Adaptive control method, device, equipment and storage medium of liquid cooling system

Through real-time monitoring and adaptive fluid dynamics simulation, the flow rate and workload of the liquid cooling system are adjusted, and the response hysteresis problem of single-phase immersion liquid cooling cabinet when the system is instable is solved, and the heat exchange efficiency and equipment safety are improved. Ensures the long-term and stable operation of the system.

CN118765100BActive Publication Date: 2024-11-15STATE GRID ZHEJIANG ELECTRIC POWER CO LTD HANGZHOU POWER SUPPLY CO +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411238355.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-05
Publication Date
2024-11-15
Estimated Expiration
2044-09-05

AI Technical Summary

Technical Problem

When the system is instable, the flow rate adjustment response is lagging, making it difficult to deal with the problem of excessive bubbles in a timely manner, resulting in a reduced heat exchange efficiency and may cause the risk of equipment performance degradation or damage.

Method used

Through real-time monitoring of data input risk assessment models, potential instability areas are identified, adaptive fluid dynamics simulations are performed, flow rate is adjusted to optimize heat exchange efficiency, and flow rate and workload are adjusted through flow instability analysis to ensure system stability and response speed and speed are ensured through flow instability analysis. .

Benefits of technology

The rapid response and stability of the liquid cooling system are achieved, and the performance degradation or risk of equipment damage caused by local overheating is avoided, long-term efficient and stable operation is ensured, and the reliability and safety of overall operation are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118765100B_ABST
    Figure CN118765100B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of liquid cooling control technology, and discloses an adaptive control method, device, equipment and storage medium for a liquid cooling system, including obtaining real-time monitoring data of the liquid cooling system, and inputting the risk assessment model to obtain a risk assessment index; judging whether the risk assessment index is greater than a risk threshold, and if so, determining a potential instability region based on the real-time monitoring data; obtaining first fluid dynamics data of the potential instability region, performing adaptive fluid dynamics simulation, and obtaining an optimal flow rate; judging whether the current flow rate of the liquid cooling system is consistent with the optimal flow rate, and if not, adjusting the current flow rate based on the optimal flow rate; obtaining second fluid dynamics data of the potential instability region after the flow rate adjustment, performing flow instability analysis, and adjusting the liquid cooling system based on the analysis results. The present invention not only effectively improves the stability and thermal management efficiency of the liquid cooling system, but also improves the reliability and safety of the overall operation of the liquid cooling system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of liquid cooling control technology, and in particular to an adaptive control method, device, equipment and storage medium for a liquid cooling system. Background Art

[0002] In single-phase immersion liquid cooling cabinets used in data centers, it is a challenge to maintain system stability while ensuring thermal management efficiency. The key task in liquid cooling systems is to maximize heat exchange and control bubble generation through fluid flow rate management to optimize heat transfer efficiency. However, once the system becomes unstable, such as excessive bubbles caused by local overheating, this may greatly affect the heat exchange efficiency, causing the temperature to rise, which in turn brings the risk of equipment performance degradation or even damage. To this end, the system needs to use complex predictive models to identify and prevent conditions that may lead to instability, such as sudden load changes and ambient temperature changes.

[0003] At present, the problem with this treatment method is that these models often consume computing resources and have a long response time. When the system detects the risk of instability and prompts the need to adjust the flow rate, the actual system may not be able to adapt to these changes in time due to the lag in the model's response. Even if an adaptive control method is adopted, it is difficult to completely eliminate the impact of instability. In addition to the possibility of failing to effectively reduce the risk of instability, this delayed response may further aggravate the problem due to improper bubble control. For example, the presence of bubbles may hinder heat transfer and cause local hot spots, while untimely flow rate adjustment may cause bubble aggregation or even the formation of a large number of bubbles, which will deteriorate the heat conduction environment. In other words, in the operation of single-phase immersion liquid cooling cabinets, there is a technical contradiction between system stability and fast response. On the one hand, complex prediction models are needed to evaluate potential instability risks, and on the other hand, the system needs to be able to adjust the flow rate and bubble control in real time to maintain efficient heat transfer. The uniqueness of this contradiction lies in the conflict between the real-time requirements of flow rate adjustment and bubble control and the complexity and response time of the prediction model, which determines the cooling efficiency of the entire system and the basis for safe operation of the equipment. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides an adaptive control method, device, equipment and storage medium for a liquid cooling system, so as to solve the technical contradiction between system stability and rapid response in the operation of a single-phase immersion liquid cooling cabinet, thereby achieving the technical effect of improving the system cooling efficiency and the safety of equipment operation.

[0005] In a first aspect, the present invention provides an adaptive control method for a liquid cooling system, the method comprising:

[0006] Obtain real-time monitoring data of the liquid cooling system and input it into a pre-established risk assessment model to obtain risk assessment indicators;

[0007] Determining whether the risk assessment index is greater than a risk threshold, and if so, determining a potential instability area in the liquid cooling system based on the real-time monitoring data;

[0008] Acquiring first fluid dynamics data of the potential instability region, and performing adaptive fluid dynamics simulation according to the first fluid dynamics data to obtain an optimal flow velocity;

[0009] Determine whether the current flow rate of the liquid cooling system is consistent with the optimal flow rate, and if not, adjust the current flow rate according to the optimal flow rate;

[0010] The second fluid dynamic data of the potential instability area after the flow velocity adjustment is obtained, a flow instability analysis is performed on the second fluid dynamic data, and the liquid cooling system is adjusted according to the analysis result.

[0011] Furthermore, the step of determining the potential instability area in the liquid cooling system according to the real-time monitoring data includes:

[0012] According to the geometric parameters of the liquid cooling system, the flow and heat transfer of the cooling liquid in the liquid cooling system are numerically simulated;

[0013] The real-time monitoring data is compared with the simulation results, and a potential unstable area in the liquid cooling system is determined according to the comparison results.

[0014] Furthermore, the step of obtaining first fluid dynamics data of the potential instability region, performing adaptive fluid dynamics simulation according to the first fluid dynamics data, and obtaining an optimal flow velocity comprises:

[0015] Acquiring first fluid dynamic data of the potential instability region and geometric parameters of the liquid cooling system, performing liquid flow simulation according to a turbulence model, and generating initial simulation results;

[0016] A heat transfer analysis is performed on the initial simulation results according to the heat transfer characteristics, and an optimal flow rate of the liquid cooling system is determined according to the analysis results.

[0017] Furthermore, the steps of obtaining the second fluid dynamic data of the potential instability area after the flow rate is adjusted, performing flow instability analysis on the second fluid dynamic data, and adjusting the liquid cooling system according to the analysis results include:

[0018] Acquire second fluid dynamic data of the potential instability region after the flow rate is adjusted, wherein the second fluid dynamic data includes flow rate data, bubble data and temperature data;

[0019] Performing flow stability analysis on the second fluid dynamic data to determine whether there is abnormal data;

[0020] If so, flow instability analysis is performed based on the correlation of each abnormal data, and the flow rate and / or workload of the liquid cooling system is adjusted based on the analysis results.

