Data center immersion cooling method and system based on dryness sensing cooperative regulation

Abstract

The application discloses a data center immersion cooling method and system based on dryness sensing cooperative regulation, and relates to the technical field of data center cooling control. The data center immersion cooling method and system based on dryness sensing cooperative regulation comprises the following steps: S1: collecting key physical parameters in the immersion liquid cooling environment, obtaining cooling working condition collection data and performing pretreatment; S2: based on the immersion server and the upper and lower layered heat exchange structure, the heat release state of the cold plate is characterized; S3: the cooling flow adjustment range is determined according to the dryness deviation; S4: the regulation response situation is identified in combination with the dryness and temperature change trend in the regulation process. The application solves the problems that the existing two-phase immersion liquid cooling data center lacks real-time monitoring and automatic flow regulation mechanism based on dryness change, cannot dynamically respond to the phase change disturbance caused by bubble generation, and leads to the problems of disconnection of cooling intensity and heat load and lag of heat dissipation response.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data center cooling control, in particular to a data center immersion cooling method and system based on dryness sensing cooperative regulation. BACKGROUND

[0002] With the expansion of data centers and the continuous increase of heat flux density, traditional air cooling methods are gradually replaced by liquid cooling technology due to low energy efficiency and obvious heat dissipation bottlenecks. Phase change immersion liquid cooling directly covers server devices with low boiling point cooling liquid to achieve full contact heat dissipation, and combines with upper PCM cold plate structure to improve gas-liquid conversion efficiency, which is an important development direction in the current liquid cooling field. Dryness, as a key parameter representing the gas-liquid state of the cooling liquid, reflects the phase change intensity and heat transfer efficiency, and its change trend is closely related to the bubble release characteristics, which is of great significance to the dynamic adjustment of the cooling process.

[0003] For example, the invention patent with publication number CN115097877B discloses a liquid cooling system vaporization prevention control method and a liquid cooling system. The method includes: calculating the difference between the real-time boiling point value and the real-time temperature value of the secondary side return liquid of the liquid cooling system; if the difference between the real-time boiling point value and the real-time temperature value is less than a first preset threshold and greater than or equal to a second preset threshold, then performing a first level vaporization prevention control on the liquid cooling system; if the difference between the real-time boiling point value and the real-time temperature value is less than the second preset threshold and greater than or equal to a third preset threshold, then performing a second level vaporization prevention control on the liquid cooling system; if the difference between the real-time boiling point value and the real-time temperature value is less than the third preset threshold, then performing a third level vaporization prevention control on the liquid cooling system; according to the difference between the real-time boiling point value and the real-time temperature value of the secondary side return liquid, the corresponding level of vaporization prevention control is adopted, and different means are used to control the pressure and temperature of the cooling liquid to prevent vaporization, so that the operation of the liquid cooling system is more reliable.

[0004] For example, the invention patent with publication number CN111552331B discloses a cooling control method and a cooling control system, relating to the field of electric vehicles. The method collects the temperature difference between the motor winding and the cooling liquid, the size of the stator winding current, the motor winding temperature and the magnetic field frequency in real time, and adjusts the water cooling flow according to the four variables. The cooling loop control system can judge the response value of the database according to the actual situation and execute it, and the cooling effect is good. In addition, since the current, frequency and temperature variables can be detected and fed back in real time, the response speed is fast, and the cooling system is also avoided from being in high gear for a long time. This method can not only solve the slow response speed and power consumption problem based on gear adjustment, but also can macroscopically adjust the variables affecting the flow size of the cooling loop, and overcome the large inertia of non-linear adjustment of water flow, which is easier to realize on vehicles.

[0005] In the existing two-phase immersion liquid cooling data center, there is generally a lack of real-time monitoring of the dryness of the cooling liquid and automatic flow regulation mechanism based on the change of dryness, which cannot dynamically respond to the phase change disturbance caused by bubble generation, and it is difficult to suppress the local bubble aggregation and flow disturbance phenomenon in time, resulting in the disconnection of cooling strength and heat load under high heat flux conditions, causing heat dissipation response lag, which restricts the stable application of liquid cooling technology in high-performance computing scenarios.

[0006] In view of the above problems, there is an urgent need for a data center immersion cooling method and system based on dryness sensing and cooperative regulation. SUMMARY

[0007] Technical problems solved

[0008] In view of the deficiencies of the prior art, the present application provides a data center immersion cooling method and system based on dryness sensing and cooperative regulation, which solves the problem that the existing two-phase immersion liquid cooling data center lacks real-time monitoring and automatic flow regulation mechanism based on the change of dryness, cannot dynamically respond to the phase change disturbance caused by bubble generation, and causes the disconnection of cooling strength and heat load, heat dissipation response lag.

[0009] Technical scheme

[0010] To achieve the above purpose, the present application is realized by the following technical scheme: a data center immersion cooling method based on dryness sensing and cooperative regulation, comprising the following steps: S1: collecting key physical parameters in the immersion liquid cooling environment, obtaining cooling working condition collection data and preprocessing; S2: based on the immersion server and the upper and lower layered heat exchange structure, representing the heat release state of the cold plate, and determining whether to enter the dryness linkage regulation process; S3: in the dryness linkage regulation process, determining the cooling flow regulation amplitude according to the dryness deviation, controlling the electromagnetic valve opening degree, pump control frequency and heat exchange structure configuration according to the cooling flow regulation amplitude; S4: combining the dryness and temperature change trend in the regulation process to identify the regulation response, and adjusting the liquid supply calculation frequency, valve control step and heat exchange area scheduling level according to the situation.

[0011] Further, the collection of key physical parameters in the immersion liquid cooling environment, acquisition of cooling condition collection data and pretreatment specific steps are as follows: collecting key physical parameters in the immersion liquid cooling environment, acquiring cooling condition collection data process includes: arranging temperature sensor in two-phase cooling liquid area to collect cooling liquid temperature; setting thermocouple measuring point above PCM cold plate to acquire cold plate surface temperature; arranging dryness sensor in cooling liquid vaporization core area, generating real-time dryness value by sensing gas-liquid phase state ratio; installing flow sensor at liquid supply inlet section and reading initial liquid inlet rate to form basic liquid supply flow rate value; extracting data upload corresponding time field from SCADA system to determine sampling time point; finally forming cooling condition collection data covering cooling state, heat conduction response and flow reference; performing sequence smoothing and noise suppression on cooling condition collection data by sliding median filter algorithm and local linear regression method; performing data completion on field missing problem caused by sampling delay and short interruption by time step alignment rule and linear interpolation method; performing standardization and normalization processing on processed cooling condition collection data.

[0012] Further, the specific steps based on the immersion server and the upper and lower layered heat exchange structure are as follows: the server mainboard is arranged in a horizontal paving manner and is completely immersed in low-boiling-point cooling liquid, and heat absorption is completed by the heated vaporization of the cooling liquid; a phase change energy storage material cold plate is arranged above the server mainboard, so that the bubbles generated by the vaporization of the cooling liquid are quickly liquefied after releasing latent heat; a cold water pipeline arrangement structure is arranged below the cold plate, and the heat conducted downward by the phase change cold plate is absorbed by the cold water; the cold water pipeline is connected to an external plate heat exchanger, and the heat conduction cold water is introduced to external industrial and domestic heat scenes through a closed loop; the heat exchanger flux is dynamically adjusted according to the heat flow intensity during the heat exchange process; the above immersion arrangement of the server and the layered form of the heat exchange structure are taken as the structure basis for releasing the characteristics of the sensing cold plate in the heat conduction process.

