Cooling liquid flow regulation and control method and system of immersed liquid cooling system

Abstract

The invention provides a cooling liquid flow regulation and control method and system for an immersed liquid cooling system, and the method comprises the steps: obtaining a set temperature of the liquid cooling system, obtaining a power interval of a liquid cooling distribution unit and a rotating speed interval of a cooling water pump based on the set temperature of the liquid cooling system, defining a first random process based on a Monte Carlo algorithm, and obtaining a first random liquid cooling sample; defining a second random process based on a random forest algorithm and obtaining a second random liquid cooling sample, executing a particle swarm optimization algorithm on the first random liquid cooling sample to obtain a first optimized liquid cooling regulation and control parameter, executing the random forest algorithm on the second random liquid cooling sample to obtain a second optimized liquid cooling regulation and control parameter, based on the root locus diagram of the liquid cooling second-order transfer function, stability judgment is carried out, and optimized liquid cooling regulation and control parameters or fused liquid cooling regulation and control parameters are obtained; and importing the optimized liquid cooling regulation and control parameters or the fused liquid cooling regulation and control parameters into a liquid cooling regulation and control module, thereby improving the cooling effect of the immersed liquid cooling system on a server or IT equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid cooling heat dissipation, and more specifically, to a method and system for regulating the coolant flow rate of an immersion liquid cooling system. Background Art

[0002] In the digital age, the computing power of data centers, artificial intelligence server clusters, and high-performance computing devices has increased exponentially, and the heat generated during device operation has also increased sharply. Due to the limited heat dissipation efficiency of traditional air cooling methods, it is difficult to meet the heat dissipation requirements of high heat density devices, and immersion liquid cooling systems stand out due to their heat dissipation performance.

[0003] In the field of traditional immersion liquid cooling system regulation, existing technologies often adopt flow regulation strategies based on PID control or single optimization algorithms, such as genetic algorithms and gradient descent methods. Although they can achieve basic temperature control functions, they have limitations.

[0004] First, single optimization algorithms are prone to falling into local optima, especially in high-dimensional parameter spaces with multi-variable coupling, such as the joint optimization of the power of the liquid cooling distribution unit and the rotation speed of the cooling water pump, which is difficult to balance. Second, traditional coolant flow rate regulation methods often use single parameter settings, which can lead to temperature control lags and a sharp increase in energy consumption. Therefore, the present invention provides a regulation method that combines a hybrid algorithm and can optimize the synergistic cooling effect. Summary of the Invention

[0005] In view of the above-mentioned problems, combined with the first aspect of the present invention, the purpose and efficacy of a method and system for regulating the coolant flow rate of an immersion liquid cooling system are achieved by the following specific technical means: A method for regulating the coolant flow rate of an immersion liquid cooling system includes: S1: Obtain the set temperature of the liquid cooling system, and obtain the power range of the liquid cooling distribution unit and the rotation speed range of the cooling water pump based on the set temperature of the liquid cooling system; S2: Define a first stochastic process of the power of the liquid cooling distribution unit and the rotation speed of the cooling water pump based on the Monte Carlo algorithm, obtain a first random liquid cooling sample based on the first stochastic process, define a second stochastic process of the power of the liquid cooling distribution unit and the rotation speed of the cooling water pump based on the random forest algorithm, and obtain a second random liquid cooling sample based on the second stochastic process; S3: Execute the particle swarm optimization algorithm on the first random liquid cooling sample, and perform repeated simulation updates based on the particle swarm optimization algorithm to obtain a first optimized liquid cooling regulation parameter; S4: Execute the random forest algorithm on the second random liquid cooling sample, and perform repeated simulation updates based on the random forest algorithm to obtain a second optimized liquid cooling regulation parameter; S5: Establish a liquid cooling second-order transfer function, plot the root locus of the liquid cooling second-order transfer function, perform stability judgment based on the root locus of the liquid cooling second-order transfer function, and obtain optimized liquid cooling control parameters or integrated liquid cooling control parameters based on the stability judgment; S6: Import the optimized liquid cooling control parameters or the integrated liquid cooling control parameters into the liquid cooling control module, and the liquid cooling distribution unit power and the cooling water pump speed can be controlled in real time through the liquid cooling control module.

[0006] As a further solution of the present invention, the stability judgment based on the root locus of the liquid cooling second-order transfer function, and obtaining optimized liquid cooling control parameters or integrated liquid cooling control parameters based on the stability judgment includes: Import the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter into the liquid cooling second-order transfer function, and find the positions of the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter on the root locus of the liquid cooling transfer function; If both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in the second or third quadrant of the root locus diagram, then both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in a stable state, perform a fusion operation on the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter, and obtain the integrated liquid cooling control parameter based on the fusion operation; If one of the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter is in the first quadrant or the fourth quadrant of the root locus diagram, then one of the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter is in an unstable state, and obtain the other set of parameters in the stable state as the optimized liquid cooling control parameter; If both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in the first quadrant or the fourth quadrant of the root locus diagram, then both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in an unstable state, and re-execute step S2.

[0007] As a further solution of the present invention, the fusion operation on the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter, and obtaining the integrated liquid cooling control parameter based on the fusion operation includes: The fusion operation is expressed as performing a weighted sum on the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter, and the calculation formula of the weighted sum is expressed as: ; Where, represents the integrated liquid cooling control parameter, represents the first optimized liquid cooling control parameter, Denoted as the second optimized liquid cooling control parameter, Denoted as the calculation weight of the first optimized liquid cooling control parameter, Denoted as the calculation weight of the second optimized liquid cooling control parameter; When performing calculations, a constraint condition is introduced, and the constraint condition is denoted as: , and .

