltage or current signal finally output to the digital-to-analog converter module will never exceed the upper limit, thereby controlling the proportional valve to operate within a safe range. To continuously monitor the dew point, an independent background asynchronous communication task is configured in the controller operating system. This task uses a remote terminal unit (RTU) protocol to continuously poll data from the dew point transmitter inside the cabinet via a serial communication interface. Each time a response message is successfully received, the background task parses the current internal dew point temperature value and writes it to the controller's shared data area for the main control logic to access at any time.

[0080] S206. When the current internal dew point temperature is detected to drop to less than or equal to the preset standard liquid supply dew point threshold, the interception is lifted and the liquid inlet proportional valve is controlled to return to the normal heat dissipation control mode.

[0081] The preset standard coolant dew point threshold refers to the upper limit of the safe dew point temperature that the internal environment of the server rack must reach under normal design conditions when a large flow of low-temperature coolant is allowed. This threshold is derived from the data center room environment design specifications and the rated water supply temperature parameters of the server rack. "Release interception" means that the system removes the previous blocking state for switching to the normal cooling control mode. During continuous monitoring of the dew point temperature inside the server rack, once it is found to have dropped below the safe standard line, it indicates that the humid and hot air that entered the server rack due to door opening has been completely dissipated or dehumidified, and the microenvironment has returned to normal. At this time, the system actively releases the restriction on coolant flow, allowing the valves to freely adjust according to the actual cooling needs of the equipment. By promptly releasing the interception and restoring the normal mode, the system maximizes the cooling system's heat dissipation efficiency while ensuring no condensation risk, improves the operating performance of heat-generating components inside the server rack, and reduces the negative impact of prolonged flow restriction on equipment performance.

[0082] This step can be implemented through software state machine conditional transitions and flag resets. The system's main control program uses a finite state machine architecture for task scheduling. During the recovery period, the state machine is in an anti-condensation flow-limiting state. In each main loop cycle, the state machine logic reads the continuously monitored and updated current internal dew point temperature and performs a floating-point comparison with the preset standard supply dew point threshold stored in non-volatile memory. When the comparison result shows that the current internal dew point temperature is less than or equal to the preset standard supply dew point threshold, the state machine triggers the state transition condition. First, the system resets the previously set normal mode enable global Boolean variable to true, thereby removing the interception of the normal heat dissipation control mode at the software level. Then, the state machine switches the current running state back to the normal heat dissipation state. In the normal heat dissipation state, the underlying valve control task is no longer constrained by the upper limit of the throttling opening during the recovery period, but directly receives and executes the unrestricted target valve opening calculated by the temperature proportional integral differential algorithm module based on the real-time heat load, controlling the inlet proportional valve to restore normal high-flow dynamic adjustment.

[0083] This step can also be implemented by triggering the control handover logic through a hardware comparator interrupt. The system's main control board integrates a high-precision analog hardware comparator circuit. The system connects the analog voltage reference signal corresponding to the preset standard liquid supply dew point threshold to the inverting input of the comparator, and the current internal dew point temperature signal, which is collected in real-time and converted into an analog voltage, to the non-inverting input. When the current internal dew point temperature drops, causing the voltage at the non-inverting input to fall below the reference voltage at the inverting input, the output pin level of the hardware comparator flips, generating a falling-edge interrupt signal. After the microprocessor captures this interrupt signal, it enters a specific interrupt clearing service routine. In this routine, the microprocessor executes the control handover logic, stripping the control of the liquid inlet proportional valve from the anti-condensation safety calculation module and re-handing it over to the main heat dissipation control module. After taking over control, the main heat dissipation control module immediately clears any limiting instructions on the output channel and, based on the real-time temperature feedback of each heat-generating device inside the target cabinet, restarts the conventional temperature closed-loop control algorithm, directly sending the opening drive signal under the conventional heat dissipation control mode to the liquid inlet proportional valve.