[0021] Furthermore, after the step of adjusting the liquid cooling system according to the analysis result, the method further includes:

[0022] Acquire the adjusted real-time temperature data and real-time pressure data of the liquid cooling system, and determine whether there is an abnormal area according to the real-time temperature data and the real-time pressure data, wherein the abnormal area includes a bubble accumulation area and / or a local overheating area;

[0023] If so, the flow rate of the liquid cooling system is adjusted, and based on the feedback loop, it is determined whether to continue adjusting the flow rate.

[0024] Furthermore, after the step of adjusting the liquid cooling system according to the analysis result, the method further includes:

[0025] The response time and performance data of the adjusted liquid cooling system are obtained, analysis of the performance data is triggered based on the response time, and dynamic resource allocation is performed on the liquid cooling system according to the analysis result.

[0026] Furthermore, after the step of adjusting the liquid cooling system according to the analysis result, the method further includes:

[0027] Obtaining performance data of the adjusted liquid cooling system and inputting the data into a pre-established performance prediction model to obtain a performance prediction result;

[0028] According to the performance prediction result, the control strategy of the liquid cooling system is optimized, and the optimization includes cooling liquid flow optimization and fan speed optimization.

[0029] In a second aspect, the present invention provides an adaptive control device for a liquid cooling system, the device comprising:

[0030] The risk assessment module is used to obtain the real-time monitoring data of the liquid cooling system and input the pre-established risk assessment model to obtain the risk assessment indicators;

[0031] an instability determination module, for determining whether the risk assessment index is greater than a risk threshold, and if so, determining a potential instability region in the liquid cooling system based on the real-time monitoring data;

[0032] A flow velocity simulation module, used for acquiring first fluid dynamics data of the potential instability region, and performing adaptive fluid dynamics simulation according to the first fluid dynamics data to obtain an optimal flow velocity;

[0033] A flow rate adjustment module, used to determine whether the current flow rate of the liquid cooling system is consistent with the optimal flow rate, and if not, to adjust the current flow rate according to the optimal flow rate;

[0034] The analysis and adjustment module is used to obtain the second fluid dynamic data of the potential instability area after the flow rate is adjusted, perform flow instability analysis on the second fluid dynamic data, and adjust the liquid cooling system according to the analysis results.

[0035] In a third aspect, an embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0036] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0037] The present invention provides an adaptive control method, device, equipment and storage medium for a liquid cooling system. Through real-time monitoring and dynamic adjustment, the present invention can timely identify and prevent conditions that may cause system instability, ensure that the equipment operates at an optimal temperature, thereby avoiding the risk of performance degradation or equipment damage caused by local overheating, and can quickly respond to changes under various working conditions while continuously self-optimizing, thereby ensuring that the liquid cooling system can operate efficiently and stably for a long time. The present invention can not only improve the stability and thermal management efficiency of the liquid cooling system, but also improve the reliability and safety of the overall operation of the liquid cooling system. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a flow chart of an adaptive control method for a liquid cooling system in an embodiment of the present invention;

[0039] Figure 2 is a schematic structural diagram of an adaptive control device for a liquid cooling system in an embodiment of the present invention;

[0040] Figure 3 It is a diagram of the internal structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1 The first embodiment of the present invention provides an adaptive control method for a liquid cooling system, which includes steps S10 to S50:

[0043] Step S10, acquiring real-time monitoring data of the liquid cooling system and inputting the data into a pre-established risk assessment model to obtain risk assessment indicators.

[0044] In this embodiment, a risk assessment model of the liquid cooling system is first constructed based on the historical data of the single-phase immersion liquid cooling system, and then the real-time monitoring data of the liquid cooling system is input into the risk assessment model to identify the patterns of temperature anomalies, flow rate changes, and bubble generation, thereby obtaining the current risk assessment index of the liquid cooling system. The data used to establish the risk assessment model include temperature data, flow rate data, pressure data, bubble data, operation data, and climate data.

[0045] Specifically, the training data set of the model is first constructed based on historical data, in which sensors are used to obtain key temperature indicators of the single-phase immersion liquid cooling system, including inlet temperature, outlet temperature, ambient temperature and temperature change rate data. According to the real-time flow rate data provided by the flow rate sensor, combined with the flow rate change rate, the flow condition of the coolant is judged. If the flow rate drops suddenly, it indicates that there is a blockage or the pump function is attenuated. The inlet pressure, outlet pressure and pressure change rate are obtained by the pressure sensor, and the pressure data is analyzed to identify potential leaks or pressure anomalies. If the bubble sensor detects an increase in the bubble generation rate, it is combined with the bubble size and distribution data to determine whether there is cavitation or bubble accumulation inside the liquid cooling system. Combined with the turbulence sensor and the vibration and noise level sensor data, the stability of the fluid flow and the mechanical integrity of the liquid cooling system are evaluated. Through the liquid cooling system operation data, including pump status, valve position, and coolant type and concentration information, it is determined whether the liquid cooling system is operating according to the predetermined parameters and the risk of operational deviation is predicted. If the ambient humidity and climate condition sensors show that environmental factors have changed, the potential impact of these changes on the efficiency of the cooling liquid cooling system and bubble behavior is evaluated. At the same time, according to the maintenance and fault history records, the weak links and historical fault modes of the liquid cooling system are determined to optimize the evaluation model and improve the accuracy of fault diagnosis. Based on the above collected data, a training data set is constructed, and the pre-established risk assessment model is trained. Preferably, the risk assessment model is constructed using a random forest algorithm. Of course, other machine learning algorithms or neural network models can also be used to construct it, which is not specifically limited here.

[0046] After the risk assessment model is trained, the real-time monitoring data of the liquid cooling system can be input into the risk assessment model. By identifying the patterns of temperature anomalies, flow rate changes and bubble generation, the status of the liquid cooling system can be assessed in real time. That is, it is predicted whether there is a risk of abnormality in the operation of the liquid cooling system, and the predicted specific abnormal pattern and the probability of abnormal occurrence are output as risk assessment indicators. It should be noted that the specific construction process of the risk assessment model in this embodiment and other subsequent models can refer to the conventional model construction and training steps, which will not be repeated here.