[0008] As a further solution of the present invention, establishing a liquid cooling second-order transfer function and plotting the root locus of the liquid cooling second-order transfer function includes: When plotting the root locus of the liquid cooling second-order transfer function, a plane coordinate system is established. In the plane coordinate system, the X-axis represents real numbers, and the Y-axis represents complex numbers. When the root locus of the liquid cooling second-order transfer function is in the second or third quadrant of the plane coordinate system, it is in a stable state. When the root locus of the liquid cooling second-order transfer function is in the first or fourth quadrant of the plane coordinate system, it is in an unstable state.

[0009] As a further solution of the present invention, defining a first random process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the Monte Carlo algorithm, and obtaining a first random liquid cooling sample based on the first random process, includes: Dividing the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump into ten liquid cooling random ranges respectively. The random probability of each liquid cooling random range is equal, and the random probability of each liquid cooling random range is ten percent; Obtaining two extreme value ranges within the liquid cooling random range. The extreme value ranges are denoted as the two lowest and highest extreme value ranges in the liquid cooling random range, and expanding the boundaries of the two extreme value ranges; For the lowest extreme value range, reducing the minimum value of the extreme value range by 10%. For the highest extreme value range, increasing the maximum value of the extreme value range by 10%; Performing random sampling based on the liquid cooling random ranges corresponding to the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump to obtain a first random liquid cooling sample; When performing random sampling, the sampling density of each liquid cooling random range is equal, and the distance between each first random liquid cooling sample sampled is greater than a preset threshold.

[0010] As a further solution of the present invention, defining a second random process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the random forest algorithm, and obtaining a second random liquid cooling sample based on the second random process, includes: Obtain the key features of liquid cooling, where the key features of liquid cooling are represented as the physical operating characteristics of an immersion liquid cooling system during operation. Random samples are randomly generated within the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump based on the random forest algorithm. The random samples are imported into the random forest algorithm, and decision tree splitting is performed based on the key features of liquid cooling to obtain the second random liquid cooling samples.

[0011] As a further solution of the present invention, perform the particle swarm optimization algorithm on the first random liquid cooling samples, and perform repeated simulation updates based on the particle swarm optimization algorithm to obtain the first optimized liquid cooling control parameters, including: Establish a two-dimensional space, map the first random liquid cooling samples into the two-dimensional space, and use the first random liquid cooling samples as the liquid cooling particle swarm. Use the random liquid cooling distribution unit power and random cooling water pump rotational speed included in the first random liquid cooling samples as liquid cooling particles; Initialize the liquid cooling particles, and randomly mutate 20% of the liquid cooling particles, where the amplitude of the random mutation is within the interval; Based on the particle swarm optimization algorithm, perform simulation updates on the initial positions of the liquid cooling particles, and retain 30% of the optimal liquid cooling particles in each simulation update for the next round of simulation updates; When the particle swarm optimization algorithm reaches the optimal solution, output the first optimized liquid cooling control parameters.

[0012] As a further solution of the present invention, perform the random forest algorithm on the second random liquid cooling samples, and perform repeated simulation updates based on the random forest algorithm to obtain the second optimized liquid cooling control parameters, including: Use the depth regression tree in the random forest algorithm to construct a liquid cooling prediction model. Import the second random liquid cooling samples into the liquid cooling prediction model, and perform simulation updates based on the liquid cooling prediction model. When the liquid cooling prediction model reaches the optimal solution, output the second optimized liquid cooling control parameters.

[0013] As a further solution of the present invention, obtain the set temperature of the liquid cooling system, and obtain the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump based on the set temperature of the liquid cooling system, including: Perform physical condition restrictions on the obtained power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump. The physical condition restrictions mean removing the liquid cooling distribution unit power that exceeds the actual operating power of the liquid cooling distribution unit within the power range of the liquid cooling distribution unit, and removing the cooling water pump rotational speed that exceeds the actual operating rotational speed of the cooling water pump within the rotational speed range of the cooling water pump.

[0014] A coolant flow rate control system for an immersion liquid cooling system, including: An acquisition module, which is used to acquire the set temperature of the liquid cooling system, and can acquire the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump based on the set temperature of the liquid cooling system; A first solution module, which can define a first stochastic process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the Monte Carlo algorithm, acquire a first random liquid cooling sample based on the first stochastic process, and can also acquire a first optimized liquid cooling control parameter based on the particle swarm optimization algorithm; A second solution module, which can define a second stochastic process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the random forest algorithm, acquire a second random liquid cooling sample based on the second stochastic process, and can also acquire a second optimized liquid cooling control parameter based on the random forest algorithm; A stability module, which is used to establish a second-order transfer function of the liquid cooling, perform stability judgment based on the root locus diagram of the second-order transfer function of the liquid cooling, and can also acquire an optimized liquid cooling control parameter or a fused liquid cooling control parameter based on the stability judgment; A liquid cooling control module, which can perform real-time control on the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the optimized liquid cooling control parameter or the fused liquid cooling control parameter.

