Dryness perception and branch impedance regulation two-phase liquid cooling system liquid supply control method

Abstract

The application discloses a two-phase liquid cooling system liquid supply control method for dryness perception and branch impedance regulation, and particularly relates to the technical field of two-phase liquid cooling heat dissipation control; a dryness evolution relationship model is constructed based on temperature sequences of the inlet and outlet of an evaporation section to obtain a dryness prediction value; and equivalent impedance is calculated in combination with branch inlet pressure, outlet pressure and mass flow rate to determine the target opening of a liquid supply control valve; a dry-out risk evaluation model is established through the dryness prediction value and the wall temperature difference, the branch impedance model is corrected when the risk exceeds a threshold value, and the updated impedance is used for next cycle regulation; further based on the dryness prediction value and the impedance, liquid supply distribution is simulated, a liquid supply uniformity parameter is constructed, and the dryness evolution coefficient is automatically corrected when the liquid supply deviation is large; the application can improve the dryness prediction accuracy, inhibit flow backflow and air lock, reduce the dry-out risk of the evaporation section, and improve the stability and temperature field uniformity of the two-phase liquid cooling system.

Description

Technical Field

[0001] This invention relates to the field of two-phase liquid cooling heat dissipation control technology, specifically to a liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation. Background Technology

[0002] With the continuous development of high-power-density electronic devices, two-phase liquid cooling has become a key cooling method due to its ability to efficiently dissipate heat by utilizing the latent heat of fluid phase change. However, the vapor-liquid distribution within the two-phase region is extremely unstable, and the local dryness of the evaporation section fluctuates transiently with heat flux density, easily leading to liquid film rupture and desiccation, causing the wall temperature to rise sharply in milliseconds. Simultaneously, the flow impedance of each branch in the system exhibits strong nonlinearity due to changes in the vapor-liquid integral, resulting in random fluctuations in branch flow distribution. This can cause repeated gas lock formation and significant insufficient liquid at the evaporator inlet, or even lead to overall thermal failure. Currently, liquid supply control generally relies on open-loop regulation based on pressure or temperature differences, which cannot accurately obtain dryness or adjust branch impedance in real time, making it difficult for two-phase liquid cooling systems to maintain stable operation under sudden load changes or space constraints. Therefore, there is an urgent need for a liquid supply control method that can sense dryness in real time and dynamically adjust branch impedance to ensure that the two-phase liquid cooling system can maintain uniform vapor-liquid distribution and evaporation section safety under complex operating conditions. Summary of the Invention

[0003] The purpose of this invention is to provide a liquid supply control method for a two-phase liquid cooling system that uses dryness sensing and branch impedance regulation to address the shortcomings in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation, comprising:

[0005] S100: Obtain the temperature sequence T between the inlet and outlet of the evaporation section of the two-phase liquid cooling system, and calculate the initial dryness estimate X0 of the evaporation section based on the rate of change of the temperature sequence T.

[0006] S200: Based on the initial dryness estimate X0 and the structural parameters of the evaporation section, construct the dryness evolution relationship model SX of the two-phase region, and obtain the dryness evolution coefficient S1 based on the model;

[0007] S300: Based on the dryness evolution coefficient S1, the inlet pressure and mass flow rate of each branch in the two-phase liquid cooling system are jointly analyzed to obtain the equivalent impedance Zi of each branch.

[0008] S400: Based on the equivalent impedance Zi, determine the target opening Vi of the liquid supply control valve so that the branch with high impedance gets high liquid supply priority; and input the target opening Vi into the dryness evolution relationship model SX to calculate the dryness prediction value X1.

[0009] S500: Calculate the drying risk parameter D based on the predicted dryness value X1 and the temperature difference between the evaporation section wall;

[0010] S600: If the drying risk parameter D exceeds the preset threshold, the branch impedance model SZ is corrected based on the dryness prediction value X1 and the equivalent impedance Zi to obtain a new branch impedance Zi', and the process returns to step S400; otherwise, proceed to step S700.

[0011] S700: Based on the dryness prediction value X1 and the equivalent impedance Zi or Zi', the overall liquid supply distribution of the two-phase liquid cooling system is simulated to obtain the liquid supply balance parameter W.