[0084] In the above embodiments, upon detecting that the physical door lock is closed, the immediate switch of the liquid inlet proportional valve to the conventional heat dissipation control mode is blocked, preventing condensation caused by a sudden drop in pipe temperature due to a large influx of cold water and contact with the humid and hot air entering the cabinet. By acquiring the real-time dew point temperature inside the cabinet after the door is closed and adding a safety margin to generate a dynamic safe lower limit for the pipe temperature, and combining this with the real-time heat load power, the upper limit of the throttling opening during the recovery period is calculated. This mechanism provides a reasonable upper limit constraint on the coolant flow rate during the humidity recovery period after the door is closed, ensuring that the pipe surface temperature remains above the current internal dew point temperature. As the internal environment of the cabinet recovers, the blockage is lifted only when the internal dew point temperature gradually drops below the standard supply dew point threshold, reducing the risk of latent condensation in the initial period after maintenance and improving the safety of the cabinet's internal microenvironment recovery.

[0085] The above embodiments effectively avoid the risk of condensation caused by a sudden drop in pipeline temperature under macroscopic conditions by limiting the upper limit of the opening of the liquid inlet proportional valve after the door is closed. However, in practical applications, the internal structure of the cabinet is often complex, and some heat dissipation blind spots may have already generated trace amounts of latent condensation. Relying solely on monitoring the overall internal dew point temperature is insufficient to ensure that the moisture in these local microenvironments has completely evaporated. If the interception is lifted prematurely and normal heat dissipation is restored before the latent condensation is completely eliminated, it will still pose a safety hazard of short circuits in electronic components. Therefore, this application embodiment further provides another anti-condensation control method for CDU units before performing the unblocking step. The following is combined with Figure 3 The following describes another method for preventing condensation in a CDU unit as described in the embodiments of this application: Please see Figure 3This is another flowchart illustrating an anti-condensation control method for a CDU unit in this application.

[0086] S301. Calculate the theoretical sensible heat rise rate of the heating device based on the real-time heat load power of the heating device inside the target cabinet, the actual coolant flow rate of the inlet proportional valve under the upper limit of the throttling opening during the recovery period, and the preset heat capacity parameters of the heating device. Real-time thermal load power refers to the total heat generated per unit time by the heat-generating devices inside the target cabinet under its current operating state. These devices are not limited to core computing components such as the central processing unit, graphics processing unit, memory modules, or network switching chips. Actual coolant flow rate refers to the actual physical flow rate of coolant through the cabinet's heat exchange pipes when the inlet proportional valve is limited by its maximum throttling opening during the recovery period. Preset heat capacity parameters of the heat-generating devices refer to the inherent physical properties of the devices themselves when their material and encapsulation structure absorb heat, leading to a temperature increase. Theoretical sensible heat rise rate refers to the rate of temperature increase that the heat-generating devices should exhibit solely due to the difference between their own heat generation and coolant dissipation, assuming the cabinet interior is completely dry and there is no moisture evaporation and heat absorption. The system collects these fundamental physical quantities to construct an ideal thermodynamic model of the heat-generating devices. In the complex microenvironment after the door is closed, directly measuring latent condensation is extremely difficult. By calculating a theoretical baseline without phase change interference, a reliable mathematical basis can be provided for subsequent comparisons of abnormal states, improving the system's perception depth of the cabinet's internal thermodynamic state and reducing the probability of misjudging the dryness of the cabinet interior due to a lack of reference.

[0087] This can be achieved by combining the arithmetic logic unit of the main control microprocessor with a thermodynamic fundamental algorithm module. The system's main control microprocessor, through an out-of-band management network interface, uses the intelligent platform management interface protocol to read in real-time the power consumption sensor data of each heat-generating device in the target cabinet, summing the power consumption data to obtain the real-time heat load power. Simultaneously, the microprocessor reads the analog voltage signal from the electromagnetic flowmeter installed on the inlet pipe through an analog-to-digital converter, converting it into the actual coolant flow rate. Subsequently, the microprocessor retrieves pre-written preset heat capacity parameters of the heat-generating devices from its internal electrically erasable programmable read-only memory. In the thermodynamic fundamental algorithm module, the microprocessor's arithmetic logic unit first calculates the net heat difference between the real-time heat load power and the heat carried away by the coolant. This heat carry-away is calculated from the actual coolant flow rate combined with the coolant specific heat capacity parameter. Next, the arithmetic logic unit divides this net heat difference by the preset heat capacity parameter of the heat-generating device, and the quotient is the theoretical sensible heat rise rate, which is stored in a designated random access memory address.