[0047] The following example illustrates the above risk assessment steps. Assume that a single-phase immersion liquid cooling system is being monitored. The system is used to cool a high-performance computing server. A series of sensors are used to collect key performance indicator data and use this data to evaluate the health and performance of the system. Specifically, through temperature indicator monitoring, the inlet temperature is 25°C, the outlet temperature is 35°C, the ambient temperature is 22°C, and the temperature change rate is 10°C higher than the inlet temperature. Through flow rate and flow rate change rate monitoring, the real-time flow rate is 4L / min, and the flow rate suddenly drops to 2L / min. Through pressure monitoring, the inlet pressure is 1.5bar and the outlet pressure is 1.3bar. The bubble sensor detects a slight increase in the bubble generation rate, but it does not reach the warning level. Through turbulence and vibration / noise levels, the turbulence is low, and the vibration and noise levels are within the normal range. Through the liquid cooling system operation data monitoring, the pump status is normal operation, the valve position is fully open, the coolant type is synthetic oil, and the coolant concentration is unchanged. Then the impact of ambient humidity and climatic conditions was analyzed, where ambient humidity: 45%, climatic conditions: mild, no significant changes. Historical data showed no major failures in the past 6 months, and regular maintenance was done properly. The collected data was input into the risk assessment model for working mode identification, where the temperature indicator indicated that the system was effectively removing heat, and the flow rate indicator indicated that there might be blockage or pump performance degradation. The pressure indicator indicated that the fluid was flowing normally, and the turbulence and vibration / noise levels indicated that the fluid flow was stable and there was no obvious mechanical damage. The liquid cooling system operation data showed that it met the predetermined parameters. The ambient humidity and climatic conditions indicated that the current environment would not have a negative impact on the cooling efficiency. The historical data indicated that there were no obvious weak links in the system. Therefore, the mode identified by the model is the normal working mode, and the current system status has not triggered any abnormal mode, but the model will continue to monitor the decrease in flow rate and the slight increase in bubble generation rate.

[0048] Step S20, determining whether the risk assessment index is greater than a risk threshold, and if so, determining a potential instability area in the liquid cooling system based on the real-time monitoring data.

[0049] In this embodiment, if the system is identified to have abnormal risk through the above steps and the predicted risk assessment index exceeds the risk threshold, for example, the risk probability of a certain abnormal working mode is greater than 30%, it is necessary to analyze the real-time monitored data to determine the potential instability area in the liquid cooling system, and the specific steps include:

[0050] Step S201, numerically simulating the flow and heat transfer of cooling liquid in the liquid cooling system according to geometric parameters of the liquid cooling system;

[0051] Step S202: compare the real-time monitoring data with the simulation results, and determine the potential unstable area in the liquid cooling system according to the comparison results.

[0052] Specifically, we first use a sensor network to continuously collect data to obtain real-time monitoring data of the liquid cooling system, including temperature, flow rate, pressure and other data. Then, we set parameter thresholds based on the physical properties of the coolant, including using the boiling point of the coolant as the upper temperature threshold, and setting a reasonable range of flow rate based on the flow characteristics of the coolant. We use numerical simulation methods to combine the geometric parameters of the system, that is, the geometric structure and physical parameters, to simulate the flow and heat transfer process of the cooling fluid in the system.

[0053] By comparing the real-time monitoring data with the results of numerical simulation, the difference between the actual operating status of each area in the system and the theoretical model can be judged. If the monitoring data shows that the temperature of a certain area exceeds the threshold, numerical simulation is used to predict whether there is a risk of boiling or bubble formation in the area. If the flow rate data is lower than the set minimum value, it is judged that the cooling efficiency of the area is insufficient, which will increase the risk of overheating and bubble formation. Based on the pressure data of the liquid cooling system, if the pressure of a certain area is much lower than the average level of the system, the simulation predicts whether there is boiling and bubble generation caused by low pressure in the area. Finally, the heat transfer efficiency of the liquid cooling system is calculated based on parameters such as the thermal conductivity, specific heat capacity and flow rate of the cooling liquid. Based on the calculation results of the heat transfer efficiency, the areas where the heat transfer efficiency is lower than the standard value are identified, and these areas are identified as potential unstable areas. Based on the simulation prediction results of the potential unstable areas, high-risk areas in the entire liquid cooling system can be identified, thereby predicting the instability of the potential unstable areas.

[0054] The following example illustrates the above-mentioned potential instability area analysis steps. It is assumed that a sensor network is used to continuously collect temperature, flow rate and pressure data, and the sensors are deployed at key points of the system, including the inlet, outlet and server hot spots. If the coolant is synthetic oil with a boiling point of 120°C, the upper temperature threshold is set to 115°C to provide a safety margin. Based on the flow characteristics of the coolant, the reasonable flow rate range is set to 3-5L / min. Based on the system geometry and the physical properties of the coolant, numerical simulation software is used to simulate the flow of the cooling fluid in the system and the heat transfer process monitoring data. For example, through real-time monitoring, it can be seen that the temperature of a hot spot area is 117°C, the flow rate is 2L / min, and the pressure is 0.8bar, while the numerical simulation results predict that the theoretical temperature of the hot spot area is 112°C, the flow rate is 4L / min, and the pressure is 1.2bar. Since the monitored temperature exceeds the threshold of 115°C, there is a risk of boiling or bubble formation. The flow rate is lower than the minimum value of 3L / min, indicating insufficient cooling efficiency and the risk of overheating and bubble formation. The pressure is lower than the system average, indicating boiling and bubble generation caused by low pressure. If the calculated heat transfer efficiency of the hot spot area is lower than the standard value of 80%, the area is judged to be a potential unstable area. Based on the differences between the monitoring data and the numerical simulation, as well as the heat transfer efficiency analysis, the hot spot area is identified as a high-risk area with a significant risk of instability. Further, based on the analysis results, measures may need to be taken such as clearing blockages in the system, increasing the output flow rate of the pump, or optimizing the server layout to reduce the temperature of the hot spot area.

[0055] Step S30, obtaining first fluid dynamics data of the potential instability area, and performing adaptive fluid dynamics simulation according to the first fluid dynamics data to obtain an optimal flow velocity.

[0056] In this embodiment, by performing fluid dynamics simulation on the fluid in the potential instability area, the optimal flow rate of the fluid in the liquid cooling system can be simulated and predicted, and the specific steps include:

[0057] Step S301, obtaining first fluid dynamic data of the potential instability area and geometric parameters of the liquid cooling system, performing liquid flow simulation according to a turbulence model, and generating initial simulation results;

[0058] Step S302: performing heat transfer analysis on the initial simulation results according to the heat transfer characteristics, and determining the optimal flow rate of the liquid cooling system according to the analysis results.

[0059] Specifically, we first obtain the initial data for fluid dynamics simulation, including fluid properties (including density, viscosity, specific heat capacity, thermal conductivity, surface tension) and geometric parameters (including pipe size and shape), and use these data to establish a basic simulation environment. Using a specific turbulence model k-ε, combined with fluid flow characteristics (including flow velocity, flow pattern and stability), we generate initial simulation results of fluid flow, which reflect the flow behavior at different flow velocities.