[0012] S800: If the liquid supply balance parameter W does not reach the stable condition, then input W into the dryness evolution relationship model SX to obtain the updated dryness evolution coefficient S2, and return to step S300; if the stable condition is reached, then output the final valve opening of each branch and the steady-state dryness distribution of the system.

[0013] Preferably, S100 further includes:

[0014] A first temperature sensor and a second temperature sensor are installed at the inlet and outlet of the evaporation section, respectively, and multiple sets of temperature data are collected synchronously at a fixed sampling period to form a temperature sequence T ordered by time.

[0015] The instantaneous temperature gradient between any adjacent sampling points is calculated based on the temperature sequence T, and a moving average is performed through a set time window to obtain the stable temperature change rate parameter R1 of the evaporation section.

[0016] Input the steady temperature change rate parameter R1 into the preset dryness-sensitive response function model, and modify the model according to the structural parameters of the evaporation section to obtain the intermediate quantity X01 for dryness estimation.

[0017] A linear mapping relationship is established between the intermediate dryness estimate X01 and the reference temperature difference at the inlet of the evaporation section to obtain the initial dryness estimate X0 of the evaporation section.

[0018] Preferably, S200 further includes:

[0019] Obtain the structural parameters of the evaporation section, including the length of the evaporation section, the cross-sectional area of ​​the flow channel, the wetted perimeter of the flow channel, and the surface roughness of the evaporation section wall, and calculate the equivalent heat transfer area parameter A1 of the evaporation section based on the structural parameters;

[0020] Using the initial dryness estimate X0 as the input variable, and combining it with the equivalent heat transfer area parameter A1, a dryness evolution relationship model SX for the two-phase region is constructed. The dryness evolution relationship model SX establishes a functional relationship between the dryness change rate dX / dt and the instantaneous liquid supply and heat flux density of the evaporation section in a time-continuous differential form.

[0021] The parameters of the dryness evolution model SX are identified. The error between the rate of change of the inlet temperature of the evaporation section and the dryness predicted by the model is calculated. The dryness evolution parameter set P1 is solved by the least squares method.

[0022] Input the set of dryness evolution parameters P1 into the dryness evolution relationship model SX, and linearize the model output to obtain the dryness evolution coefficient S1.

[0023] Preferably, S300 further includes:

[0024] Inlet and outlet pressure data are collected at the inlet and outlet locations of each branch. The pressure difference between the inlet and outlet pressures is calculated to obtain the pressure difference parameter ΔPi, which reflects the resistance distribution characteristics of the two-phase flow.

[0025] Based on the dryness evolution coefficient S1, the mass flow rate measured in real time for each branch is normalized.

[0026] After obtaining the pressure difference parameter ΔPi and the normalized mass flow rate Mi1, a two-phase flow correction factor K1 is constructed based on the dryness evolution coefficient S1.

[0027] The pressure difference parameter ΔPi, the normalized mass flow rate Mi1, and the two-phase flow correction factor K1 are input into the impedance calculation model to calculate the equivalent impedance Zi corresponding to each branch.

[0028] Preferably, S400 further includes:

[0029] Sort the equivalent impedance Zi of each branch according to its numerical value, and generate an impedance priority sequence P1 according to the impedance from largest to smallest.

[0030] Based on the impedance priority sequence P1, the equivalent impedance Zi of each branch is input into the preset valve opening mapping function to calculate the target opening Vi of the liquid supply control valve.

[0031] The target opening degree Vi is input into the dryness evolution relationship model SX, and the liquid supply function m(t) in the model is updated by using the influence of valve opening degree on instantaneous liquid supply.

[0032] Based on the updated liquid supply function m(t) and the parameter set in the dryness evolution relationship model SX, the predicted dryness value X1 is calculated.

[0033] Preferably, S500 further includes:

[0034] Wall temperature data is acquired by multiple wall temperature sensors installed at different locations on the evaporation section wall, and the wall temperature data is compared with the evaporation section inlet temperature to form a wall temperature difference parameter that can reflect the degree of local overheating in the evaporation section.

[0035] Input the predicted dryness value X1 into the preset dryness-sensitive model, and the model outputs a dryness response factor that can characterize the strength of the influence of dryness changes on liquid film stability.

[0036] Based on the wall temperature difference parameters, the wall heat load factor is generated by utilizing the correspondence between the temperature difference and the wall heating condition.