[0088] This can also be achieved by combining the script parsing engine of the edge computing gateway with a digital twin thermodynamic model. The system deploys an edge computing gateway on the data center side. The gateway continuously subscribes to real-time heat load power of the heat-generating devices inside the rack and the actual coolant flow rate data at the underlying level via a message queue telemetry transmission protocol. Internally, the edge computing gateway runs a digital twin thermodynamic model written in a high-level scripting language. During initialization, this model loads a device configuration file containing preset heat capacity parameters for the heat-generating devices. Upon receiving the latest power and flow data, the script parsing engine uses this data as input parameters into the digital twin thermodynamic model. The model internally calls a numerical integration and differentiation algorithm library to simulate the heat accumulation process of the heat-generating devices under the current heat input and cold output conditions, thereby deriving the theoretical sensible heat rise rate of the heat-generating devices under conditions without phase change interference. After the calculation is completed, the edge computing gateway transmits this rate value back to the system's core control unit via a high-speed Ethernet bus.

[0089] S302. Obtain the current actual temperature rise rate of the heating element; The actual temperature rise rate refers to the rate at which the surface or core temperature of a heat-generating device inside a target cabinet rises over time in a real operating environment, influenced by a combination of factors including its own heat generation, coolant dissipation, and potential heat absorption from the evaporation of ambient moisture. After obtaining the theoretical baseline, the system must simultaneously acquire the actual temperature changes of the heat-generating device in the real physical world. Since the evaporation process of latent condensation absorbs a significant amount of latent heat, the actual temperature rise rate of the heat-generating device will inevitably be slower than theoretically expected. By acquiring this actual dynamic temperature change rate, the system can use it as a key clue to reveal whether there are abnormal heat absorption phenomena inside the cabinet, improving the sensitivity to capturing microscopic thermodynamic changes inside the cabinet and reducing the risk of neglecting dynamic phase transition processes by focusing only on static temperature values.

[0090] This can be achieved by combining high-frequency temperature sampling with a sliding window least squares linear regression algorithm. The system's main control unit continuously reads real-time temperature values ​​from a high-precision digital temperature sensor attached to the surface of the heating element at an extremely high sampling frequency via an internal integrated circuit bus. These continuous temperature data points are sequentially stored in a pre-allocated first-in-first-out sliding window queue in the microcontroller's memory. When the sliding window is full, the microcontroller calls its built-in mathematical operation library to perform a least squares linear regression algorithm on all temperature data points and their corresponding timestamps within the window. This algorithm calculates the slope of the fitted line by finding a straight line that minimizes the sum of squared errors of all data points. This slope value accurately represents the trend of temperature change of the heating element within the current time period, i.e., the actual temperature rise rate. As new data is continuously stored and old data is removed, the system updates this rate value in real time.

[0091] S303. The difference between the theoretical sensible heat temperature rise rate and the actual temperature rise rate is taken as the phase change latent heat loss deviation. Phase change latent heat loss deviation refers to the numerical difference between the theoretical sensible heat rise rate and the actual temperature rise rate of a heat-generating device under the same time dimension and thermodynamic boundary conditions. The system calculates this difference to quantify and isolate the hidden phase change process. In an ideal dry environment, the theoretical and actual rates should be highly consistent, with the difference approaching zero. However, when latent condensation exists in the cabinet's blind spots, moisture evaporation steals some of the heat that should be used to raise the device temperature, resulting in an actual rate lower than the theoretical rate. By calculating this deviation, the system can intuitively measure the scale of latent heat consumed by moisture evaporation, improving the quantitative analysis capability of the latent condensation state, reducing over-reliance on data from a single humidity sensor, and making the anti-condensation control strategy more closely aligned with the actual physical phase change situation inside the cabinet.