[0060] Then, the initial simulation results are analyzed for heat transfer based on the heat transfer characteristics (including convective heat transfer coefficient, radiative heat transfer, thermal conductivity, and interfacial thermal resistance) to determine how heat propagates in the fluid. If the initial simulation shows flow instability areas, these areas are simulated by adjusting the flow rate control strategy and fluid inlet conditions to obtain detailed information about the instability areas. After obtaining the instability areas, the simulation is further refined by modifying the boundary conditions (such as temperature and pressure boundary conditions) to observe how changes in boundary conditions affect heat transfer and bubble behavior. Then, combined with phase change characteristics (including boiling point, latent heat, and phase change dynamics), the bubble generation frequency and size distribution are simulated to observe how bubbles form and grow at different flow rates. By adjusting the numerical solution parameters (such as mesh density, time step, and convergence criteria), the simulation accuracy is optimized to ensure the reliability and practicality of the simulation results. Through the analysis of the simulation results, the optimal fluid flow rate is determined. The optimal flow rate should maximize the heat exchange efficiency and control bubble generation to avoid affecting the heat transfer performance due to excessive bubbles. This embodiment adjusts the control strategy parameters of the fluid dynamics simulation software through simulation results to obtain the optimal flow rate, which can achieve maximum heat exchange efficiency and control bubble generation in actual operation.

[0061] The following example illustrates the above optimal flow rate calculation steps. Assume that the initial data is obtained first. The fluid properties are synthetic oil as the coolant. The specific parameters are density 860kg / m³, viscosity 0.025Pa·s, specific heat capacity 1800J / (kg·K), thermal conductivity 0.15W / (m·K), and surface tension 0.030N / m. The geometric parameters are that the cooling system contains a set of pipes with a diameter of 25mm and a length of 1m, and the layout is designed to optimize the cooling efficiency. The k-ε turbulence model is used, and the flow rate setting range is 0.5m / s to 2m / s to simulate the flow behavior at different flow rates. The convective heat transfer coefficient is calculated based on the flow rate and fluid properties, and the data obtained is 500W / (m²·K). Here, heat transfer is mainly considered through convection and heat conduction, while the influence of radiation heat transfer and interface thermal resistance is ignored. Initial simulation results showed that flow instability occurred in some narrow parts of the system when the flow rate was close to 2m / s. By reducing the flow rate to 1.5m / s, the instability was significantly improved. The boundary conditions were modified to set the inlet temperature to 60°C and the pressure to 1bar. Further simulations showed that heat transfer and fluid flow were more stable under these conditions. Combining the boiling point of the coolant at 120°C and the latent heat of 225kJ / kg, it was found that at a flow rate of 1.5m / s, the bubble generation frequency and size distribution were within a controllable range and did not significantly affect the heat exchange efficiency. The accuracy of the simulation was optimized by adjusting the grid density to 10 grids per cubic millimeter, setting the time step to 0.01 seconds, and setting the convergence criterion to a residual less than 1e-5. After analyzing the simulation results, 1.5m / s was determined to be the optimal flow rate, at which the heat exchange efficiency is the highest and bubble generation can be effectively controlled to maintain good cooling performance. Subsequently, the flow rate control strategy of the liquid cooling system can be adjusted according to the simulation results, and the pump speed can be set to maintain the coolant flow rate at 1.5m / s to maximize the heat exchange efficiency and control bubble generation in actual operation.

[0062] Step S40, determining whether the current flow rate of the liquid cooling system is consistent with the optimal flow rate, and if not, adjusting the current flow rate according to the optimal flow rate.

[0063] In this embodiment, a flow rate sensor is first used to monitor the actual flow rate of the liquid in the system to obtain accurate data on the flow conditions in real time. The temperature sensor is used to obtain liquid temperature information at key points in the system. The pressure sensor is used to measure the pressure value inside the system to evaluate the flow resistance and the working condition of the pump. The controller / processor receives data from the flow rate, temperature and pressure sensors, analyzes these data, and determines whether the flow rate needs to be adjusted. If the controller or processor determines that the current flow rate does not meet the optimal flow rate, it will send an instruction to the pump speed control to instruct it to adjust the pump speed to change the flow rate.

[0064] Specifically, by simulating fluid dynamics, the theoretical optimal flow rate is obtained and compared with the real-time flow rate data collected by the sensor. If the real-time flow rate is inconsistent with the optimal flow rate obtained by simulation, the controller or processor will send an adjustment instruction to the electronic valve or adjustable fluid path component through the communication interface to further adjust the flow rate. According to the adjustment results of the controller or processor, the working state of the radiator / heat transfer element is adjusted to ensure that the heat transfer or dissipation efficiency matches the requirements after the flow rate adjustment. The controller or processor uses safety features for monitoring. If the flow rate or temperature exceeds the safety range, immediate countermeasures are taken, including reducing the speed of the pump or starting the automatic shutdown / warning mechanism to prevent system overload.

[0065] The following example illustrates the above flow rate adjustment steps. Assume that there is a single-phase immersion liquid cooling system designed to cool a high-performance computing server. The optimal flow rate of the system is 1.5 m / s through software simulation, which is designed to ensure the highest heat exchange efficiency and control bubble generation. The real-time monitoring of the flow rate sensor shows that the flow rate is 1.2 m / s, which is lower than the optimal flow rate of 1.5 m / s obtained by simulation. The temperature information is obtained through the temperature sensor. The key point temperature monitoring results show that the outlet temperature is 38°C, which is higher than the set optimal operating temperature of 35°C. Based on the pressure value measured by the pressure sensor, the internal pressure monitoring result of the system is 1.1 bar, indicating that there is a slight flow resistance inside the system. After analyzing the received data, the controller determines that the current flow rate of 1.2 m / s is lower than the theoretical optimal flow rate of 1.5 m / s and needs to be adjusted. At this time, the controller sends a command to the pump speed control unit to instruct it to increase the pump speed to increase the flow rate to 1.5 m / s. The optimal flow rate simulated by the software is 1.5 m / s. The controller receives the real-time data after the flow rate adjustment and confirms that the flow rate has been adjusted to 1.5m / s. Since the flow rate has reached the optimal value through the pump speed adjustment, this step does not require additional adjustment in the current scenario. Based on the data after the flow rate adjustment, the controller adjusts the working state of the radiator to match the heat dissipation demand after the flow rate is increased to ensure that the temperature is controlled within the optimal range. The controller continuously monitors the flow rate and temperature data. During the adjustment process, ensure that all parameters remain within the safe range to avoid system overload.

[0066] Step S50, obtaining second fluid dynamic data of the potential instability area after the flow rate is adjusted, performing flow instability analysis on the second fluid dynamic data, and adjusting the liquid cooling system according to the analysis result.