[0015] Based on the above aspects, in the embodiment of the present application, first, the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump are acquired through the set temperature of the liquid cooling system. A first stochastic process is defined based on the Monte Carlo algorithm and a first random liquid cooling sample is acquired. A second stochastic process is defined based on the random forest algorithm and a second random liquid cooling sample is acquired. Subsequently, the particle swarm optimization algorithm is executed on the first random liquid cooling sample, the random forest algorithm is executed on the second random liquid cooling sample, and repeated simulations are performed for updating to acquire a first optimized liquid cooling control parameter and a second optimized liquid cooling control parameter. Then, a second-order transfer function of the liquid cooling is established, and an optimized liquid cooling control parameter or a fused liquid cooling control parameter is acquired through stability judgment. Finally, the liquid cooling control module can perform real-time control on the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the optimized liquid cooling control parameter or the fused liquid cooling control parameter, so as to regulate the coolant flow rate of the immersion liquid cooling system. By applying a hybrid algorithm to the regulation of the immersion liquid cooling system, it is avoided that a single algorithm falls into a local optimal solution, resulting in poor regulation of the coolant flow rate of the immersion liquid cooling system. And by performing real-time control on the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump through the optimized liquid cooling control parameter or the fused liquid cooling control parameter, the limitation of a single parameter setting is avoided. Through this method, the cooling effect of the immersion liquid cooling system on servers or IT devices can be improved. Description of the Drawings

[0016] Figure 1 It is a schematic execution flow diagram of a method for regulating the coolant flow rate of an immersion liquid cooling system provided by an embodiment of the present invention.

[0017] Figure 2 It is a schematic diagram of the execution process of stability judgment in a coolant flow rate regulation method for an immersion liquid cooling system provided by an embodiment of the present invention.

[0018] Figure 3 It is a flowchart of a coolant flow rate regulation method for an immersion liquid cooling system provided by an embodiment of the present invention.

[0019] Figure 4 It is a schematic diagram of a coolant flow rate regulation system for an immersion liquid cooling system provided by an embodiment of the present invention. Specific Embodiments

[0020] The following further describes in detail the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. The following embodiments are used to illustrate the technical solutions of the present invention, but cannot be used to limit the protection scope of the present invention. Embodiment:

[0021] As shown in the attached Figure 1 , Figure 2 , Figure 3 , Figure 4 : An embodiment of the present invention provides a coolant flow rate regulation method for an immersion liquid cooling system, which is applicable to an immersion liquid cooling system and includes the following steps: An embodiment of the present invention provides a coolant flow rate regulation system for an immersion liquid cooling system, which is applicable to liquid cooling control and includes: Step S1, obtain the set temperature of the liquid cooling system, and obtain the power range of the liquid cooling distribution unit and the speed range of the cooling water pump based on the set temperature of the liquid cooling system.

[0022] Specifically, physical condition restrictions are imposed on the obtained power range of the liquid cooling distribution unit and the speed range of the cooling water pump. The physical condition restrictions are expressed as removing the power of the liquid cooling distribution unit in the power range of the liquid cooling distribution unit that exceeds the actual operating power of the liquid cooling distribution unit, and removing the speed of the cooling water pump in the speed range of the cooling water pump that exceeds the actual operating speed of the cooling water pump.

[0023] Further, when the user sets the cooling temperature of the immersion liquid cooling cabinet through the liquid cooling system, when obtaining the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump, due to calculation errors or other reasons, the power and rotational speed within the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump may accidentally exceed the actual operating power range of the liquid cooling distribution unit or the actual operating rotational speed range of the cooling water pump. Without introducing physical condition restrictions, it will lead to incorrect regulation parameters in subsequent algorithm processing, resulting in the liquid cooling distribution unit or the cooling water pump being unable to achieve the expected optimization regulation effect when inputting to the liquid cooling regulation module and regulating the liquid cooling distribution unit and the cooling water pump through the liquid cooling regulation module. By introducing physical condition restrictions, the power and rotational speed within the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump are always within the actual operating power range of the liquid cooling distribution unit or the actual operating rotational speed range of the cooling water pump, providing a reliable data basis for subsequent algorithm processing.

[0024] Step S2, define the first random process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the Monte Carlo algorithm, define the second random process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the random forest algorithm, obtain the first random liquid cooling sample based on the first random process, and obtain the second random liquid cooling sample based on the second random process.

[0025] In this embodiment, the first random process includes: Step S21, divide the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump into ten liquid cooling random intervals respectively, the random probability of each liquid cooling random interval is equal, and the random probability of each liquid cooling random interval is ten percent.

[0026] It can be understood that the form adopted is the uniform distribution method in the Monte Carlo algorithm. The probability density of each point within the specified interval in the uniform distribution is constant, and a completely random sample coverage can be achieved without complex parameter adjustment, thereby ensuring the efficiency and fairness in the subsequent random sampling process. Moreover, a similar uniform sampling method is also adopted in the subsequent random sampling process, which can also reduce the computational amount of the overall process and increase the processing speed. The uniform distribution can effectively avoid sampling bias and provide a basic support for the subsequent algorithm random simulation.

[0027] Step S22, obtain two extreme value intervals within the liquid cooling random interval, the extreme value intervals are expressed as the two lowest and highest extreme value intervals in the liquid cooling random interval, and widen the boundaries of the two extreme value intervals.