[0092] This can be achieved through direct manipulation of registers by the microprocessor's floating-point unit. After completing the preliminary calculations and acquisitions, the system's main microprocessor loads the theoretical sensible heat rise rate data stored in random access memory into the first data register of the floating-point unit, and simultaneously loads the actual temperature rise rate data into the second data register. Subsequently, the microprocessor's control unit issues a single-precision or double-precision floating-point subtraction instruction to the floating-point unit. The floating-point unit executes this instruction, subtracting the value in the second register from the value in the first register, obtaining a floating-point result representing the rate difference. After the calculation, the microprocessor outputs the result and overwrites it into a specially named global variable memory address. The data at this address is then recognized by the system as the phase change latent heat loss deviation, which can be called upon by subsequent conditional judgment logic at any time.

[0093] This can also be achieved by mapping mathematical operation function blocks and data blocks within the programmable logic controller (PLC). In a PLC-based system architecture, the theoretical sensible heat rise rate and the actual temperature rise rate are updated in real time and mapped to two specific floating-point data block addresses within the controller. In the controller's loop-scanning ladder diagram program, a standard mathematical subtraction function block is called. The data block address mapping the theoretical sensible heat rise rate is connected to the minuend input pin of the subtraction function block, and the data block address mapping the actual temperature rise rate is connected to the subtrahend input pin of the subtraction function block. When the PLC's program pointer scans and executes this function block, the underlying firmware automatically reads the values ​​from the two pins, performs the subtraction operation, and outputs the calculated difference to the third data block address bound to the function block's output pin. The content of this newly generated data block represents the phase change latent heat loss deviation.

[0094] S304. When the latent heat loss of phase change is greater than the preset microenvironment liquefaction residue threshold, it is determined that there is hidden condensation in the heat dissipation blind zone of the target cabinet in the evaporative heat absorption state. The preset microenvironment liquefaction residue threshold is a critical value used to determine whether the heat absorption effect of moisture evaporation in small local spaces inside the server rack is significant. This threshold is derived from a large amount of high-humidity environment simulation test data conducted during the system's development phase for specific server rack structures, combined with the measurement error range of temperature sensors. Heat dissipation blind spots refer to localized areas inside the server rack where airflow is extremely slow due to structural obstructions, dense cabling, or airflow design limitations, making it difficult for heat and moisture to dissipate quickly. These are not limited to corners at the bottom of the rack, behind the back panel of high-density servers, or inside cable management channels. Latent condensation refers to tiny liquid droplets that quietly form in these heat dissipation blind spots due to local microenvironment deterioration and are difficult to detect directly by conventional macroscopic sensors. The system makes a final physical state judgment by comparing the calculated deviation with the preset threshold. If the deviation exceeds the allowable error range, it indicates the presence of a significant additional heat source, namely moisture evaporation. This judgment process improves the system's diagnostic accuracy for moisture residue in complex dead spaces inside the server rack and reduces the potential hardware damage risk caused by undetected condensation in blind spots.

[0095] This can be achieved through software logic comparison instructions and state machine flag settings. In the main control program's main loop's logic judgment branch, the phase change latent heat loss deviation value stored in memory is read, and the preset microenvironment liquefaction residue threshold is read from read-only memory. The microprocessor calls its internal logic comparison instructions to compare these two floating-point numbers. When the comparison instruction's execution result indicates that the phase change latent heat loss deviation is strictly greater than the preset microenvironment liquefaction residue threshold, the microprocessor's judgment condition is met. At this point, the system immediately performs a state update operation, forcibly setting a Boolean variable named "Hidden Condensation Existence Flag" in memory from false to true. Subsequent control tasks, during scheduling, will prioritize checking the state of this flag. Once it is found to be true, it confirms the existence of hidden condensation in an evaporative heat absorption state within the target cabinet's heat dissipation blind zone, thereby triggering the corresponding maintenance and interception protection mechanism.