[0067] In order to improve the cooling efficiency of the liquid cooling system, in this embodiment, flow instability analysis is further performed on the potential instability area after the flow rate is adjusted, so as to determine whether further adjustment is required to the liquid cooling system. The specific analysis steps include:

[0068] Step S501, obtaining second fluid dynamic data of the potential instability area after the flow rate is adjusted, wherein the second fluid dynamic data includes flow rate data, bubble data and temperature data;

[0069] Step S502, performing flow stability analysis on the second fluid dynamic data to determine whether there is abnormal data;

[0070] Step S503: If yes, flow instability analysis is performed according to the correlation of each abnormal data, and the flow rate and / or workload of the liquid cooling system is adjusted according to the analysis result.

[0071] Specifically, firstly, a high-precision flow velocity sensor is used to obtain the real-time flow velocity value of the fluid in the pipeline. If the flow velocity fluctuates abnormally, the flow direction and flow data are further analyzed to determine whether there is flow instability. According to the flow velocity data, the fluid flow conditions are adjusted to ensure flow stability. If the flow velocity exceeds the preset safety range, the flow velocity will be automatically reduced or increased to maintain flow stability. A bubble monitoring sensor is used to obtain real-time data on bubble size and number in the fluid. If the number of bubbles exceeds the preset normal range, the bubble distribution and dynamic properties are analyzed to determine whether the previously determined potential instability area has been reached. The temperature monitoring sensor is used to obtain the temperature value of the fluid and its surrounding environment. If the monitored temperature gradient exceeds the normal range, the thermal flow data is used to further determine whether it affects the stability of the fluid flow. The data acquisition system is used to synchronously obtain data from the flow velocity, bubble and temperature sensors. Through the data processing process, useful information is cleaned and screened out, and the data is analyzed. The flow behavior and development trend are further predicted through the correlation of abnormal data. If potential instability areas are predicted, the fluid flow conditions are adjusted to avoid entering these areas.

[0072] Then, through data visualization technology, the processed data is displayed in the form of charts or images. If the bubble monitoring or temperature monitoring sensor data shows abnormalities, the problem area is highlighted through visualization tools. The automatic adjustment function is used to make necessary adjustments to the flow rate and temperature based on the real-time feedback monitoring data and control algorithm. Through this automatic adjustment, the system of this embodiment maintains the flow process in an optimal state and reduces the risk of instability. Through the fault diagnosis mechanism, when an abnormal situation is detected, the system will automatically identify the problem and try to solve it. If the problem persists, an alarm will be issued through the liquid cooling system, indicating that manual intervention is required to ensure the safe operation of the system.

[0073] The following example illustrates the adjustment steps of the above liquid cooling system. Assume that a single-phase immersion liquid cooling system is used to cool a group of high-performance servers. The system is designed to optimize thermal management and ensure stable operation of the servers. Real-time flow rate monitoring with high-precision flow rate monitoring shows that the flow rate fluctuates, dropping from 1.5m / s to 1.0m / s. Based on this abnormal prompt, the flow direction and flow rate are further analyzed. According to the simulation prediction, the safe flow rate range is set to 1.3 to 1.7m / s. The bubble monitoring sensor data shows that the number of bubbles increases from 2 to 10 per cubic centimeter, exceeding the normal range. After analysis, it is known that the bubbles are mainly concentrated in the hot spot area. The temperature sensor shows that the temperature gradient of the fluid near a server is abnormal, rising from 35°C to 40°C, exceeding the normal range of 5°C to 35°C, indicating that heat transfer may be blocked. The data acquisition system synchronously processes the flow rate, bubble and temperature data. The cleaned data shows that the decrease in flow rate and the increase in bubbles are directly related to the abnormal temperature gradient. Then the data is analyzed and predicted whether these abnormal conditions may cause flow instability. And according to the analysis results, the flow rate is adjusted and the workload of certain servers is changed to reduce the burden on hot spots. In this embodiment, visualization technology highlights problem areas and abnormal data through charts and images, helping operators quickly identify and focus on key issues. The liquid cooling system can automatically adjust the flow rate and temperature based on feedback and try to restore to the optimal conditions. At the same time, the fault diagnosis mechanism is activated to identify the possible fault point as an overheating server. If the automatic adjustment fails to solve the problem, the system issues an alarm to indicate that manual intervention is required. Maintenance personnel can check the overheated server and adjust the cooling strategy based on the visualization data and fault diagnosis results.

[0074] In a preferred embodiment, after adjusting the liquid cooling system according to the analysis results, the present invention further comprises:

[0075] Acquire the adjusted real-time temperature data and real-time pressure data of the liquid cooling system, and determine whether there is an abnormal area according to the real-time temperature data and the real-time pressure data, wherein the abnormal area includes a bubble accumulation area and / or a local overheating area;

[0076] If so, the flow rate of the liquid cooling system is adjusted, and based on the feedback loop, it is determined whether to continue adjusting the flow rate.

[0077] In this embodiment, by installing high-precision temperature and pressure sensors, and based on a preset sampling rate, real-time temperature and pressure data inside the liquid cooling system are obtained. The stable communication protocol MQTT is used to transmit the data collected by the sensor to the data processing platform in real time, reducing the delay and packet loss rate during data transmission. A real-time data processing algorithm is deployed in the data processing platform to analyze the collected data and identify signs of bubble accumulation or local overheating. Specifically, a temperature sensor and a pressure sensor are installed in the liquid cooling system, and the sampling rate is set to sample once per second to capture real-time temperature and pressure data. The temperature sensor is installed at the server outlet, and the pressure sensor is installed at the cooling system inlet and outlet. Using the MQTT communication protocol, the temperature and pressure data are transmitted to the data processing platform in real time to determine whether the pressure data is lower than the pressure threshold. If so, the bubbles in the liquid are detected by an ultrasonic sensor. If the sensor feedbacks a bubble signal, the bubble processing process is started. The density of the bubbles is determined according to the distribution and size of the bubbles, so as to determine whether there is a bubble accumulation area. At the same time, based on the temperature data provided by the thermocouple or thermistor in the temperature sensor, it is determined whether there is local overheating. If the temperature exceeds the set threshold, it indicates local overheating, and a signal of bubble accumulation or local overheating will be issued.

[0078] Then the current flow rate and temperature data are analyzed to calculate the new flow rate adjustment value, for example, the latest flow rate can be determined by numerical simulation. The flow controller dynamically adjusts the flow rate according to the calculated adjustment value, and uses a regulating valve or a variable speed pump to accurately control the flow of the fluid. At the same time, the sensor in the feedback loop is used to re-monitor the adjusted flow rate, bubble situation and temperature to confirm the adjustment effect. If the sensor data shows that the bubbles are not effectively dispersed or the local hot spots are not eliminated, the flow rate is fine-tuned according to the preset software algorithm. If the continuous monitoring data shows that the adjustment is effective, the adjustment parameters are dynamically saved as a reference for similar situations in the future. The adjustment results are sent to the user interface through the communication interface to let the operator understand the current status of the liquid cooling system and the adjustment measures performed. The actuator responds to the command of the control unit and adjusts the motor of the pump to achieve fine-tuning of the flow rate.