[0013] Further, the specific steps of the characterization of the cold plate heat conduction release state are as follows: screening a time interval with small dryness fluctuation, stable liquid supply and constant temperature to form a cooling stable period; constructing a square value sequence by extracting non-zero samples of the reference dryness in multiple cooling stable periods, selecting the smallest non-zero item, performing scaling combined with the normalized accuracy boundary to obtain a temperature difference compensation constant; consulting the thermal conductivity parameter table of the selected PCM material to obtain a phase change heat conduction capacity value; obtaining a server mainboard surface structure image, combining the cold plate arrangement size, and using an image projection overlap algorithm to calculate the direct contact area between the cold plate and the mainboard to obtain an effective heat exchange area; retrieving the cooling liquid temperature and the cold plate surface temperature from the cooling working condition data; taking the difference between the cooling liquid temperature and the cold plate surface temperature as the numerator, the sum of the cold plate surface temperature and the temperature difference compensation constant as the denominator, and dividing the numerator by the denominator to obtain a ratio; multiplying the phase change heat conduction capacity value by the 0.8 power of the effective heat exchange area to obtain a product; multiplying the ratio by the product to finally obtain the cold plate heat conduction output value.

[0014] Further, the specific steps of the judgment of whether to enter the dryness linkage regulation process are as follows: real-time comparison of the cold plate heat conduction output value and the heat conduction threshold value, the heat conduction threshold value including a first-level heat conduction threshold value and a second-level heat conduction threshold value; when the cold plate heat conduction output value is greater than or equal to the second-level heat conduction threshold value, it is determined that the region is in a high response state and enters the dryness linkage regulation process; when the cold plate heat conduction output value is greater than or equal to the first-level heat conduction threshold value and less than the second-level heat conduction threshold value, it is determined that the region is in a stable state and does not enter the dryness linkage regulation process; when the cold plate heat conduction output value is less than the first-level heat conduction threshold value, it is determined that the region is in heat conduction lag and enters the dryness linkage regulation process.

[0015] Further, the specific steps of determining the cooling flow adjustment range according to the dryness deviation in the dryness linkage regulation process are as follows: input the high response and the heat conduction release state of the heat conduction lagged cold plate into the dryness linkage regulation process, dynamically calculate the target liquid supply amount of the cooling water according to the deviation degree between the current dryness and the reference dryness; in the cooling process, select a time period in which the dryness, the liquid supply flow rate and the cold plate temperature change range do not exceed the corresponding single sampling accuracy, extract the dryness sensor original data in the time period and calculate the mean value to form the reference dryness value, when the determination area is in a stable state and has not entered the running stage of the dryness linkage regulation process; backtrack the cooling running samples with stable dryness regulation, extract all non-zero reference dryness values and select the smallest effective item, multiply the value by the minimum resolution unit of the dryness sensor to obtain the deviation stable compensation value; obtain the real-time dryness value and the basic liquid supply flow rate value; subtract the reference dryness value from the real-time dryness value, square the difference value, multiply by the dryness deviation adjustment coefficient as the numerator; square the reference dryness and add the deviation stable compensation value as the denominator; divide the numerator by the denominator to obtain a fraction value; add one to the fraction value to form the whole in the parentheses; multiply the result in the parentheses by the basic liquid supply flow rate value to obtain the liquid supply adjustment response value.

[0016] Further, the specific steps of controlling the electromagnetic valve opening degree, pump control frequency and heat exchange structure configuration according to the cooling flow adjustment range are as follows: execute the liquid supply response hierarchical efficiency strategy according to the liquid supply adjustment response value: compare the liquid supply adjustment response value with the dryness deviation allowable threshold value in real time; when the liquid supply adjustment response value is greater than or equal to the dryness deviation allowable threshold value, adjust the cooling liquid electromagnetic valve opening degree according to the relative amplitude between the liquid supply adjustment response value and the dryness deviation allowable threshold value, calculate the liquid supply change amplitude value according to the difference between the actual liquid supply flow rate in the current sampling period and the actual liquid supply flow rate in the last sampling period, increase the cooling pump operating frequency according to the liquid supply change amplitude value and limit the minimum operating period, check the number of active plates in the heat exchanger area and expand the enabled range when the maximum value is not reached, and perform positive correction on the liquid supply branch flow in the dryness rising area; when the liquid supply adjustment response value is less than the dryness deviation allowable threshold value, the cooling liquid electromagnetic valve opening degree is maintained at the current value, the cooling pump operating frequency is kept at the original state, and the heat exchange plate configuration remains unchanged.

[0017] Further, the combination of the regulation process of the dryness and the temperature change trend identification regulation response condition specific steps as follows: tracking the deviation between the cooling results and the regulation response, optimizing the regulation strategy execution; calling the real-time dryness value and the sampling time point; obtaining the real-time dryness value and the sampling time point at the ith moment and the i-1 moment, and recording them as the dryness value at the ith moment, the dryness value at the i-1 moment, the sampling time point at the ith moment and the sampling time point at the i-1 moment; extracting the time interval between every two frames in the continuous dryness collection record, screening out the minimum non-zero value among them, multiplying it by the minimum time resolution granularity of the SCADA system to form the time difference compensation constant; performing difference calculation between the adjacent frame dryness values in the continuous dryness collection record to form the dryness change data sequence, constructing the dryness change data sequence under multiple different continuous frame numbers, evaluating the smooth error and the response delay change trend of each sequence, and selecting the minimum effective frame number with the minimum error and no response lag as the statistical sample number; taking the statistical sample number as the total cycle number, in each group of continuous sampling data, first take the dryness value at the ith moment minus the dryness value at the i-1 moment to get the dryness change amount; then take the sampling time point at the ith moment minus the sampling time point at the i-1 moment, and add the time difference compensation constant in the difference to form the corresponding time interval value; divide the dryness change amount by the corresponding time interval value to calculate the unit time dryness change rate corresponding to the ith group of samples; after completing the calculation of all n groups of samples in turn, sum up the n dryness change rates, and then divide by the statistical sample number to get the dryness fluctuation rate value after averaging all samples.

[0018] Further, the specific steps of adjusting the liquid supply calculation frequency, the valve control step size and the heat exchange region scheduling level according to the dryness fluctuation rate value are as follows: performing dryness disturbance grading release strategy according to the dryness fluctuation rate value: real-time comparison of the dryness fluctuation rate value and the fluctuation threshold value, the fluctuation threshold value including a first-level fluctuation threshold value and a second-level fluctuation threshold value; when the dryness fluctuation rate value is less than the first-level fluctuation threshold value, keep the calculation frequency of the liquid supply adjustment response value unchanged, control the electromagnetic valve opening degree according to the corresponding relationship between the cold plate heat conduction output value and the liquid flow rate; when the dryness fluctuation rate value is greater than or equal to the first-level fluctuation threshold value and less than the second-level fluctuation threshold value, reduce the update frequency of the liquid supply adjustment response value, limit the electromagnetic valve opening degree within the minimum allowed adjustment step size range based on the current value, cancel the liquid redistribution between the cooling regions, increase the cold plate temperature sensor sampling density and update the cold plate heat conduction output value; when the dryness fluctuation rate value is greater than or equal to the second-level fluctuation threshold value, suspend the calculation process of the liquid supply adjustment response value, keep the electromagnetic valve opening degree at the current value, close the flow dynamic adjustment channel of the corresponding heat exchanger partition, and set the execution level to the lowest.