[0096] This can also be achieved through the rule engine and event-driven publishing mechanism of the edge computing gateway. The system pushes the calculated latent heat loss deviation value of phase change to the lightweight rule engine deployed within the edge computing gateway in real time. The rule engine preloads a set of judgment rule bases written in Extensible Markup Language, which explicitly defines the specific values ​​of the preset microenvironment liquefaction residue threshold and the triggering conditions. The rule engine automatically parses the rule base and compares the input deviation value with the threshold according to the set evaluation cycle. When the rule engine finds that the deviation value is greater than the threshold, the judgment rule is triggered. The rule engine then generates an event payload containing the alarm level, rack number, and latent condensation judgment result, and publishes it to the core control module of the system in an event-driven manner through the internal message bus. After subscribing to the event, the core control module immediately confirms the presence of latent condensation in the heat dissipation blind spot.

[0097] S305. Maintain the interception state of the liquid inlet proportional valve and control the liquid inlet proportional valve to continue operating at the upper limit of the throttling opening during the recovery period until the phase change latent heat loss deviation is continuously monitored to decrease to less than or equal to the preset microenvironment liquefaction residue threshold, and then execute the step of releasing the interception.

[0098] The system maintains the interception state of the inlet proportional valve and controls it to continue operating at the upper limit of the throttling opening during the recovery period until the phase change latent heat loss deviation is continuously monitored to decrease to less than or equal to the preset microenvironment liquefaction residue threshold. Then, the interception is released. This release can be achieved as follows: While maintaining the interception state of the inlet proportional valve, a preset amplitude and preset frequency micro-perturbation signal is superimposed on the upper limit of the throttling opening during the recovery period, causing the actual coolant flow rate to oscillate periodically with a small amplitude. Based on the periodically oscillating actual coolant flow rate, the theoretical temperature ripple amplitude of the heating element at the preset frequency is calculated. The actual temperature ripple amplitude of the current actual temperature signal of the heating element at the preset frequency is extracted. The ripple attenuation rate of the actual temperature ripple amplitude relative to the theoretical temperature ripple amplitude is calculated. The phase change latent heat loss deviation is weighted and corrected using the ripple attenuation rate to obtain the net latent heat deviation. When the net latent heat deviation decreases to less than or equal to the preset microenvironment liquefaction residue threshold, and the ripple attenuation rate approaches zero, the interception is released.

[0099] First, let's explain the terms used in this step: Preset amplitude refers to the pre-defined value of the minute fluctuation amplitude added to the valve opening command. Preset frequency refers to the number of times this minute fluctuation signal oscillates per second. Opening micro-perturbation signal refers to a periodic control command, not limited to sine or triangular wave form, used to superimpose on the basic opening command. Periodic micro-amplitude oscillation refers to the regular, small-amplitude increase or decrease in coolant flow rate around the basic flow rate. Theoretical temperature ripple amplitude refers to the expected temperature fluctuation amplitude of the heating element in response to this flow oscillation, without phase change interference. Actual temperature ripple amplitude refers to the actual measured temperature fluctuation amplitude of the heating element. Ripple attenuation rate refers to the reduction ratio of the actual amplitude to the theoretical amplitude. Net latent heat deviation refers to the deviation correction value caused purely by water evaporation after excluding interferences such as heat capacity and heat transfer delay. The preset microenvironment liquefaction residue threshold is consistent with the previous description. While maintaining interception, the system actively injects micro-perturbation signals and analyzes the dynamic response of temperature. Utilizing the low-pass filtering smoothing effect of moisture evaporation on temperature fluctuations, it accurately separates the impact of latent heat absorption. This active detection mechanism improves the system's reliability in confirming whether latent condensation has been completely eliminated under complex operating conditions, reduces the risk of misjudgment due to sensor noise or load fluctuations, and ensures that normal heat dissipation is only resumed after the moisture has truly and completely evaporated.