[0079] Assume that a single-phase immersion liquid cooling system is used to cool a high-performance computing center. The system design includes monitoring and control strategies to ensure that the server always runs at the optimal temperature. Ultrasonic sensors are deployed in the liquid cooling system to monitor bubbles in the coolant. The sensor detects the presence of bubble signals in the coolant in a certain area. The thermocouple monitors the temperature of the area to be 45°C, which exceeds the normal operating temperature threshold of 40°C set by the system, indicating that there is local overheating. The control unit receives the bubble and overheating signals. The current flow rate is 1.2m / s. According to the preset algorithm, the new flow rate adjustment value is calculated to be 1.5m / s to enhance the cooling effect and dissipate the bubbles. The flow controller adjusts the flow rate from 1.2m / s to 1.5m / s through the regulating valve and variable speed pump. After the adjustment, the sensor re-monitors the flow rate to 1.5m / s, the temperature drops to 39°C, and the number of bubbles is significantly reduced. Despite the improvement, the control unit continues to fine-tune the flow rate to 1.55m / s to ensure that the bubbles are completely dissipated and the temperature is further reduced. Continuous monitoring data shows that after the flow rate is adjusted to 1.55m / s, the temperature stabilizes at 38°C and no bubbles are generated. This parameter setting is then dynamically saved as a reference for similar situations in the future, and the adjustment results, including the flow rate adjustment to 1.55m / s and the current temperature of 38°C, are sent to the user interface through the communication interface, so that the operator can understand the adjustment measures and the real-time status of the liquid cooling system. The actuator responds to further commands from the control unit and fine-tunes the pump speed to ensure that the flow rate is maintained at 1.55m / s to maximize heat exchange efficiency and eliminate bubble problems.

[0080] In addition, if the data processing algorithm analysis results show that the temperature or pressure is not within the preset safety threshold, the alarm mechanism of the liquid cooling system can be triggered to alert relevant personnel. The alarm information obtained will be notified to the maintenance team or operator through a specified communication method, such as SMS or email. At the same time, real-time monitoring data is displayed on the user interface. If an abnormal situation is found, the liquid cooling system settings are adjusted or the alarm is responded to through the interactive interface. All acquired data is stored, and it is determined whether there are periodic thermal management problems in the liquid cooling system. Redundant design is implemented on key components, and regular maintenance is performed to improve the reliability of the liquid cooling system and the stability of data processing. Furthermore, high-level encryption and access control measures can be provided for data to prevent unauthorized access, thereby ensuring the security of data transmission and storage.

[0081] For example, in the above case, because the temperature exceeds the preset safety threshold, the system automatically triggers the alarm mechanism to notify the maintenance team. The maintenance team receives alerts via text messages and emails, including the detailed location of the abnormal area and recommended inspection measures. Real-time monitoring data is displayed on the user interface. After the operator observes abnormal data, he can remotely adjust the flow rate or start the cooling system's backup route to try to reduce the temperature. All collected data is stored in the database to analyze whether there are periodic thermal management problems in the liquid cooling system, such as finding that the temperature is high during a specific time period every afternoon. Redundant design is implemented on key sensors and control systems to ensure that even if some components fail, the stability of the entire system will not be affected. Regular system maintenance is performed to maintain optimal performance. Advanced encryption measures are implemented for transmitted and stored data, and only authorized users can access system data to prevent data leakage or unauthorized access.

[0082] In a preferred embodiment, after adjusting the liquid cooling system according to the analysis results, the present invention further comprises:

[0083] The response time and performance data of the adjusted liquid cooling system are obtained, analysis of the performance data is triggered based on the response time, and dynamic resource allocation is performed on the liquid cooling system according to the analysis result.

[0084] In this embodiment, an application performance management tool is used to monitor the key performance indicators of the liquid cooling system, and based on the system response time, such as when the response time exceeds the threshold, a preliminary diagnosis process for performance problems is triggered. The preliminary diagnosis process includes obtaining the resource utilization of the liquid cooling system, including CPU, memory, disk I / O and network usage, through real-time data flow analysis. If the resource utilization is found to be close to the upper limit, the dynamic resource allocation strategy is adjusted. At the same time, the number of concurrent users is analyzed according to the trend of historical data to determine the performance of the liquid cooling system during high-load periods. If the concurrent volume is predicted to increase, the computing resources are increased in advance according to the automatic expansion rules. In addition, a real-time alarm mechanism of the monitoring software is set. If the error rate of the liquid cooling system increases, a notification is immediately sent to the liquid cooling system (such as email, SMS, mobile phone push) to prompt corrective measures. If the monitoring tool detects a decrease in throughput, the allocation request is set through load balancing to reduce the pressure on a single server and restore or improve the overall performance of the liquid cooling system.

[0085] At the same time, log management tools are used to analyze the operation logs of the liquid cooling system to determine the cause of the abnormality of the liquid cooling system, and the configuration management is adjusted based on this information to optimize the liquid cooling system settings. For example, if the service quality monitoring QoS indicates that the service level does not meet the service level agreement SLA, the failover mechanism is activated to maintain service stability and avoid service interruptions. In addition, when unauthorized access is detected through security monitoring, access control measures can be enabled to protect the liquid cooling system from malicious attacks, and related security events can be recorded at the same time. According to the change management process, the liquid cooling system is updated or modified as necessary. If the performance test results after the change are good, the new configuration will be promoted to the production environment. The following is an example to illustrate.

[0086] Assume that a single-phase immersion liquid cooling system is used to cool a server cluster in a cloud computing data center. The system uses the application performance management tool APM to ensure that key performance indicators (KPIs) meet expected standards. The APM tool sets the response time threshold to 200 milliseconds. Once the response time reaches 250 milliseconds, the initial diagnosis process of the performance problem is immediately triggered. Real-time monitoring finds that the CPU utilization rate reaches 95%, the memory utilization rate is 90%, and the disk I / O and network utilization rates are also close to the upper limit. The system then adjusts the dynamic resource allocation and migrates some computing tasks to servers with lower utilization rates. Historical data trend analysis shows that during high-load periods, the number of concurrent users is expected to increase from 1,000 to 1,500. The system automatically starts the expansion rule to increase computing resources to cope with the expected load increase. The monitoring software finds that the error rate has increased from 0.1% to 1%, and immediately sends a notification to the operation and maintenance team via email and SMS, prompting them to check and take corrective measures. When the throughput drops by 10%, the load balancer automatically redistributes requests, reducing the reliance on the busiest server and restoring system performance. The log management tool analysis found that the hot spot area of ​​a server frequently overheated due to unbalanced task distribution. By adjusting the configuration management, the temperature of the server was reduced, making the system run more stably. QoS monitoring pointed out that the availability of a service dropped to 99%, failing to meet the 99.9% specified in the SLA. The system automatically started the failover mechanism to direct traffic to the backup system to avoid service interruption. Security monitoring detected several unauthorized access attempts. The system immediately enabled access control measures to block these attempts and record related security events. Based on the above optimization steps, the liquid cooling system was updated and tested. The updated performance test showed that the response time was improved to 180 milliseconds, and the CPU and memory utilization dropped to 80%. At this time, it was considered that the test results met the performance requirements, and then the new configuration was officially deployed to the production environment.