[0019] The second aspect of the present application provides a data center immersion cooling system based on dryness sensing cooperative regulation, comprising: a cooling data acquisition module, a cooling heat exchange structure module, a dryness linkage regulation module and a state feedback cooperation module, characterized by: the cooling data acquisition module is used for collecting key physical parameters in the immersion liquid cooling environment, acquiring cooling working condition collection data and preprocessing; the cooling heat exchange structure module is used for representing the cooling plate heat conduction release state based on the immersion server and the upper and lower layered heat exchange structure, and judging whether to enter the dryness linkage regulation process; the dryness linkage regulation module is used for determining the cooling flow regulation amplitude according to the dryness deviation in the dryness linkage regulation process, controlling the electromagnetic valve opening degree, pump control frequency and heat exchange structure configuration according to the cooling flow regulation amplitude; and the state feedback cooperation module is used for identifying the regulation response situation in combination with the dryness and temperature change trend in the regulation process, and adjusting the liquid supply calculation frequency, valve control step and heat exchange area scheduling level in different cases.

[0020] Advantages

[0021] The present application has the following advantages:

[0022] (1) The present application constructs cooling working condition collection data covering cooling liquid temperature, cooling plate temperature, dryness value and liquid supply rate field, and realizes standardized processing through filtering, complementing and normalizing operation, effectively improves the processing capacity of multi-source sensing data in time sequence alignment, structural consistency and noise robustness, and provides stable data input basis for subsequent linkage regulation strategy.

[0023] (2) The present application performs heat conduction state identification based on the cooling plate heat conduction output value and the set double threshold value structure, determines whether to enter the dryness linkage regulation process on the basis of distinguishing high response, stable and lagging regions, so as to realize regional triggering and conditional execution of cooling control strategy, and improve the execution efficiency and distribution accuracy of regulation logic under dynamic thermal load.

[0024] (3) The present application constructs a continuous response liquid supply regulation calculation model by fusing reference dryness, minimum resolution compensation and deviation adjustment coefficient, so that the cooling flow regulation process has the characteristics of dynamic adjustment, quantitative continuity and deviation sensitivity, which improves the adaptation ability and refined control granularity of different heat conduction structures while ensuring the heat exchange efficiency.

[0025] (4) The present application introduces the dryness fluctuation rate value and sets the hierarchical fluctuation threshold value in the regulation execution process, so that the liquid supply calculation frequency, electromagnetic valve control step and heat exchange area scheduling level can be dynamically adjusted according to the dryness fluctuation degree, realizing rhythm compression of high disturbance state, rhythm maintenance of low disturbance state, effectively enhancing the flexible response ability and execution stability of the control process.

[0026] Of course, implementing any product of the application does not necessarily require achieving all the advantages described above at the same time. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 Flow chart of the data center immersion cooling method based on dryness sensing cooperative regulation of the application;

[0028] Figure 2 Structure diagram of the data center immersion cooling system based on dryness sensing cooperative regulation of the application;

[0029] Figure 3 Distribution diagram of the cold plate heat conduction output value of the application;

[0030] Figure 4 Regulation effect evaluation curve of the application;

[0031] Figure 5 Design diagram of the data center two-phase immersion liquid cooling PCM module of the application;

[0032] Figure 6 Working principle diagram of the two-phase immersion liquid cooling PCM data center refrigeration system of the application;

[0033] Figure 7 System structure diagram of the data center two-phase immersion PCM module of the application. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the application.

[0035] Please refer to Figures 1-7 The embodiments of the application provide a technical solution: a data center immersion cooling method based on dryness sensing cooperative regulation, comprising the following steps: S1: collecting key physical parameters in the immersion liquid cooling environment, obtaining cooling working condition collection data and performing preprocessing; S2: based on the immersion server and the upper and lower layered heat exchange structure, representing the cold plate heat conduction release state, and determining whether to enter the dryness linkage regulation process; S3: in the dryness linkage regulation process, determining the cooling flow adjustment amplitude according to the dryness deviation, and controlling the electromagnetic valve opening degree, pump control frequency and heat exchange structure configuration according to the cooling flow adjustment amplitude; S4: identifying the regulation response situation in combination with the dryness and temperature change trend in the regulation process, and adjusting the liquid supply calculation frequency, valve control step size and heat exchange region scheduling level according to the situation.

[0036] As Figure 1As shown, the method flow of the present application is mainly based on continuous sensing and dryness change in cooling condition. Firstly, the temperature, dryness and liquid supply state in the immersed liquid cooling environment are uniformly collected and pretreated. On this basis, the heat release state of the cold plate is characterized by combining the immersed server and the upper and lower layered heat exchange structure, and whether to enter the dryness linkage regulation process is judged by the heat conduction threshold. After entering the regulation process, the cooling flow regulation amplitude is calculated according to the dryness deviation, and the regulation result is applied to the liquid supply execution unit. At the same time, the change trend of dryness and temperature is continuously tracked, which is used to identify the regulation response state and adjust the supply calculation frequency, valve control step and heat exchange region scheduling level, so as to form a complete closed-loop control process of sensing-judgment-regulation-feedback.

[0037] Specifically, the key physical parameters in the immersed liquid cooling environment are collected, the cooling condition collection data are obtained and the specific steps of pretreatment are as follows: in the process of collecting cooling condition data, first, temperature sensors are arranged in the two-phase cooling liquid area to continuously record the temperature rise state of the liquid after being heated, so as to obtain the cooling liquid temperature change trajectory; a thermocouple measuring point is arranged above the PCM cold plate to obtain the cold plate surface temperature, reflecting the real-time reaction characteristics of phase change interface heat conduction; a dryness sensor is arranged in the cooling liquid vaporization core area to generate real-time dryness value by using the continuous sensing mechanism of gas-liquid phase state ratio, enhancing the dynamic judgment ability of boiling state and bubble distribution; a flow sensor is installed at the inlet section of the liquid supply to read the initial rate of incoming liquid in real time, which is used as the calculation basis for subsequent liquid supply regulation; at the same time, the time field attached to the data upload from the SCADA system is extracted to establish the sampling time point of various types of sensing data, ensuring the time sequence comparability of multi-field data; finally, a structured cooling condition data set containing cooling state, heat conduction response, liquid supply benchmark and other information is formed. In the data processing stage, the sliding median filter algorithm and local linear regression method are used to smooth the sequence and suppress noise of the collected data, improving the data stability; for the missing fields caused by sampling delay and short interruption, the time step alignment rule and linear interpolation method are used to perform data completion; finally, standardization and normalization operations are uniformly performed on all fields to establish a data input structure that meets the linkage control requirements.