[0028] Specifically, for the lowest extreme value interval, reduce the minimum value of the extreme value interval by 10%, and for the highest extreme value interval, increase the maximum value of the extreme value interval by 10%.

[0029] It is understandable that after dividing into the liquid cooling random intervals, it is necessary to widen the boundaries of the two extreme value intervals of the liquid cooling random intervals, so as to provide redundant intervals for subsequent algorithm processing and retain a safety margin.

[0030] Step S23: Based on the liquid cooling random intervals corresponding to the power interval of the liquid cooling distribution unit and the rotational speed interval of the cooling water pump, perform random sampling to obtain the first random liquid cooling samples.

[0031] Furthermore, when performing random sampling, the sampling density of each liquid cooling random interval is equal, and the distance between each of the first random liquid cooling samples sampled is greater than a preset threshold.

[0032] It is understandable that when performing random sampling, uniform sampling similar to the uniform distribution mentioned above is adopted, so that the number of the first random liquid cooling samples sampled within each liquid cooling random interval is equal, ensuring that random sample sampling can be realized within each liquid cooling random interval.

[0033] Furthermore, keeping the distance between each of the first random liquid cooling samples sampled greater than the preset threshold can maintain the minimum distance between the first random liquid cooling samples, preventing the random sampling from falling into local sampling. If it falls into local sampling, the difference between the first random liquid cooling samples sampled is small, and in the subsequent algorithm processing process, it is easy for the algorithm to fall into a local optimal solution. By keeping the distance at the minimum preset threshold, the randomness and discreteness between the first random liquid cooling samples are ensured, and the preset threshold can be adjusted based on the number of samples set by the user.

[0034] In this embodiment, the second random process is expressed as obtaining the key liquid cooling features. The key liquid cooling features are expressed as the physical operation features of the immersion liquid cooling system during operation. Based on the random forest algorithm, random samples are randomly generated within the power interval of the liquid cooling distribution unit and the rotational speed interval of the cooling water pump, and the random samples are imported into the random forest algorithm. Based on the key liquid cooling features, decision tree splitting is performed to obtain the second random liquid cooling samples.

[0035] It can be understood that the feature sampling based on the random forest algorithm is a data-driven sampling method based on ensemble learning and feature importance analysis. Its core lies in constructing a multi-dimensional feature space and a decision tree network to identify sensitive areas and guide the generation of parameter samples. When implementing specifically, by obtaining the key liquid cooling features, the key liquid cooling features can be expressed as parameters of feature categories such as thermodynamics, fluid mechanics, equipment status, and environmental parameters. The specific physical meanings can be expressed as physical meanings such as heat dissipation requirements, flow states, equipment health, and heat transfer boundary conditions. Then, a decision tree split is constructed through the key liquid cooling features. Subsequently, the randomly generated samples can be imported into the decision tree split, and the randomly generated random samples are screened through the decision tree split to obtain the second random liquid cooling samples.

[0036] Step S3: Perform a particle swarm optimization algorithm on the first random liquid cooling sample, and perform repeated simulation updates based on the particle swarm optimization algorithm to obtain the first optimized liquid cooling control parameters.

[0037] In this embodiment, step S3 includes: Step S31: Establish a two-dimensional space, map the first random liquid cooling sample into the two-dimensional space, and use the first random liquid cooling sample as the liquid cooling particle swarm. Use the power of the random liquid cooling distribution unit and the rotational speed of the random cooling water pump included in the first random liquid cooling sample as the liquid cooling particles.

[0038] Step S32: Initialize the liquid cooling particles, and randomly mutate 20% of them. The range of the random mutation is within the interval.

[0039] Step S33: Based on the particle swarm optimization algorithm, perform simulation updates on the initial positions of the liquid cooling particles. Each time a simulation update is performed, 30% of the optimal liquid cooling particles are retained and enter the next round of simulation update.

[0040] Step S34: When the particle swarm optimization algorithm reaches the optimal solution, output the first optimized liquid cooling control parameters.

[0041] In this embodiment, the particle swarm optimization algorithm uses the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump as decision variables, constructs an initial particle population through the first random sample. The position of each particle corresponds to a combination of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump. The velocity vector determines its search direction and step size, and a mutation mechanism is introduced therein. 20% of the liquid cooling particles are randomly mutated, which can be simply understood as randomly resetting the initial positions of 20% of the liquid cooling particles to escape from the local optimal solution. Subsequently, simulation updates can be performed through the objective function of the particle swarm optimization algorithm to solve for the optimal solution.

[0042] It can be understood that when generating and simulating the first random sample, the connection and mixing of the Monte Carlo algorithm and the particle swarm optimization algorithm are used to improve the diversity of the sample and the rapidity of solution. The Monte Carlo algorithm provides a diverse initial solution distribution for the particle swarm through global random sampling, avoiding the premature convergence problem caused by the aggregation of the initial particle population in the traditional particle swarm algorithm. At the same time, its large-scale exploration ability can enable the particle swarm optimization algorithm to maintain the diversity of the sample during the simulation update process and prevent the loss of potential optimal solutions. The particle swarm optimization algorithm directionally accelerates and optimizes the candidate solutions generated by the Monte Carlo through the group cooperation mechanism, and uses the dual guidance of the individual historical optimum and the group optimum to transform the random search into an efficient gradient approximation, improving the convergence speed and accuracy. The combination of the two not only retains the robustness of the Monte Carlo algorithm to multi-dimensional, non-linear and uncertain problems, but also can make up for the blind calculation redundancy of the Monte Carlo algorithm through the local refined search of the particle swarm optimization algorithm. In the application scenario of the immersion liquid cooling system in this embodiment, the problem of falling into local optimal solutions is reduced, and at the same time, the overall computing power can be reduced.