[0100] This can be achieved through active excitation injection of a digital signal processor (DSP) and a fast Fourier transform (FFT) spectral analysis algorithm. While maintaining the interception state, the DSP uses an internal sine wave generator algorithm to generate a micro-perturbation signal with a preset amplitude and frequency for the opening. This signal is added to the upper limit of the throttling opening during the recovery period and output to the inlet proportional valve, causing the actual coolant flow rate to oscillate periodically with a small amplitude. The processor calculates the theoretical temperature ripple amplitude based on the flow oscillation data and the heat conduction differential equation. Simultaneously, the actual temperature sequence of the heating device is acquired, and a fast Fourier transform algorithm is applied to convert the time-domain signal to the frequency domain, extracting the actual temperature ripple amplitude at the preset frequency point. The actual amplitude is divided by the theoretical amplitude to obtain the ripple attenuation rate. Subsequently, the processor multiplies the phase change latent heat loss deviation by this attenuation rate to calculate the net latent heat deviation. Finally, the processor continuously compares the net latent heat deviation with a preset microenvironment liquefaction residue threshold and checks whether the attenuation rate approaches zero. When the net latent heat deviation is less than or equal to the threshold and the attenuation rate is below the minimum tolerance, the processor modifies the state machine variables and executes the unlocking step.

[0101] This can also be achieved through pulse width modulation superposition and digital bandpass filter envelope detection using a programmable logic controller (PLC). The PLC, based on the analog signal representing the upper limit of the throttling opening during the output recovery period, uses an internal timer to superimpose a square wave micro-perturbation signal with a preset amplitude and frequency. The internal transfer function block of the controller calculates the theoretical temperature ripple amplitude based on this square wave flow oscillation. For the acquired actual temperature signal, the controller inputs it into an infinite impulse response digital bandpass filter with a center frequency set to a preset frequency to filter out temperature noise in other frequency bands. Then, the actual temperature ripple amplitude is extracted using an envelope detection algorithm. The ripple attenuation rate is calculated using a division instruction, and this attenuation rate is used as a weighting coefficient to linearly weight and correct the phase change latent heat loss deviation, yielding the net latent heat deviation. The controller's logic comparator continuously monitors these variables. When it is confirmed that the net latent heat deviation drops to or below the preset microenvironment liquefaction residue threshold, and the ripple attenuation rate drops to near-zero dead zone, the controller triggers a reset coil, cancels the blocking enable for the normal heat dissipation mode, and executes the deblocking step.

[0102] In the above embodiments, the theoretical sensible heat rise rate is calculated based on the real-time heat load power, actual coolant flow rate, and heat capacity parameters of the heating devices. This calculation is then compared with the obtained actual temperature rise rate to obtain the phase change latent heat loss deviation. When this deviation exceeds a preset microenvironment liquefaction residue threshold, the system can sensitively detect situations where the actual temperature rise is lower than expected, thus accurately determining the presence of latent condensation in the cabinet's heat dissipation blind zone that is in an evaporative heat absorption state. Based on this determination, the system maintains the interception state of the liquid inlet proportional valve and operates at the upper limit of the throttling opening during the recovery period, forcing the latent condensation to continue evaporating under controlled pipe temperature conditions until the phase change latent heat loss deviation drops to a safe range before releasing the interception. This improves the ability to identify and remove latent condensation that is difficult to directly observe in the complex internal structure of the cabinet, and reduces the risk of secondary condensation or short circuits in electronic components caused by residual moisture in the blind zone when the system prematurely resumes normal high-flow heat dissipation.

[0103] The system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 4 This is a schematic diagram of the physical device structure of an anti-condensation control system for a CDU unit provided in an embodiment of this application.

[0104] It should be noted that, Figure 4 The structure of the system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0105] like Figure 4As shown, the system includes a Central Processing Unit (CPU) 401, which can perform various appropriate actions and processes based on a program stored in Read-Only Memory (ROM) 402 or a program loaded from storage portion 408 into Random Access Memory (RAM) 403, such as executing the methods described in the above embodiments. The RAM 403 also stores various programs and data required for system operation. The CPU 401, ROM 402, and RAM 403 are interconnected via a bus 404. An Input / Output (I / O) interface 405 is also connected to the bus 404.

[0106] The following components are connected to I / O interface 405: input section 406 including a camera, infrared sensor, etc.; output section 407 including a liquid crystal display (LCD) and speakers, etc.; storage section 408 including a hard disk, etc.; and communication section 409 including a network interface card such as a LAN (Local Area Network) card and a modem, etc. Communication section 409 performs communication processing via a network such as the Internet. Drive 410 is also connected to I / O interface 405 as needed. Removable media 411, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 410 as needed so that computer programs read from it can be installed into storage section 408 as needed.