[0038] In this embodiment, by arranging temperature, dryness and flow sensors, the cooling liquid temperature, cold plate surface temperature, real-time dryness value and basic liquid supply flow rate are collected, and the complete cooling condition data are formed in combination with the SCADA system time field. In the pretreatment process, filtering and regression operations are performed to smooth the data, missing fields are completed by time alignment and interpolation method, and finally all kinds of parameters are standardized and normalized to improve the continuity and consistency of the data structure, providing a high reliability input basis for the subsequent regulation logic.

[0039] Specifically, the specific steps of the immersion server and the upper and lower layered heat exchange structure are as follows: the server mainboard is arranged in a horizontal paving manner to avoid bubble accumulation on the surface and to completely immerse the mainboard in low-boiling-point cooling liquid, and the cooling liquid directly absorbs the heat of the mainboard by using the heat vaporization process of the cooling liquid; a phase change energy storage material cold plate is arranged above the server mainboard to quickly liquefy the bubbles generated by the vaporized cooling liquid after releasing latent heat, forming an efficient gas-liquid conversion interface; a cold water pipeline arrangement structure is arranged below the cold plate to absorb the heat conducted downward by the phase change cold plate through cold water, realizing heat transfer at the liquid-solid interface; the cold water pipeline is connected to an external plate heat exchanger to build a closed loop, and the heat-conducting cold water is guided to industrial and domestic heat scenes for waste heat utilization; the cooling water flow and the effective heat exchange area of the heat exchanger are dynamically adjusted according to the real-time heat flow intensity during the heat exchange process to enhance the overall heat exchange capacity; and the overall immersion arrangement and the upper and lower layered structure jointly build a structure basis that can be used to characterize the cold plate heat conduction characteristics.

[0040] In the embodiment, by constructing the immersion server and the upper and lower layered heat exchange structure, a continuous conduction process of heat release from the mainboard to external recovery is realized. The cooling liquid directly contacts the mainboard to absorb heat and vaporize, the upper cold plate promotes the rapid liquefaction of the bubbles, the lower cold water pipeline timely takes away the conduction heat, and the heat energy is guided to the external scene through the closed loop. The overall structure realizes the effective connection of cooling, heat conduction and waste heat utilization, and provides physical support for heat conduction state monitoring and control strategy execution.

[0041] Specifically, the specific steps of the cold plate heat conduction release state are as follows: a representative cooling stable period is constructed by selecting a time interval with small dryness fluctuation, stable liquid supply and constant temperature, which is used as a sample basis for subsequent heat conduction state analysis; non-zero samples of the reference dryness are extracted in multiple cooling stable periods, a square value sequence is constructed, the minimum non-zero item is selected, scaling processing is performed combined with the normalization accuracy boundary, and a temperature difference compensation constant for formula calculation is obtained; the thermal conductivity parameter table corresponding to the selected PCM material is consulted, and its stable heat transfer capacity is extracted as the phase change heat conduction capacity value in the calculation; the server mainboard surface structure image is obtained, the cold plate arrangement size information is combined, the image projection overlap algorithm is used to measure the two-dimensional area where the cold plate and the mainboard form direct contact, and then the effective heat exchange area is obtained; the cooling liquid temperature and the cold plate surface temperature are retrieved from the cooling working condition data, the difference between the two is calculated as the numerator, the sum of the cold plate surface temperature and the temperature difference compensation constant is calculated as the denominator, and the ratio is obtained; the ratio is multiplied by the product of the phase change heat conduction capacity value and the 0.8 power of the effective heat exchange area, and finally the cold plate heat conduction output value is calculated.

[0042] The specific calculation method of the cold plate heat conduction output value is as follows:

[0043]

[0044] In the formula, This indicates the thermal conductivity output value of the cold plate. Indicates the coolant temperature. Indicates the surface temperature of the cold plate. This represents the temperature difference compensation constant. This indicates the phase change thermal conductivity value. This indicates the effective heat exchange area.

[0045] Table 1 shows the thermal conductivity output values ​​of the cold plate provided in the embodiments of this application. The coolant temperature of thermal conductivity 1 is set to 50.00°C, the surface temperature of the cold plate is set to 47.00°C, the phase change thermal conductivity value is set to 1.20, and the effective heat exchange area is set to 1.00; the coolant temperature of thermal conductivity 2 is set to 55.00°C, the surface temperature of the cold plate is set to 46.00°C, the phase change thermal conductivity value is set to 1.50, and the effective heat exchange area is set to 1.20; the coolant temperature of thermal conductivity 3 is set to 60.00°C, and the surface temperature of the cold plate is set to 45°C. 00, the phase change thermal conductivity value is set to 1.80, and the effective heat exchange area is set to 1.50; the coolant temperature of thermal conductivity 4 is set to 65.00, the cold plate surface temperature is set to 44.00, the phase change thermal conductivity value is set to 2.00, and the effective heat exchange area is set to 1.70; the coolant temperature of thermal conductivity 5 is set to 70.00, the cold plate surface temperature is set to 42.00, the phase change thermal conductivity value is set to 2.20, and the effective heat exchange area is set to 2.00.

[0046] Table 1. Data on thermal conductivity output of cold plate

[0047]

[0048] like Figure 3 The figure shows the distribution of thermal conductivity output values ​​of the cold plate provided in this embodiment of the application. According to the data in the image and table, the set primary thermal conductivity threshold is 0.90, the secondary thermal conductivity threshold is 1.80, and the introduced temperature difference compensation constant is 0.01. The thermal conductivity output values ​​corresponding to the five groups of thermal conductivity numbers are: thermal conductivity 1 is 0.0766, thermal conductivity 2 is 0.3395, thermal conductivity 3 is 0.8297, thermal conductivity 4 is 1.4590, and thermal conductivity 5 is 2.5530. From the distribution trend, the thermal conductivity output values ​​generally show an increasing relationship. Thermal conductivity 1 to 3 are below the primary thermal conductivity threshold, and are judged to be in a thermal conductivity lag state; thermal conductivity 4 is between the primary and secondary thresholds, in a stable range; thermal conductivity 5 exceeds the secondary thermal conductivity threshold, belonging to a high-response state. This figure can be used to intuitively judge the current thermal conductivity level of each area, providing a basis for determining whether to enter the dryness-linkage control process.

[0049] In the embodiment, the temperature difference compensation parameters are constructed by extracting the dryness samples in the cooling stability cycle, and the heat conduction capacity of the cold plate per unit time is quantitatively calculated by combining the heat conduction characteristics of the PCM material and the contact area of the cold plate, and the temperature difference between the cooling liquid and the cold plate. The obtained cold plate heat conduction output value can be used as a key indicator to measure the current heat conduction state, and provides an accurate basis for determining whether to enter the regulation process and the execution intensity.