[0043] Step S4, perform a random forest algorithm on the second random liquid cooling sample, and perform repeated simulation updates based on the random forest algorithm to obtain a second optimized liquid cooling control parameter.

[0044] Specifically, a depth regression tree in the random forest algorithm is used to construct a liquid cooling prediction model. The second random liquid cooling sample is imported into the liquid cooling prediction model, and simulation updates are performed based on the liquid cooling prediction model. When the liquid cooling prediction model reaches the optimal solution, the second optimized liquid cooling control parameter is output.

[0045] Specifically, the depth regression tree in the random forest algorithm constructs a liquid cooling prediction model by integrating multiple depth regression trees. After inputting the second random liquid cooling sample, that is, inputting the parameter combination of the power of the liquid cooling distribution unit and the rotation speed of the cooling water pump, based on multi-dimensional features such as temperature gradient and equipment efficiency decay rate, the performance of energy efficiency, stability and response speed is predicted. In actual use, strategies such as Bagging and feature random masking can also be added to enhance the generalization model of the liquid cooling prediction model. After predicting the performance, the multi-objective optimization problem can be transformed into a single-objective sequence, gradually relaxed to generate multiple groups of optimal solutions, and then the feasibility of multiple groups of optimal solutions can be quickly evaluated through the updated simulation of the liquid cooling prediction model to achieve the global optimal solution.

[0046] It can be understood that in this embodiment, the first optimized liquid cooling control parameters are obtained through the Monte Carlo algorithm and the particle swarm optimization algorithm, and the second optimized liquid cooling control parameters are obtained through the random forest algorithm, providing a basis for subsequent stability judgment. When both of them meet the stability requirements, the first optimized liquid cooling control parameters and the second optimized liquid cooling control parameters can be combined to obtain a better cooling effect. If only one of the groups meets the stability requirements, it will be directly used.

[0047] Step S5: Establish a second-order liquid cooling transfer function, draw the root locus diagram of the second-order liquid cooling transfer function, and perform stability judgment based on the root locus diagram of the second-order liquid cooling transfer function.

[0048] Furthermore, when establishing the second-order liquid cooling transfer function, the thermal equation can be Laplace-transformed to obtain a second-order transfer function containing two inertia links and one integral link. The constants in the second-order liquid cooling transfer function can be identified through a frequency sweep experiment. When constructing the second-order liquid cooling transfer function, a non-linear compensation can be added to describe the flow state mutation in the immersion liquid cooling system and characterize the phase characteristics near the turbulent critical point. After the second-order liquid cooling transfer function, the root locus diagram of the second-order liquid cooling transfer function can be drawn based on the existing root locus drawing method.

[0049] Specifically, when drawing the root locus diagram of the second-order liquid cooling transfer function, a plane coordinate system is established. The X-axis in the plane coordinate system represents the real number, and the Y-axis in the plane coordinate system represents the complex number. When the root locus diagram of the second-order liquid cooling transfer function is in the second or third quadrant of the plane coordinate system, it is in a stable state. When the root locus diagram of the second-order liquid cooling transfer function is in the first or fourth quadrant of the plane coordinate system, it is in an unstable state.

[0050] It can be understood that the root locus diagram is a graphical tool used for analysis in a control system. It can intuitively show the movement trajectory when the closed-loop poles of the system change, thereby judging the stability and dynamic performance of the system. In the root locus diagram, the poles move along a specific path. When the poles are located in the left half of the complex plane, the system is in a stable state. The closer the poles are to the imaginary axis, the slower but smoother the response. The farther away from the imaginary axis, the faster the response. If the poles move to the right half plane, the system will be in an unstable state.

[0051] In this embodiment, step S5 includes: Step S51: Import the first optimized liquid cooling control parameters and the second optimized liquid cooling control parameters into the second-order liquid cooling transfer function, and find the positions of the first optimized liquid cooling control parameters and the second optimized liquid cooling control parameters on the root locus diagram of the liquid cooling transfer function.

[0052] Step S52: If both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in the second or third quadrant of the root locus diagram, then both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in a stable state. Perform a fusion operation on the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter, and obtain the fused liquid cooling control parameter based on the fusion operation.

[0053] In this embodiment, if both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in the second or third quadrant of the root locus diagram, it indicates that these two parameters are in a stable state and can be used for the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter.

[0054] Specifically, the fusion operation is expressed as performing a weighted sum on the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter. The calculation formula for the weighted sum is expressed as: ; where, represents the fused liquid cooling control parameter, represents the first optimized liquid cooling control parameter, represents the second optimized liquid cooling control parameter, represents the calculation weight of the first optimized liquid cooling control parameter, represents the calculation weight of the second optimized liquid cooling control parameter; When performing the calculation, a limiting condition is introduced, and the limiting condition is expressed as: , and .