[0107] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 409, and / or installed from removable medium 411. When the computer program is executed by central processing unit (CPU) 401, it performs the various functions defined in the present invention.

[0108] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0109] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0110] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the system described in the above embodiments; or it may exist independently and not assembled into the system. The storage medium carries one or more computer programs that, when executed by a processor of a system, cause the system to implement the methods provided in the above embodiments.

[0111] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0112] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0113] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and 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 instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0114] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for preventing condensation control in a CDU unit, characterized in that, include: Receive the access control unlocking request command from the target cabinet and intercept the opening of the physical door lock; In response to the access control unlocking request command, the current external dew point temperature of the external environment where the target cabinet is located, and the real-time heat load power of the heat-generating devices inside the target cabinet are obtained. Obtain the current surface temperature of the coolant piping inside the target cabinet; When the current surface temperature is lower than the current external dew point temperature, calculate the anti-condensation temperature difference between the current external dew point temperature and the current surface temperature; Based on the real-time heat load power and the anti-condensation temperature difference, the flow reduction control coefficient of the liquid inlet proportional valve is calculated. The flow reduction control coefficient represents the reduction ratio of the current coolant flow. The flow reduction control coefficient is negatively correlated with the real-time heat load power and positively correlated with the anti-condensation temperature difference. Calculate the target throttling opening based on the flow reduction control coefficient and the current valve opening of the inlet proportional valve; The larger of the target throttling opening and the preset minimum opening for safe heat dissipation of the equipment is taken as the actual opening. The liquid inlet proportional valve is controlled to operate according to the actual opening degree, and the current surface temperature is continuously monitored; When the current surface temperature is detected to reach the safety anti-condensation threshold, or the continuous running time of the actual opening reaches the preset waiting timeout threshold, an opening execution command is sent to the physical door lock. When the physical door lock is detected to switch from the open state to the closed state, the switching of the liquid inlet proportional valve to the normal heat dissipation control mode is intercepted; Obtain the current internal dew point temperature inside the target cabinet; The current internal dew point temperature is added to the preset anti-condensation safety margin to obtain the dynamic safety lower limit of the pipe temperature; Calculate the upper limit of the throttling opening during the recovery period based on the lower limit of the dynamic safety pipe temperature and the real-time heat load power of the heat-generating devices inside the target cabinet; The liquid inlet proportional valve is controlled to operate within the range not exceeding the upper limit of the throttling opening during the recovery period, and the current internal dew point temperature is continuously monitored; When the current internal dew point temperature is detected to drop to less than or equal to the preset standard supply dew point threshold, the theoretical sensible heat rise rate of the heating device is calculated based on the real-time heat load power of the heating device inside the target cabinet, the actual coolant flow rate of the inlet proportional valve under the upper limit of the throttling opening during the recovery period, and the preset heat capacity parameters of the heating device. Obtain the current actual temperature rise rate of the heating device; The difference between the theoretical sensible heat temperature rise rate and the actual temperature rise rate is taken as the phase change latent heat loss deviation. When the phase change latent heat loss deviation is greater than the preset microenvironment liquefaction residue threshold, it is determined that there is hidden condensation in the heat dissipation blind zone of the target cabinet in an evaporative heat absorption state. Maintain the interception state of the liquid inlet proportional valve and control the liquid inlet proportional valve to continue operating at the upper limit of the throttling opening during the recovery period until the phase change latent heat loss deviation is continuously monitored to decrease to less than or equal to the preset microenvironment liquefaction residue threshold. Then, release the interception and control the liquid inlet proportional valve to return to the normal heat dissipation control mode.