[0050] Specifically, the specific steps of determining whether to enter the dryness linkage regulation process are as follows: comparing the real-time calculated cold plate heat conduction output value with the heat conduction threshold value, the heat conduction threshold value including a first-level heat conduction threshold value and a second-level heat conduction threshold value, which is used to divide the level interval of the regional heat conduction state; when the heat conduction output value is greater than or equal to the second-level heat conduction threshold value, it is determined that the region is in a high response state, indicating that the cold plate heat conduction capacity is sufficient and the cooling intensity is strong, and the dryness linkage regulation process needs to be executed to suppress the intensity; when the heat conduction output value is greater than or equal to the first-level heat conduction threshold value and less than the second-level heat conduction threshold value, it is determined that the regional heat conduction state is stable, and there is no dryness regulation trigger condition, so the current control parameters remain unchanged; when the heat conduction output value is lower than the first-level heat conduction threshold value, it is determined that there is a heat conduction lag risk in the region, and the dryness linkage regulation process is entered to enhance the local heat conduction capacity.

[0051] In the embodiment, by comparing the cold plate heat conduction output value with the set double threshold value in real time, the level to which the current regional heat conduction state belongs is identified, and the accurate division of the three states of high response, stability and heat conduction lag is realized. Whether to enter the dryness linkage regulation process can be determined, and the start or not of the regulation strategy is controlled accordingly, so as to realize the hierarchical triggering and regional execution of the regulation process, and improve the pertinence and efficiency of the cooling control.

[0052] Specifically, in the dryness linkage regulation process, the specific steps of determining the cooling flow adjustment range according to the dryness deviation are as follows: the cold plate heat release state identified as high response and heat lag is taken as input into the dryness linkage regulation process, and the target liquid supply amount is dynamically calculated according to the deviation between the current dryness and the reference dryness; during the cooling process, in the running stage where the determination region is in a stable state and has not entered the dryness linkage regulation process, a time period in which the dryness, the liquid supply flow rate and the cold plate temperature change range do not exceed the corresponding single sampling precision is selected, the dryness sensor original data in the selected time period is extracted and the mean value is calculated as the reference dryness value, wherein the single sampling precision represents the minimum numerical value change unit that can be recognized by the sensor in one effective sampling; the running samples in the regulation stable stage are traced back, all non-zero reference dryness data are extracted, the smallest effective item is selected, and the smallest resolution unit of the dryness sensor is multiplied to generate a deviation stable compensation value; the current real-time dryness value and the basic liquid supply flow rate value are obtained; the real-time dryness value is subtracted from the reference dryness value, squared, multiplied by the dryness deviation adjustment coefficient as the numerator, the dryness deviation adjustment coefficient is obtained by constructing a nonlinear function between the dryness change rate and the response time delay, and the value range is generally between 0.1 and 0.5; the reference dryness square value is added to the deviation stable compensation value as the denominator; the numerator is divided by the denominator to obtain a fractional value, which is added to one to form a bracket as a whole; the bracket is multiplied by the basic liquid supply flow rate value to obtain the final liquid supply adjustment response value.

[0053] The specific calculation method of the liquid supply adjustment response value is:

[0054]

[0055] In the formula, the liquid supply adjustment response value, the real-time dryness value, the reference dryness value, the dryness deviation adjustment coefficient, the deviation stable compensation value, the basic liquid supply flow rate value.

[0056] In the embodiment, by constructing a calculation formula with dryness deviation as the core, combining reference dryness, compensation value, adjustment coefficient and other parameters, the dynamic calculation of the cooling flow adjustment range is realized. The liquid supply adjustment response value can be automatically adjusted according to the dryness change degree, has the characteristics of continuity and difference, and is helpful to realize the accurate matching of cooling capacity to heat state and enhance the independent regulation and control ability of high response and lag area.

[0057] Specifically, the specific steps of controlling the opening degree of the electromagnetic valve, the pump control frequency and the heat exchange structure configuration according to the cooling flow adjustment amplitude are as follows: a liquid supply response grading efficiency strategy is executed according to the liquid supply adjustment response value, and the response value is compared with the dryness deviation tolerance threshold in real time; when the liquid supply adjustment response value is greater than or equal to the threshold, the opening degree of the cooling liquid electromagnetic valve is adjusted according to the relative amplitude between the response value and the threshold, and the liquid supply change amplitude value is calculated according to the difference between the actual liquid supply flow in the current sampling period and the actual liquid supply flow in the last sampling period; the cooling pump operating frequency is increased according to the change amplitude, and the minimum operating period is set to prevent frequent start-stop, the minimum operating period being the shortest duration for which the cooling pump maintains constant output in one flow adjustment, and being used to avoid equipment jitter and flow disturbance; the number of active plates in the heat exchanger area is checked synchronously, and if the maximum configuration is not reached, the range to be enabled is expanded, and the liquid supply branch flow in the dryness rising area is positively corrected; when the liquid supply adjustment response value is lower than the dryness deviation tolerance threshold, the current opening degree of the electromagnetic valve, the pump control frequency and the heat exchange plate configuration are kept unchanged.

[0058] In the embodiment, by adjusting the opening degree of the electromagnetic valve, the pump control frequency and the heat exchange structure configuration, fine regulation of the cooling flow under different heat conduction states is realized. The minimum operating period is set to avoid flow disturbance caused by frequent start-stop, and to enhance the continuity and stability of the liquid supply process. The overall control strategy has the characteristics of response grading and adjustment, which can reduce unnecessary resource fluctuations while ensuring heat exchange efficiency, and improve the execution reliability of the cooling regulation process.

[0059] Specifically, the specific steps of identifying the regulation response by combining the dryness and temperature change trend in the regulation process are as follows: the deviation between the cooling result and the regulation response is tracked to optimize the actual execution effect of the regulation strategy; the real-time dryness value recorded in the SCADA system and the corresponding sampling time point are called, and the dryness value and the sampling time at the ith moment and the (i-1)th moment are obtained, which are respectively recorded as the ith moment dryness value, the (i-1)th moment dryness value, the ith moment sampling time point and the (i-1)th moment sampling time point; the time interval between each frame in the continuous dryness collection record is extracted, the smallest non-zero value among them is selected, and the minimum time resolution granularity of the SCADA system is multiplied to generate a time difference compensation constant for avoiding division by zero error; difference calculation is performed on the continuous dryness frame value to form a dryness change data sequence, and multiple groups of sequences are constructed according to different continuous frame numbers, the smooth error and response delay trend of each group are evaluated, and the smallest effective frame number with the smallest error and no response lag is selected as the statistical sample number; the dryness change rate calculation is performed in groups with the statistical sample number as the number of cycles: each group first obtains the dryness change amount by subtracting the dryness value of the previous frame from the ith frame, and then obtains the corresponding time interval value by adding the time difference compensation constant to the difference between the two frame sampling time points; the dryness change amount is divided by the time interval to obtain the dryness change rate per unit time; finally, the dryness change rates of all samples are summed and averaged to obtain the dryness fluctuation rate value.

[0060] The specific calculation method for the dryness fluctuation rate value is as follows:

[0061]

[0062] In the formula, This represents the rate of change in dryness. This represents the dryness value at time i. This represents the dryness value at time i-1. This represents the sampling time point at time i. This represents the sampling time point at time i-1. Represents the time difference compensation constant. This indicates the number of statistical samples.