[0087] In a preferred embodiment, after adjusting the liquid cooling system according to the analysis results, the present invention further comprises:

[0088] Obtaining performance data of the adjusted liquid cooling system and inputting the data into a pre-established performance prediction model to obtain a performance prediction result;

[0089] According to the performance prediction result, the control strategy of the liquid cooling system is optimized, and the optimization includes cooling liquid flow optimization and fan speed optimization.

[0090] In this embodiment, the performance data of the liquid cooling system obtained by sensors and monitoring equipment, including temperature, flow rate, pressure and other data, are used to obtain the real-time operating status of the liquid cooling system in the data center. Through data acquisition and transmission technology, these real-time data are sent to the data processing center. The received data is processed using a performance prediction model to obtain a prediction result about the future performance of the liquid cooling system. The performance prediction model here can be constructed using a machine learning algorithm or a neural network model, and there are no excessive restrictions here.

[0091] If the performance prediction model indicates the presence of potential instability factors, the control strategy and decision algorithm are used to determine the corresponding adjustment measures. According to the determined measures, the actuators and control devices are dynamically adjusted to deal with the predicted problems, such as adjusting the coolant flow or changing the fan speed. After the adjustment measures are implemented, a feedback mechanism is established to ensure that the latest liquid cooling system performance data is transmitted back to the data processing center to provide new feedback information for the closed-loop liquid cooling system. After obtaining the new feedback information, the optimization algorithm analyzes the data and optimizes the performance prediction model and control strategy. At the same time, the performance of the improved liquid cooling system can be analyzed through the stability analysis tool to determine the stability of the liquid cooling system after the implementation of the new control strategy. The optimized liquid cooling system is monitored using a safety and fault-tolerant mechanism to prevent potential errors or failures and ensure the continuous and stable operation of the liquid cooling system.

[0092] The following example illustrates the above performance prediction steps. Assume that a single-phase immersion liquid cooling system is used to cool a high-performance server in a data center. The system uses a series of sensors to monitor key performance indicators such as temperature, flow and pressure, and uses machine learning to optimize its operation. The data monitored by the sensors include the coolant inlet temperature of 35°C, the outlet temperature of 45°C, the flow rate of 4L / min, and the system pressure of 1.2bar. Real-time data is sent to the data processing center through a high-speed network, using the latest IoT communication technology to ensure the instant transmission and accuracy of data. The data processing center processes the received data through a pre-established performance prediction model. The model can predict that the coolant outlet temperature may rise to 50°C in the next hour, causing overheating. The control strategy is adjusted according to the prediction results. For example, the adjustment steps of the control strategy are determined based on the numerical model. The control strategy determines that the coolant flow rate needs to be increased to 5L / min and the fan speed needs to be increased from 1500rpm to 2000rpm to prevent overheating. The actuator dynamically adjusts the coolant flow rate and fan speed to deal with the predicted overheating problem. After the adjustment measures, the new performance data showed that the coolant outlet temperature was stable at 42°C, the flow rate was successfully increased to 5L / min, and the system pressure was adjusted to 1.3bar. The optimization algorithm is used to analyze the new feedback data and fine-tune the performance prediction model to improve the accuracy of future predictions and the efficiency of the control strategy. Verified by the stability analysis tool, the adjusted liquid cooling system is more stable and can effectively cope with the predicted high load conditions. This embodiment continuously monitors the operation of the liquid cooling system based on the safety monitoring system, and ensures that any potential hardware failures or data transmission errors will not cause system interruptions through fault-tolerant design.

[0093] This embodiment provides an adaptive control method for a liquid cooling system. The present invention timely identifies and prevents conditions that may cause system instability, such as temperature anomalies and flow rate changes, through real-time monitoring and data analysis of the liquid cooling system, thereby ensuring that the equipment operates at the optimal temperature and avoiding the risk of performance degradation or equipment damage due to local overheating. Through real-time feedback and dynamic adjustment mechanisms, the present invention can not only quickly respond to changes in various working conditions, but also continuously self-optimize, thereby ensuring the long-term, efficient and stable operation of the liquid cooling system. The present invention not only effectively improves the stability and thermal management efficiency of the liquid cooling system, but also improves the reliability and safety of the overall operation of the liquid cooling system.

[0094] See also Figure 2 Based on the same inventive concept, a second embodiment of the present invention provides an adaptive control device for a liquid cooling system, comprising:

[0095] The risk assessment module 10 is used to obtain real-time monitoring data of the liquid cooling system and input the pre-established risk assessment model to obtain risk assessment indicators;

[0096] An instability determination module 20, for determining whether the risk assessment index is greater than a risk threshold, and if so, determining a potential instability region in the liquid cooling system according to the real-time monitoring data;

[0097] A flow velocity simulation module 30, for acquiring first fluid dynamics data of the potential instability region, and performing adaptive fluid dynamics simulation according to the first fluid dynamics data to obtain an optimal flow velocity;

[0098] A flow rate adjustment module 40, used to determine whether the current flow rate of the liquid cooling system is consistent with the optimal flow rate, and if not, to adjust the current flow rate according to the optimal flow rate;

[0099] The analysis and adjustment module 50 is used to obtain the second fluid dynamic data of the potential instability area after the flow rate is adjusted, perform flow instability analysis on the second fluid dynamic data, and adjust the liquid cooling system according to the analysis result.

[0100] The technical features and technical effects of the adaptive control device for a liquid cooling system proposed in an embodiment of the present invention are the same as those of the method proposed in an embodiment of the present invention, and are not described in detail here. Each module in the adaptive control device for a liquid cooling system can be implemented in whole or in part by software, hardware, and a combination thereof. Each of the above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0101] In addition, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0102] See also Figure 3, an internal structure diagram of a computer device in one embodiment, the computer device can specifically be a terminal or a server. The computer device includes a processor, a memory, a network interface, a display and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an adaptive control method for a liquid cooling system is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0103] It can be understood by those skilled in the art that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computing device may include more or fewer components than shown in the figure, or combine certain components, or have the same component arrangement.