2. The anti-condensation control method for CDU units according to claim 1, characterized in that, The step of calculating the flow reduction control coefficient of the liquid inlet proportional valve based on the real-time heat load power and the anti-condensation temperature difference specifically includes: The real-time heat load power and the anti-condensation temperature difference are input into a preset flow reduction matrix to determine the basic flow reduction coefficient. The basic flow reduction coefficient is negatively correlated with the real-time heat load power and positively correlated with the anti-condensation temperature difference. Obtain the current liquid supply temperature delivered by the CDU unit to the target cabinet; Calculate the temperature deviation between the preset standard design water temperature and the current supply temperature; The temperature compensation factor is calculated based on the temperature deviation value. When the current liquid supply temperature is lower than the preset standard design water temperature, the temperature compensation factor is greater than 1. Multiplying the basic flow reduction coefficient by the temperature compensation factor yields the flow reduction control coefficient.

3. The anti-condensation control method for CDU units according to claim 1, characterized in that, The step of using the larger of the target throttling opening and the preset minimum safe heat dissipation opening of the equipment as the actual execution opening specifically includes: The real-time highest component temperature of the heat-generating devices inside the target cabinet is obtained through the Baseboard Management Controller (BMC) of the target cabinet. The difference between the preset hardware alarm temperature and the real-time highest component temperature is used as the real-time temperature margin. When the real-time temperature margin is greater than the preset safety margin threshold, the preset static minimum opening degree will be used as the preset minimum opening degree for safe heat dissipation of the device. When the real-time temperature margin is less than or equal to the preset safety margin threshold, a dynamic minimum opening degree that is inversely proportional to the real-time temperature margin is calculated, and the dynamic minimum opening degree is used as the preset minimum opening degree for safe heat dissipation of the device. The larger of the target throttling opening and the preset minimum opening for safe heat dissipation of the equipment is taken as the actual opening.

4. The anti-condensation control method for CDU units according to claim 1, characterized in that, The step of calculating the upper limit of the throttling opening during the recovery period based on the lower limit of the dynamic safety pipe temperature and the real-time heat load power of the heat-generating devices inside the target cabinet specifically includes: Obtain the current coolant supply temperature of the CDU unit delivered to the target cabinet; The difference between the dynamic safety pipe temperature lower limit and the coolant supply temperature is taken as the target heat exchange temperature difference; The maximum allowable coolant flow rate is calculated by dividing the real-time heat load power by the target heat exchange temperature difference and combining it with the specific heat capacity parameter of the coolant. Based on the inherent flow coefficient of the inlet proportional valve, the maximum allowable coolant flow rate is converted into the corresponding upper limit of the throttling opening during the recovery period.

5. The anti-condensation control method for CDU units according to claim 1, characterized in that, The step of maintaining the interception state of the liquid inlet proportional valve and controlling the liquid inlet proportional valve to continue operating at the upper limit of the throttling opening during the recovery period until the latent heat loss deviation of the phase change is continuously monitored to decrease to less than or equal to the preset microenvironment liquefaction residue threshold, and then performing the step of releasing the interception, specifically includes: While maintaining the interception state of the inlet proportional valve, a micro-perturbation signal with a preset amplitude and preset frequency is superimposed on the upper limit of the throttling opening during the recovery period, so that the actual coolant flow rate exhibits periodic micro-oscillation. Based on the actual coolant flow rate of the periodic micro-oscillation, calculate the theoretical temperature ripple amplitude of the heating device at the preset frequency; Extract the actual temperature ripple amplitude of the current actual temperature signal of the heating device at the preset frequency; Calculate the ripple attenuation rate of the actual temperature ripple amplitude relative to the theoretical temperature ripple amplitude; The net latent heat loss deviation of the phase change is obtained by weighting the ripple attenuation rate. When the net latent heat deviation decreases to less than or equal to the preset microenvironment liquefaction residue threshold, and the ripple attenuation rate approaches zero, the step of releasing the interception is executed.

6. A condensation prevention control system for a CDU unit, characterized in that, The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the anti-condensation control method for a CDU unit as described in any one of claims 1-5.

7. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the system, the system performs the anti-condensation control method for the CDU unit as described in any one of claims 1-5.

8. A computer program product, characterized in that, When the computer program product is run on the system, the system performs the anti-condensation control method for the CDU unit as described in any one of claims 1-5.

Images