[0063] like Figure 4 The figure shows the evaluation curve of the control effect provided in the embodiment of this application. The figure shows the change trend of the thermal conductivity output value of the cold plate at 8 consecutive time points in three stages: before control (cycle 1), during control (cycle 2), and after control (cycle 3). The set primary thermal conductivity threshold is 0.90, and the secondary thermal conductivity threshold is 1.80. Before control, the curve is generally lower than the primary thermal conductivity threshold, and the thermal conductivity is in a lagging state; during control, the curve gradually rises and gradually approaches the secondary threshold at time points 7 and 8, indicating that the control process is enhancing the regional thermal conductivity performance; after control, the curve remains between the primary and secondary thresholds throughout, with a peak value exceeding 1.6, showing strong thermal conductivity stability and control response effect. This figure can be used to compare the thermal conductivity state of each cycle and evaluate the execution effect and stability improvement of the control measures in the time dimension.

[0064] In this implementation scheme, a dryness fluctuation rate value is constructed to measure the intensity of cooling response fluctuations by continuously calculating the rate of dryness change. The timing accuracy of the rate calculation is ensured by combining the sampling time point and the time difference compensation constant, and the optimal statistical sample is selected through comparison of multiple sets of frames to improve the smoothness and real-time performance of the results. The final dryness fluctuation rate value can be used to reflect the response stability in the current cooling process, providing a quantitative basis for the dynamic adjustment of the control rhythm.

[0065] Specifically, the specific steps of adjusting the liquid supply calculation frequency, valve control step, and heat exchange region scheduling level are as follows: according to the dryness fluctuation rate value, a dryness disturbance grading slow-release strategy is executed, and the rate value is compared with the first and second fluctuation thresholds in real time, serving as the basis for adjusting the rhythm and execution intensity; when the dryness fluctuation rate value is lower than the first threshold, it is determined that the cooling process is stable, the calculation frequency of the liquid supply adjustment response value is kept unchanged, and the electromagnetic valve opening is directly controlled according to the relationship between the cold plate heat conduction output value and the liquid supply flow; when the dryness fluctuation rate value is between the first and second thresholds, the update frequency of the liquid supply adjustment response value is reduced, the electromagnetic valve opening is controlled within the minimum allowed adjustment step based on the current value, the liquid supply redistribution operation between different regions is cancelled, and the sampling density of the cold plate temperature sensor is increased to update the cold plate heat conduction output value in real time; when the dryness fluctuation rate value is greater than or equal to the second threshold, it is determined that the control response is intense, the calculation process of the liquid supply adjustment response value is suspended, the current electromagnetic valve opening is kept unchanged, the flow dynamic adjustment channel of the relevant heat exchanger partition is closed, and the region control execution level is set to the lowest.

[0066] In the embodiment, through the grading judgment of the dryness fluctuation rate value, dynamic adjustment of the liquid supply calculation frequency, electromagnetic valve control step, and heat exchange region scheduling level is realized. When the fluctuation is small, the original rhythm and opening control are maintained, the adjustment amplitude is tightened and the sensor sampling density is increased in the medium fluctuation state, the adjustment process is suspended and the related region control authority is frozen in the intense fluctuation state, ensuring that the cooling process has a flexible and stable response mechanism under different disturbance levels.

[0067] As Figure 2As shown, the structural schematic diagram of the data center immersion cooling system based on dryness sensing cooperative regulation provided by the embodiment of the application, the data center immersion cooling system based on dryness sensing cooperative regulation provided by the embodiment of the application, the data center immersion cooling method based on dryness sensing cooperative regulation is applied, comprising: a cooling data acquisition module, a cooling heat exchange structure module, a dryness linkage regulation module and a state feedback cooperation module: the cooling data acquisition module is used for arranging multiple sensors in the immersion liquid cooling environment, acquiring the key physical parameters of the cooling liquid temperature, the cold plate temperature, the dryness value and the liquid supply flow rate, obtaining the cooling working condition acquisition data and completing the filtering, completion and standardization processing; the cooling heat exchange structure module is used for combining the horizontally arranged immersion server and the upper and lower layered cold plate-water cooling structure, identifying the heat conduction state of the cold plate and judging whether to enter the dryness linkage regulation process according to the heat conduction output value and the double threshold value rule; the dryness linkage regulation module is used for calculating the cooling flow regulation amplitude according to the deviation between the current dryness and the reference dryness after entering the regulation process, and controlling the electromagnetic valve opening degree, the pump control frequency and the heat exchanger plate enabling range according to the control; the state feedback cooperation module is used for tracking the dryness and temperature change trend, identifying the regulation response stability, and adjusting the liquid supply calculation frequency, the valve control step and the heat exchange region scheduling level according to the dryness fluctuation rate value.

[0068] As Figure 2 shown, the system of the application is composed of a cooling data acquisition module, a cooling heat exchange structure module, a dryness linkage regulation module and a state feedback cooperation module, and each module forms a cooperative relationship around the dryness change in the immersion liquid cooling environment. The cooling data acquisition module is responsible for obtaining the temperature, dryness and liquid supply state basic information and providing uniform input to the subsequent modules; the cooling heat exchange structure module identifies the heat conduction state based on the server immersion arrangement and the layered heat exchange structure; the dryness linkage regulation module adjusts and controls the liquid supply execution unit when the trigger condition is met; the state feedback cooperation module analyzes the dryness and temperature change trend after regulation, and the results are used to update the regulation parameters, so that the system maintains stable linkage regulation ability in the running process.

[0069] In the embodiment, by constructing a multi-module structure with data acquisition, heat conduction identification, linkage regulation and feedback correction as the core, the closed-loop control of the cooling process from physical perception to regulation execution is realized. The continuous logic is formed among the modules through data driving and state linkage, which can dynamically match the cooling intensity under different heat conduction states, and improve the accuracy, response speed and local adaptability of the regulation.

[0070] As Figure 5The diagram shows a design of a two-phase immersion liquid-cooled PCM module for a data center according to an embodiment of this application. In this structure, server motherboards are stacked horizontally and submerged in a low-boiling-point coolant. The coolant vaporizes upon heating and rises, entering the PCM module area at the top of the diagram. The vaporized coolant releases latent heat, which is absorbed by the PCM material, completing the phase change process. Under gravity, it re-liquefies and falls back below the liquid surface, forming a closed coolant circulation loop. A heat exchange unit 1 is located above the PCM module to further recover the phase change heat. Since only one heat exchange unit is labeled in the diagram, the number "1" is for illustrative purposes only. In the actual structure, multiple heat exchange units can be configured according to heat dissipation requirements and are not repeatedly shown in the diagram. The entire structure, through its layered design, achieves efficient coupling of the coolant's phase change heat absorption, gas-liquid circulation, and heat conduction processes, improving heat dissipation performance and realizing waste heat utilization. This provides a physical foundation for subsequent cold plate thermal conductivity identification and dryness control.