[0037] The dryness response factor and the wall heat load factor are combined according to the set weights to obtain the dryness risk parameter D, which characterizes the severity of dryness in the evaporation section.

[0038] Preferably, S600 further includes:

[0039] The predicted dryness value X1 is compared with the initial estimated dryness value X0 to obtain the dryness deviation amount, which is used to characterize the degree of dryness deviation in the evaporation section. The correction direction and correction magnitude of the branch impedance model SZ are determined by the dryness deviation amount.

[0040] Based on the real-time changing trends of dryness deviation and equivalent impedance Zi, an impedance response factor is constructed.

[0041] The impedance response factor is correlated with the equivalent impedance Zi, and an impedance correction amount is generated based on the coupling relationship between impedance and dryness.

[0042] The impedance correction is input into the branch impedance model SZ, and the parameters of the branch impedance model SZ are updated to obtain the updated impedance Zi' for subsequent liquid supply regulation.

[0043] Preferably, S700 further includes:

[0044] The dryness prediction value X1 is integrated with the equivalent impedance Zi or updated impedance Zi' determined by the previous steps to establish a set of liquid supply simulation inputs to describe the liquid supply status of each branch.

[0045] Based on the aforementioned liquid supply simulation input set, a preset two-phase flow distribution solution model is invoked. The model calculates the achievable mass flow rate of each branch under a given impedance condition to obtain the flow distribution result used to characterize the flow distribution trend of multiple branches.

[0046] Based on the flow distribution results, the mass flow data of each branch is extracted, and a liquid supply balance evaluation function is constructed according to the flow dispersion of multiple branches to obtain the liquid supply balance parameter W, which reflects the consistency of liquid supply in multiple branches.

[0047] Preferably, S800 further includes:

[0048] The liquid supply balance parameter W is compared with the preset stability condition. When the liquid supply balance parameter W is higher than the stability condition, it is determined that the liquid supply distribution of the two-phase liquid cooling has not yet reached the flow balance state, and the dryness evolution process needs to be corrected.

[0049] Based on the degree of deviation between the liquid supply balance parameter W and the stability condition, a balance deviation quantity is generated to characterize the strength of the liquid supply imbalance, so that the balance deviation quantity can reflect the influence of the flow dispersion of each branch on the dryness change trend.

[0050] Input the equilibrium deviation into the dryness evolution relationship model SX, and adjust the weights of the liquid supply related terms in the dryness evolution relationship model SX.

[0051] Based on the dryness evolution relationship model SX after the deviation input, the model output is linearized to obtain a new dryness evolution coefficient S2.

[0052] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0053] 1. This invention achieves real-time response to transient thermal flow disturbances in a two-phase liquid cooling system by constructing a closed-loop control chain consisting of dryness prediction, impedance identification, liquid supply regulation, and risk assessment. By building a dryness evolution model within the evaporation section, the predicted dryness value accurately reflects the dynamic changes in gas-liquid distribution, thus providing a reliable basis for impedance calculation and liquid supply regulation. By introducing impedance response factors, wall heat load factors, and drying risk parameters, the system can identify potential drying risks and take preventative control measures, effectively avoiding a sharp rise in wall temperature due to localized liquid film rupture, significantly improving the thermal stability and operational safety of the evaporation section. This invention achieves visualization and controllability of key variables in the two-phase flow, improves the accuracy of dryness prediction, and allows liquid supply regulation to move beyond fixed valve openings or empirical strategies, instead relying on dynamic optimization based on actual operating conditions, greatly enhancing system reliability.

[0054] 2. This invention, by introducing liquid supply balance evaluation, adaptive impedance model correction, and iterative updating of the dryness evolution coefficient into the control process, enables the two-phase liquid cooling system to maintain stable gas-liquid distribution across multiple branches even under high heat flux density conditions. The introduction of the liquid supply balance parameter W allows the system to continuously monitor the dispersion of flow rates across multiple branches and automatically correct the dryness evolution relationship model SX based on the deviation, ensuring it remains consistent with the current operating conditions. Through the synergistic effect of the nonlinear response model, parameter calibration method, and dynamic adjustment strategy, this invention achieves closed-loop stable control of the liquid supply, dryness, and impedance even under strong disturbances and high fluctuations, effectively suppressing typical two-phase liquid cooling failure modes such as backflow, gas lock formation, and evaporation section drying, thereby improving overall heat dissipation capacity and temperature field uniformity. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0056] Figure 1 This is a flowchart of the liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to the present invention.