[0055] It can be understood that the sum of the calculation weight of the first optimized liquid cooling control parameter and the calculation weight of the second optimized liquid cooling control parameter should be equal to 1. If the sum of the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter is not 1, then the fused liquid cooling control parameter generated through the fusion operation is an incorrect parameter. After importing it into the liquid cooling control module, the expected effect cannot be achieved. Moreover, the first optimized liquid cooling control parameter generated by simulating with the Monte Carlo algorithm and the particle swarm algorithm is superior to the second optimized control parameter generated by simulating with the random forest algorithm and the random forest algorithm. Therefore, the calculation weight of the first optimized liquid cooling control parameter is greater than the calculation weight of the second optimized control parameter, and their values should be relatively close. Otherwise, it will lead to a state where a certain optimized liquid cooling control parameter is in an absolute dominant position. For example, if the calculation weight value of the first optimized liquid cooling control parameter is 0.65, then the calculation weight value of the second optimized liquid cooling control parameter is 0.35.

[0056] Step S53, if a set of parameters among the first optimized liquid cooling control parameters and the second optimized liquid cooling control parameters is in the first quadrant or the fourth quadrant of the root locus plot, then this set of parameters among the first optimized liquid cooling control parameters and the second optimized liquid cooling control parameters is in an unstable state, and the other set of parameters in a stable state is obtained as the optimized liquid cooling control parameters.

[0057] In this embodiment, if the first optimized liquid cooling control parameters or the second optimized liquid cooling control parameters are in the first quadrant or the fourth quadrant of the root locus plot, it indicates that one set of parameters is not in a stable state, and this set of parameters needs to be excluded, and the remaining set of stable parameters can be used.

[0058] Step S54, if both the first optimized liquid cooling control parameters and the second optimized liquid cooling control parameters are in the first quadrant or the fourth quadrant of the root locus plot, then both the first optimized liquid cooling control parameters and the second optimized liquid cooling control parameters are in an unstable state, and step S2 is re-executed.

[0059] In this embodiment, if both the first optimized liquid cooling control parameters and the second optimized liquid cooling control parameters are in the first quadrant or the fourth quadrant of the root locus plot, it indicates that neither of these two sets of parameters is in a stable state, and neither set of parameters can be used. At this time, step S2 needs to be re-executed to perform a new random sampling through the Monte Carlo algorithm and the random forest algorithm.

[0060] Step S6, based on the stability judgment, the optimized liquid cooling control parameters or the integrated liquid cooling control parameters are obtained, and the optimized liquid cooling control parameters or the integrated liquid cooling control parameters are imported into the liquid cooling control module, and the liquid cooling control module can perform real-time control on the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump.

[0061] It can be understood that through step S5 in this embodiment, the optimized liquid cooling control parameters or the integrated liquid cooling control parameters are obtained, and the optimized liquid cooling control parameters or the integrated liquid cooling control parameters are imported into the liquid cooling control module. The liquid cooling control module can perform real-time adjustment on the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the above-mentioned optimized liquid cooling control parameters or the integrated liquid cooling control parameters, thereby improving the cooling effect of the immersion liquid cooling system on the IT devices or servers that are working in the liquid cooling cabinet, so that the IT devices or servers are in a stable temperature environment for a long time when processing high-density and high-load computing tasks.

[0062] The embodiment of the present invention provides a coolant flow rate control system for an immersion liquid cooling system, including: An acquisition module, the acquisition module is used to acquire the set temperature of the liquid cooling system, and can acquire the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump based on the set temperature of the liquid cooling system.

[0063] The first solving module, which can define a first stochastic process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the Monte Carlo algorithm, obtain a first random liquid cooling sample based on the first stochastic process, and can also obtain a first optimized liquid cooling regulation parameter based on the particle swarm optimization algorithm.

[0064] The second solving module, which can define a second stochastic process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the random forest algorithm, obtain a second random liquid cooling sample based on the second stochastic process, and can also obtain a second optimized liquid cooling regulation parameter based on the random forest algorithm.

[0065] The stability module, which is used to establish a second-order transfer function of the liquid cooling, perform stability judgment based on the root locus diagram of the second-order transfer function of the liquid cooling, and can also obtain an optimized liquid cooling regulation parameter or a fused liquid cooling regulation parameter based on the stability judgment.

[0066] The liquid cooling regulation module, which can perform real-time control of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the optimized liquid cooling regulation parameter or the fused liquid cooling regulation parameter.

[0067] The specific usage mode and function of this embodiment: First, obtain the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump through the set temperature of the liquid cooling system, define the first stochastic process based on the Monte Carlo algorithm and obtain the first random liquid cooling sample, define the second stochastic process based on the random forest algorithm and obtain the second random liquid cooling sample. Subsequently, perform the particle swarm optimization algorithm on the first random liquid cooling sample, perform the random forest algorithm on the second random liquid cooling sample, and perform repeated simulation updates to obtain the first optimized liquid cooling regulation parameter and the second optimized liquid cooling regulation parameter. Then, establish a second-order transfer function of the liquid cooling, and obtain the optimized liquid cooling regulation parameter or the fused liquid cooling regulation parameter through stability judgment. Finally, the liquid cooling regulation module can perform real-time control of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the optimized liquid cooling regulation parameter or the fused liquid cooling regulation parameter, so as to regulate the coolant flow rate of the immersion liquid cooling system. By applying the hybrid algorithm to the regulation of the immersion liquid cooling system, it is avoided that a single algorithm falls into a local optimal solution, resulting in poor regulation of the coolant flow rate of the immersion liquid cooling system. And by performing real-time control of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump through the optimized liquid cooling regulation parameter or the fused liquid cooling regulation parameter, the limitation of single parameter setting is avoided. Through this method, the cooling effect of the immersion liquid cooling system on servers or IT devices can be improved.