[0071] like Figure 6 The diagram shown illustrates the working principle of the two-phase immersion liquid-cooled PCM data center cooling system provided in this embodiment. The system uses the data center unit module as the core heat source. The cooling process involves a circulating pump driving the coolant through a flow pump and sensor unit into the unit module, absorbing heat from the server motherboard and PCM cold plate. The heated coolant then enters the compressor, where its temperature and pressure are increased before being sent to the heat exchanger module, where it releases heat. The released heat energy is guided by the heat exchanger to the terminal heating module, enabling energy utilization for building heating and domestic heat consumption. The cooled refrigerant is then returned to the data center cooling module by the circulating pump, forming a closed cooling loop. Sensors can collect cooling status parameters in real time and upload the data to a management terminal for adjusting the cooling flow rate and heat distribution strategy, achieving coordinated operation of cooling and waste heat recovery.

[0072] like Figure 7 The diagram shows the system structure of a two-phase immersion PCM module for a data center according to an embodiment of this application. The structure adopts a multi-layer stacked arrangement, with multiple server motherboards horizontally placed within the same frame, evenly arranged vertically, and the entire system is immersable in coolant. Each layer forms an independent cooling space, facilitating the vaporization, rising, condensation, and settling of the two-phase coolant within a localized area. Multiple openings are provided on the periphery of the structure to facilitate coolant flow and external heat exchange pipe interfaces, ensuring sufficient contact and continuous flow during heat exchange. This module features a simple structure and strong scalability, serving as a standardized thermal control unit for coupled two-phase liquid cooling and PCM cold plate operation. It is suitable for zoned cooling and thermal conductivity monitoring applications under high heat flux density conditions in data centers.

[0073] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting; it is not intended to exclude myriad other embodiments of the present application that other presenters can develop. It is also possible, however, that only a single element can be present. It is further noted that such a term as "comprising" is intended to mean that the embodiments include the recited elements, but not excluding other elements. "Consisting essentially of when used herein in relation to a composition, means that the composition includes the recited elements, and can include additional elements, so long as the additional elements do not materially alter the basic and novel properties of the claimed composition. "Consisting of" when used herein in relation to a composition means that the composition includes the recited elements and nothing more.

[0074] The preferred embodiments of the application disclosed above are only to help explain the principles of the present application. The preferred embodiments do not describe all the details of the present application, nor limit the present application to only the specific embodiments described. It is apparent that many modifications and variations can be made to the present application based on the content of the present disclosure. The present disclosure selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can well understand and utilize the present application. The present application is limited only by the claims and their full scope and equivalents.

Claims

1. A method of data center immersion cooling based on dryness sensing collaborative regulation, characterized in that, The method comprises the following steps: S1: Collecting key physical parameters in the immersed liquid cooling environment, obtaining cooling working condition collection data and performing preprocessing; S2: Based on the immersed server and the upper and lower layered heat exchange structure, the heat conduction release state of the cold plate is characterized, and it is judged whether to enter the dryness linkage regulation process; S3: In the dryness linkage regulation process, the cooling flow adjustment range is determined according to the dryness deviation, and the electromagnetic valve opening degree, pump control frequency and heat exchange structure configuration are controlled according to the cooling flow adjustment range; In the dryness linkage regulation process, the cooling flow adjustment range is determined according to the dryness deviation, and the specific steps are as follows: The high response and heat conduction hysteresis of the cold plate heat conduction release state are input into the dryness linkage regulation process, and the target liquid supply amount of the cooling water is dynamically calculated according to the deviation between the current dryness and the reference dryness; In the cooling process, in the running stage where the determination area is in a stable state and does not enter the dryness linkage regulation process, a time period in which the dryness, liquid supply flow rate and cold plate temperature change amplitude do not exceed the corresponding single sampling accuracy is selected, the original data of the dryness sensor in the time period is extracted and the average value is calculated to form a reference dryness value; the cooling running sample of the dryness regulation keeping stable is traced back, all non-zero reference dryness values are extracted and the smallest effective item is selected, the value is multiplied by the smallest resolution unit of the dryness sensor to obtain a deviation stable compensation value; the real-time dryness value and the basic liquid supply flow rate value are obtained; The real-time dryness value is subtracted from the reference dryness value, the difference is squared, multiplied by the dryness deviation adjustment coefficient, and used as the numerator; The reference dryness is squared and added to the deviation stable compensation value to be used as the denominator; divide the numerator by the denominator to get a fraction value; add one to the fraction value to form the whole in the parentheses; multiply the result in the parentheses by the basic liquid supply flow rate value to get the liquid supply adjustment response value; S4: Identify the regulation response according to the dryness and temperature change trend in the regulation process, and adjust the liquid supply calculation frequency, valve control step and heat exchange region scheduling level according to the situation.

2. The data center immersion cooling method based on dryness sensing collaborative regulation according to claim 1, characterized in that: The specific steps of collecting key physical parameters in the immersed liquid cooling environment, obtaining cooling working condition collection data and performing preprocessing are as follows: The process of collecting key physical parameters in the immersed liquid cooling environment and obtaining cooling working condition collection data includes: arranging temperature sensors in the two-phase cooling liquid area to collect cooling liquid temperature; setting thermocouple measuring points above the PCM cold plate to obtain the cold plate surface temperature; arranging dryness sensors in the cooling liquid vaporization core area to generate real-time dryness values by sensing the gas-liquid phase state ratio; installing a flow sensor at the liquid supply inlet section to read the initial liquid rate and form a basic liquid supply flow rate value; extracting the corresponding time field from the SCADA system to determine the sampling time point; finally, cooling working condition collection data covering the cooling state, heat conduction response and flow reference are formed; The cooling working condition collection data is sequentially smoothed and noise suppressed by the sliding median filtering algorithm and the local linear regression method; the field missing problem caused by sampling delay and short interruption is solved by time step alignment rule and linear interpolation method; the processed cooling working condition collection data is standardized and normalized.

3. The method of claim 1, wherein the method further comprises: The specific steps of the immersion-based server and the upper and lower layered heat exchange structure are as follows: The server mainboard is arranged in a horizontal flat manner and is completely immersed in low-boiling cooling liquid, and heat absorption is completed through the heated vaporization of the cooling liquid; a phase change energy storage material cold plate is arranged above the server mainboard, so that the bubbles generated by the vaporization of the cooling liquid are quickly liquefied after releasing latent heat; A cold water pipeline arrangement structure is arranged below the cold plate, and the heat conducted downward by the phase change cold plate is absorbed by the cold water; the cold water pipeline is connected to an external plate heat exchanger, and the heat-conducting cold water is introduced to external industrial and domestic heat scenes through a closed loop; the heat exchanger flux is dynamically adjusted according to the heat flow intensity during the heat exchange process; the above immersion arrangement of the server and the layered form of the heat exchange structure are taken as the structural basis for releasing the characteristics of the sensing cold plate during heat conduction.