[0104] In addition, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0105] In summary, an adaptive control method, device, equipment and storage medium for a liquid cooling system proposed in an embodiment of the present invention, the method obtains real-time monitoring data of the liquid cooling system and inputs a pre-established risk assessment model to obtain a risk assessment index; determines whether the risk assessment index is greater than a risk threshold, and if so, determines a potential instability region in the liquid cooling system according to the real-time monitoring data; obtains first fluid dynamic data of the potential instability region, and performs adaptive fluid dynamics simulation according to the first fluid dynamic data to obtain an optimal flow rate; determines whether the current flow rate of the liquid cooling system is consistent with the optimal flow rate, and if not, adjusts the current flow rate according to the optimal flow rate; obtains second fluid dynamic data of the potential instability region after the flow rate adjustment, performs flow instability analysis on the second fluid dynamic data, and adjusts the liquid cooling system according to the analysis results. The present invention timely identifies and prevents conditions that may cause system instability, such as abnormal temperature and flow rate changes, through real-time monitoring and data analysis of the liquid cooling system, thereby ensuring that the equipment operates at an optimal temperature and avoiding the risk of performance degradation or equipment damage caused by local overheating. The present invention not only can quickly respond to changes in various working conditions through real-time feedback and dynamic adjustment mechanisms, but also can continuously self-optimize, thereby ensuring the long-term, efficient and stable operation of the liquid cooling system. The present invention not only effectively improves the stability and thermal management efficiency of the liquid cooling system, but also improves the reliability and safety of the overall operation of the liquid cooling system.

[0106] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0107] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. An adaptive control method for a liquid cooling system, characterized in that: include: Obtain real-time monitoring data of the liquid cooling system and input it into a pre-established risk assessment model to obtain risk assessment indicators; Determining whether the risk assessment index is greater than a risk threshold, and if so, determining a potential instability area in the liquid cooling system based on the real-time monitoring data; Acquiring first fluid dynamics data of the potential instability region, and performing adaptive fluid dynamics simulation according to the first fluid dynamics data to obtain an optimal flow velocity; Determine whether the current flow rate of the liquid cooling system is consistent with the optimal flow rate, and if not, adjust the current flow rate according to the optimal flow rate; Acquire second fluid dynamic data of the potential instability area after the flow velocity is adjusted, perform flow instability analysis on the second fluid dynamic data, and adjust the liquid cooling system according to the analysis result; The steps of obtaining the second fluid dynamic data of the potential instability area after the flow rate is adjusted, performing flow instability analysis on the second fluid dynamic data, and adjusting the liquid cooling system according to the analysis results include: Acquire second fluid dynamic data of the potential instability region after the flow rate is adjusted, wherein the second fluid dynamic data includes flow rate data, bubble data and temperature data; Performing flow stability analysis on the second fluid dynamic data to determine whether there is abnormal data; If so, flow instability analysis is performed based on the correlation of each abnormal data, and the flow rate and / or workload of the liquid cooling system is adjusted based on the analysis results.

2. The adaptive control method of the liquid cooling system according to claim 1, characterized in that: The step of determining the potential instability area in the liquid cooling system according to the real-time monitoring data comprises: According to the geometric parameters of the liquid cooling system, the flow and heat transfer of the cooling liquid in the liquid cooling system are numerically simulated; The real-time monitoring data is compared with the simulation results, and a potential unstable area in the liquid cooling system is determined according to the comparison results.

3. The adaptive control method of the liquid cooling system according to claim 1, characterized in that: The step of obtaining the first fluid dynamics data of the potential instability region and performing adaptive fluid dynamics simulation according to the first fluid dynamics data to obtain the optimal flow rate comprises: Acquiring first fluid dynamic data of the potential instability region and geometric parameters of the liquid cooling system, performing liquid flow simulation according to a turbulence model, and generating initial simulation results; A heat transfer analysis is performed on the initial simulation results according to the heat transfer characteristics, and an optimal flow rate of the liquid cooling system is determined according to the analysis results.

4. The adaptive control method of the liquid cooling system according to claim 1, characterized in that: After the step of adjusting the liquid cooling system according to the analysis result, the method further includes: Acquire the adjusted real-time temperature data and real-time pressure data of the liquid cooling system, and determine whether there is an abnormal area according to the real-time temperature data and the real-time pressure data, wherein the abnormal area includes a bubble accumulation area and / or a local overheating area; If so, the flow rate of the liquid cooling system is adjusted, and based on the feedback loop, it is determined whether to continue adjusting the flow rate.

5. The adaptive control method of the liquid cooling system according to claim 1, characterized in that: After the step of adjusting the liquid cooling system according to the analysis result, the method further includes: The response time and performance data of the adjusted liquid cooling system are obtained, analysis of the performance data is triggered based on the response time, and dynamic resource allocation is performed on the liquid cooling system according to the analysis result.

6. The adaptive control method of the liquid cooling system according to claim 1, characterized in that: After the step of adjusting the liquid cooling system according to the analysis result, the method further includes: Obtaining performance data of the adjusted liquid cooling system and inputting the data into a pre-established performance prediction model to obtain a performance prediction result; According to the performance prediction result, the control strategy of the liquid cooling system is optimized, and the optimization includes cooling liquid flow optimization and fan speed optimization.

7. An adaptive control device for a liquid cooling system, characterized in that: include: The risk assessment module is used to obtain the real-time monitoring data of the liquid cooling system and input it into a pre-established risk assessment model to obtain risk assessment indicators; An instability determination module, used to determine whether the risk assessment index is greater than a risk threshold, and if so, determine a potential instability area in the liquid cooling system based on the real-time monitoring data; A flow velocity simulation module, used for acquiring first fluid dynamics data of the potential instability region, and performing adaptive fluid dynamics simulation according to the first fluid dynamics data to obtain an optimal flow velocity; A flow rate adjustment module, used to determine whether the current flow rate of the liquid cooling system is consistent with the optimal flow rate, and if not, to adjust the current flow rate according to the optimal flow rate; The analysis and adjustment module is used to obtain the second fluid dynamic data of the potential instability area after the flow rate is adjusted, perform flow instability analysis on the second fluid dynamic data, and adjust the liquid cooling system according to the analysis results, specifically including: Acquire second fluid dynamic data of the potential instability region after the flow rate is adjusted, wherein the second fluid dynamic data includes flow rate data, bubble data and temperature data; Performing flow stability analysis on the second fluid dynamic data to determine whether there is abnormal data; If so, flow instability analysis is performed based on the correlation of each abnormal data, and the flow rate and / or workload of the liquid cooling system is adjusted based on the analysis results.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.