[0068] In addition, an embodiment of the present invention further provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method in the above embodiment.

[0069] The following specifically introduces each component of the electronic device: Among them, the processor is the control center of the electronic device, which can be a single processor or a collective term for multiple processing elements. For example, the processor is one or more central processing units (CPUs), or can be an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention, such as: one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs).

[0070] Among them, the processor can execute various functions of the electronic device by running or executing software programs stored in the memory and calling data stored in the memory.

[0071] The memory is used to store the software program for implementing the solution of the present invention and is controlled by the processor for execution. The specific implementation manner can refer to the above method embodiment and will not be elaborated here.

[0072] The memory can be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or can also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory can be integrated with the processor or exist independently and be coupled to the processor through the interface circuit of the electronic device. The embodiments of the present invention do not make specific limitations on this.

[0073] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0074] It should be understood that the term "and / or" in this text is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this text generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context before and after.

[0075] It should be understood that in the embodiments of the present invention, the magnitudes of the serial numbers of the above processes do not mean the sequence of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0076] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for regulating the coolant flow rate of an immersion liquid cooling system, characterized in that The method includes: S1: Obtain the set temperature of the liquid cooling system, and obtain the power range of the liquid cooling distribution unit and the rotational speed range of the cooling water pump based on the set temperature of the liquid cooling system; S2: Define a first stochastic process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the Monte Carlo algorithm, obtain a first random liquid cooling sample based on the first stochastic process, define a second stochastic process of the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump based on the random forest algorithm, and obtain a second random liquid cooling sample based on the second stochastic process; S3: Execute the particle swarm optimization algorithm on the first random liquid cooling sample, perform repeated simulation updates based on the particle swarm optimization algorithm, and obtain a first optimized liquid cooling control parameter; S4: Execute the random forest algorithm on the second random liquid cooling sample, perform repeated simulation updates based on the random forest algorithm, and obtain a second optimized liquid cooling control parameter; S5: Establish a second-order transfer function of the liquid cooling, draw the root locus diagram of the second-order transfer function of the liquid cooling, perform stability judgment based on the root locus diagram of the second-order transfer function of the liquid cooling, and obtain an optimized liquid cooling control parameter or a fused liquid cooling control parameter based on the stability judgment; S6: Import the optimized liquid cooling control parameter or the fused liquid cooling control parameter into the liquid cooling control module, and the power of the liquid cooling distribution unit and the rotational speed of the cooling water pump can be controlled in real time through the liquid cooling control module.

2. The coolant flow rate regulation method of an immersion liquid cooling system according to claim 1, wherein The stability judgment based on the root locus diagram of the second-order transfer function of the liquid cooling, and obtaining an optimized liquid cooling control parameter or a fused liquid cooling control parameter based on the stability judgment includes: Import the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter into the second-order transfer function of the liquid cooling, and find the positions of the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter on the root locus diagram of the liquid cooling transfer function; If both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in the second or third quadrant of the root locus diagram, then both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in a stable state, perform a fusion operation on the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter, and obtain the fused liquid cooling control parameter based on the fusion operation; If one of the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter is in the first quadrant or the fourth quadrant of the root locus diagram, then one of the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter is in an unstable state, and obtain the other group of parameters in the stable state as the optimized liquid cooling control parameter; If both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in the first quadrant or the fourth quadrant of the root locus diagram, then both the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter are in an unstable state, and re-execute step S2.

3. The coolant flow rate regulation method of an immersion liquid cooling system according to claim 2, wherein The fusion operation on the first optimized liquid cooling control parameter and the second optimized liquid cooling control parameter, and obtaining the fused liquid cooling control parameter based on the fusion operation includes: The fusion operation is expressed as a weighted sum of the first optimized liquid cooling regulation parameter and the second optimized liquid cooling regulation parameter, and the calculation formula for the weighted sum is expressed as: ； Among them, is expressed as the integrated liquid cooling control parameter, is expressed as the first optimized liquid cooling control parameter, is expressed as the second optimized liquid cooling control parameter, is expressed as the calculation weight of the first optimized liquid cooling control parameter, is expressed as the calculation weight of the second optimized liquid cooling control parameter; When performing the calculation, a constraint condition is introduced, and the constraint condition is expressed as: , and .

4. A method for regulating the coolant flow rate of an immersion liquid cooling system according to claim 1, characterized in that, The establishment of the liquid cooling second-order transfer function and the drawing of the root locus diagram of the liquid cooling second-order transfer function include: When drawing the root locus diagram of the liquid cooling second-order transfer function, a plane coordinate system is established. The X-axis in the plane coordinate system represents the real number, and the Y-axis in the plane coordinate system represents the complex number. When the root locus diagram of the liquid cooling second-order transfer function is in the second or third quadrant of the plane coordinate system, it is in a stable state. When the root locus diagram of the liquid cooling second-order transfer function is in the first or fourth quadrant of the plane coordinate system, it is in an unstable state.