4. The dryness-sensing coordinated regulation based data center immersion cooling method according to claim 1, characterized in that: The specific steps of the characteristic cold plate heat conduction release state are as follows: A cooling stable period is formed by screening a time interval with small dryness fluctuation, stable liquid supply and constant temperature; a square value sequence is constructed by extracting non-zero samples of the reference dryness in multiple cooling stable periods, and the minimum non-zero item is selected; scaling is performed in combination with the normalized accuracy boundary to obtain a temperature difference compensation constant; the phase change heat conduction capacity value of the selected PCM material is obtained by referring to the thermal conductivity parameter table; the server mainboard surface structure image is obtained, and the direct contact area between the cold plate and the mainboard is calculated by using an image projection overlap algorithm combined with the cold plate arrangement size to obtain the effective heat exchange area; the cooling liquid temperature and the cold plate surface temperature are obtained from the cooling working condition data; The difference between the cooling liquid temperature and the cold plate surface temperature is taken as the numerator, and the sum of the cold plate surface temperature and the temperature difference compensation constant is taken as the denominator, and a ratio is obtained by dividing the numerator by the denominator; a product is obtained by multiplying the 0.8th power of the phase change heat conduction capacity value and the effective heat exchange area; the ratio and the product are multiplied to obtain the cold plate heat conduction output value.

5. The dryness-sensing coordinated regulation based data center immersion cooling method according to claim 1, characterized in that: The specific steps of the dryness linkage regulation process are as follows: The cold plate heat conduction output value is compared with the heat conduction threshold value in real time, and the heat conduction threshold value includes a first-level heat conduction threshold value and a second-level heat conduction threshold value; When the cold plate heat conduction output value is greater than or equal to the second-level heat conduction threshold value, it is determined that the region is in a high response state, and the dryness linkage regulation process is entered; When the cold plate heat conduction output value is greater than or equal to the first-level heat conduction threshold value and less than the second-level heat conduction threshold value, it is determined that the region is in a stable state, and the dryness linkage regulation process is not entered; When the cold plate heat conduction output value is less than the first-level heat conduction threshold value, it is determined that the region is in a heat conduction lag state, and the dryness linkage regulation process is entered.

6. The dryness-sensing coordinated regulation based data center immersion cooling method according to claim 1, characterized in that: The specific steps of controlling the electromagnetic valve opening degree, pump control frequency and heat exchange structure configuration according to the cooling flow regulation amplitude are as follows: According to the liquid supply adjustment response value, a liquid supply response hierarchical enhancement strategy is executed: the liquid supply adjustment response value is compared with the dryness deviation allowable threshold value in real time; when the liquid supply adjustment response value is greater than or equal to the dryness deviation allowable threshold value, the cooling liquid electromagnetic valve opening degree is adjusted according to the relative amplitude between the liquid supply adjustment response value and the dryness deviation allowable threshold value, the liquid supply change amplitude value is calculated according to the difference between the actual liquid supply flow in the current sampling period and the actual liquid supply flow in the last sampling period, the cooling pump operating frequency is increased according to the liquid supply change amplitude value and the minimum operating period is limited, the number of active plates in the heat exchanger region is checked and the enabled range is expanded when the maximum value is not reached, and the positive correction is performed on the liquid supply branch flow in the dryness rising region; when the liquid supply adjustment response value is less than the dryness deviation allowable threshold value, the cooling liquid electromagnetic valve opening degree is maintained at the current value, the cooling pump operating frequency remains unchanged, and the heat exchanger plate configuration remains unchanged.

7. The dryness-sensing coordinated regulation based data center immersion cooling method according to claim 1, characterized in that: The specific steps of identifying the regulation response according to the dryness and temperature change trend in the regulation process are as follows: Tracking the deviation between the cooling result and the regulation response, and optimizing the regulation strategy execution; calling the real-time dryness value and the sampling time point; obtaining the real-time dryness value and the sampling time point at the ith moment and the (i-1)th moment, and recording them as the dryness value at the ith moment, the dryness value at the (i-1)th moment, the sampling time point at the ith moment and the sampling time point at the (i-1)th moment respectively; Extracting the time interval between every two frames in the continuous dryness collection record, selecting the minimum non-zero value therefrom, multiplying the minimum time resolution granularity of the SCADA system to form a time difference compensation constant; performing difference calculation between the adjacent frame dryness values in the continuous dryness collection record to form a dryness change data sequence, constructing a dryness change data sequence under a plurality of different continuous frame numbers, evaluating the smooth error and response delay change trend of each sequence, and selecting the minimum effective frame number with the minimum error and no response lag as the statistical sample number; Taking the statistical sample number as the total number of cycles, in each group of continuous sampling data, first subtract the dryness value at the (i-1)th moment from the dryness value at the ith moment to obtain the dryness change amount; then subtract the sampling time point at the (i-1)th moment from the sampling time point at the ith moment, and add the time difference compensation constant to the difference to form the corresponding time interval value; divide the dryness change amount by the corresponding time interval value to calculate the unit time dryness change rate corresponding to the ith group of samples; after the calculation of all n groups of samples is completed, sum the n dryness change rates and divide by the statistical sample number to obtain the dryness fluctuation rate value after averaging all samples.

8. The dryness-sensing coordinated regulation based data center immersion cooling method according to claim 1, characterized in that: The specific steps of adjusting the liquid supply calculation frequency, valve control step and heat exchange region scheduling level according to the situation are as follows: According to the dryness fluctuation rate value, a dryness disturbance hierarchical release strategy is executed: the dryness fluctuation rate value is compared with the fluctuation threshold value in real time, and the fluctuation threshold value includes a first-level fluctuation threshold value and a second-level fluctuation threshold value; When the dryness fluctuation rate value is less than the first-level fluctuation threshold, the calculation frequency of the liquid supply adjustment response value is kept unchanged, the electromagnetic valve opening degree is controlled according to the corresponding relationship between the cold plate heat conduction output value and the liquid supply flow; when the dryness fluctuation rate value is greater than or equal to the first-level fluctuation threshold and less than the second-level fluctuation threshold, the update frequency of the liquid supply adjustment response value is reduced, the electromagnetic valve opening degree is limited in the minimum allowed adjustment step range based on the current value, the liquid supply redistribution between the cooling regions is cancelled, the sampling density of the cold plate temperature sensor is improved, and the cold plate heat conduction output value is updated; When the dryness fluctuation rate value is greater than or equal to the second-level fluctuation threshold, the calculation process of the liquid supply adjustment response value is suspended, the electromagnetic valve opening degree is kept as the current value, the flow dynamic adjustment channel of the corresponding heat exchanger partition is closed, and the execution level is set to the lowest.

9. A data center immersion cooling system based on dryness sensing collaborative regulation, applying the data center immersion cooling method based on dryness sensing collaborative regulation of any one of claims 1-8, comprising: The cooling data acquisition module, the cooling heat exchange structure module, the dryness linkage control module and the state feedback cooperation module are characterized in that: The cooling data acquisition module is used for acquiring key physical parameters in the immersed liquid cooling environment, obtaining cooling working condition acquisition data and performing preprocessing; The cooling heat exchange structure module is used for representing the cold plate heat conduction release state based on the immersed server and the upper and lower layered heat exchange structure, and judging whether to enter the dryness linkage control process; The dryness linkage control module is used for determining the cooling flow adjustment amplitude according to the dryness deviation in the dryness linkage control process, controlling the electromagnetic valve opening degree, the pump control frequency and the heat exchange structure configuration according to the cooling flow adjustment amplitude; The state feedback cooperation module is used for identifying the control response condition in combination with the dryness and temperature change trend in the control process, and adjusting the liquid supply calculation frequency, the valve control step and the heat exchange region scheduling level according to different conditions.

Images