5. A method for regulating the coolant flow rate of an immersion liquid cooling system according to claim 1, characterized in that, The definition of the first random process of the liquid cooling distribution unit power and the cooling water pump speed based on the Monte Carlo algorithm, and the acquisition of the first random liquid cooling sample based on the first random process, include: The liquid cooling distribution unit power interval and the cooling water pump speed interval are respectively divided into ten liquid cooling random intervals, and the random probability of each liquid cooling random interval is equal, and the random probability of each liquid cooling random interval is ten percent; Two extreme value intervals within the liquid cooling random interval are obtained. The extreme value intervals are expressed as the two lowest and highest extreme value intervals in the liquid cooling random interval, and the boundaries of the two extreme value intervals are widened; For the lowest extreme value interval, the minimum value of the extreme value interval is decreased by 10%. For the highest extreme value interval, the maximum value of the extreme value interval is increased by 10%; Based on the liquid cooling random intervals corresponding to the liquid cooling distribution unit power interval and the cooling water pump speed interval, random sampling is performed to obtain the first random liquid cooling sample; When performing random sampling, the sampling density of each liquid cooling random interval is equal, and the distance between each of the first random liquid cooling samples sampled is greater than a preset threshold.

6. The coolant flow rate regulation method of an immersion liquid cooling system according to claim 1, characterized in that The definition of the second random process of the liquid cooling distribution unit power and the cooling water pump speed based on the random forest algorithm, and the acquisition of the second random liquid cooling sample based on the second random process, include: The key liquid cooling features are obtained. The key liquid cooling features are expressed as the physical operation features of the immersion liquid cooling system during operation. Based on the random forest algorithm, random samples are randomly generated within the liquid cooling distribution unit power interval and the cooling water pump speed interval, and the random samples are imported into the random forest algorithm. Based on the key liquid cooling features, decision tree splitting is performed to obtain the second random liquid cooling sample.

7. A method for regulating the coolant flow rate of an immersion liquid cooling system according to claim 1, characterized in that, The execution of the particle swarm optimization algorithm on the first random liquid cooling sample, and the repeated simulation update based on the particle swarm optimization algorithm to obtain the first optimized liquid cooling regulation parameter, include: A two-dimensional space is established, the first random liquid cooling sample is mapped into the two-dimensional space, and the first random liquid cooling sample is used as the liquid cooling particle swarm. The random liquid cooling distribution unit power and the random cooling water pump speed included in the first random liquid cooling sample are used as liquid cooling particles; Initialize the liquid-cooled particles, and randomly mutate 20% of the liquid-cooled particles, where the amplitude of the random mutation is within the interval; The initial positions of the liquid-cooled particles are simulated and updated based on the particle swarm optimization algorithm, and 30% of the optimal liquid-cooled particles are retained for the next round of simulation and update each time; When the particle swarm optimization algorithm reaches the optimal solution, the first optimized liquid-cooling control parameter is output.

8. The coolant flow rate control method of an immersion liquid cooling system according to claim 1, characterized in that, Performing a random forest algorithm on the second random liquid-cooling sample, and performing repeated simulation and update based on the random forest algorithm to obtain a second optimized liquid-cooling control parameter, including: Using a depth regression tree in the random forest algorithm to construct a liquid-cooling prediction model, importing the second random liquid-cooling sample into the liquid-cooling prediction model, performing simulation and update based on the liquid-cooling prediction model, and when the liquid-cooling prediction model reaches the optimal solution, outputting the second optimized liquid-cooling control parameter.

9. The coolant flow rate regulation method of an immersion liquid cooling system according to claim 1, characterized in that, Obtaining the set temperature of the liquid-cooling system, and obtaining the power range of the liquid-cooling distribution unit and the rotational speed range of the cooling water pump based on the set temperature of the liquid-cooling system, including: Performing physical condition restrictions on the obtained power range of the liquid-cooling distribution unit and the rotational speed range of the cooling water pump. The physical condition restrictions are expressed as eliminating the power of the liquid-cooling distribution unit in the power range of the liquid-cooling distribution unit that exceeds the actual operating power of the liquid-cooling distribution unit, and eliminating the rotational speed of the cooling water pump in the rotational speed range of the cooling water pump that exceeds the actual operating rotational speed of the cooling water pump.

10. A coolant flow rate control system for an immersion liquid cooling system, characterized in that, Including: An acquisition module, which is used to acquire the set temperature of the liquid-cooling system, and can obtain the power range of the liquid-cooling distribution unit and the rotational speed range of the cooling water pump based on the set temperature of the liquid-cooling system; A first solution module, which can define a first random process of the power of the liquid-cooling distribution unit and the rotational speed of the cooling water pump based on the Monte Carlo algorithm, obtain a first random liquid-cooling sample based on the first random process, and can also obtain a first optimized liquid-cooling control parameter based on the particle swarm optimization algorithm; A second solution module, which can define a second random process of the power of the liquid-cooling distribution unit and the rotational speed of the cooling water pump based on the random forest algorithm, obtain a second random liquid-cooling sample based on the second random process, and can also obtain a second optimized liquid-cooling control parameter based on the random forest algorithm; A stability module, which is used to establish a liquid-cooling second-order transfer function, perform stability judgment based on the root locus diagram of the liquid-cooling second-order transfer function, and can also obtain an optimized liquid-cooling control parameter or a fused liquid-cooling control parameter based on the stability judgment; A liquid-cooling control module, which can perform real-time control on the power of the liquid-cooling distribution unit and the rotational speed of the cooling water pump based on the optimized liquid-cooling control parameter or the fused liquid-cooling control parameter.

Images