[0057] Figure 2 This is a flowchart of the branch impedance model correction method of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Example 1, please refer to Figure 1 , 2 As shown in this embodiment, the liquid supply control method for a two-phase liquid cooling system with dryness sensing and branch impedance regulation includes:

[0060] S100: Obtain the temperature sequence T between the inlet and outlet of the evaporation section of the two-phase liquid cooling system, and calculate the initial dryness estimate X0 of the evaporation section based on the rate of change of the temperature sequence T.

[0061] During implementation, a first temperature sensor is installed at the inlet of the two-phase liquid-cooled evaporation section, and a second temperature sensor is installed at the outlet of the evaporation section. Both the first and second temperature sensors use the same type of temperature measuring element to ensure temperature measurement consistency. The first and second temperature sensors synchronously acquire data at a preset sampling period, which is a fixed time interval. The length of the time interval is set according to the rate of change of heat load, for example, 50 milliseconds to 200 milliseconds. After multiple sets of temperature data are continuously acquired, the temperature data from the first and second temperature sensors are arranged in chronological order of acquisition time to form a temperature sequence T ordered in ascending order of time. The temperature sequence T is the basic data set used for subsequent calculation of temperature gradient and initial dryness estimate, and it has continuity and time correlation.

[0062] After obtaining the temperature sequence T, a difference operation is performed on the temperature data corresponding to any adjacent sampling time to calculate the instantaneous temperature gradient between consecutive sampling points. The instantaneous temperature gradient is defined as the temperature difference between two adjacent temperature sampling points divided by the sampling period. Due to the strong transient fluctuation characteristics of the two-phase flow process, to avoid abnormal deviations in gradient calculation caused by instantaneous disturbances, all instantaneous temperature gradients need to be smoothed using a moving average algorithm. The length of the moving average time window is set according to the frequency of temperature fluctuations, for example, 3 to 10 sampling periods. The average of multiple instantaneous temperature gradients within the window is calculated to form the stable temperature change rate parameter R1. The stable temperature change rate parameter R1 can reflect the temperature change trend of the evaporation section on a macroscopic time scale, providing a stable input for subsequent dryness fraction estimation.

[0063] After obtaining the steady-state temperature change rate parameter R1, it needs to be input into a preset dryness-sensitive response function model. The dryness-sensitive response function model is a mathematical model expressing the relationship between the dryness of the evaporation section and the temperature change rate. The model is based on the boiling heat transfer mechanism and adopts the following functional form: Where a1, a2, and a3 are sensitivity coefficients calibrated based on the structural parameters of the evaporation section, the channel dimensions, the type of working fluid, and experimental data. The dryness-sensitive response function model adopts a quadratic function form to improve the model's nonlinear response capability to low and high dryness regions. When modifying the dryness-sensitive response function model, the sensitivity coefficients are adjusted according to the actual structural parameters of the evaporation section (including the evaporation section length, inner diameter, channel shape, wall roughness, etc.) so that the intermediate dryness estimate X01 output by the model more closely matches the actual two-phase flow conditions.

[0064] After obtaining the intermediate dryness fraction estimate X01, it is necessary to perform a linear mapping on X01 based on the baseline difference in the evaporation section inlet temperature to ensure the consistency between the estimated result and the actual dryness fraction of the evaporation section. The baseline difference in the evaporation section inlet temperature ΔTin is defined as the temperature difference between the evaporation section inlet temperature and the saturation temperature, satisfying the following relationship: Where Tin is the inlet temperature of the evaporation section, and Tsat is the saturation temperature of the working fluid at the inlet pressure. The initial dryness estimate X0 of the evaporation section is obtained by linearly combining the intermediate dryness estimate X01 with the inlet temperature baseline difference ΔTin. The calculation method is as follows: Where b1, b2, and b3 are mapping coefficients calibrated based on the properties of the working fluid and the heat transfer characteristics of the evaporation section. The initial dryness estimate X0 serves as the input parameter for the subsequent dryness evolution relationship model, providing the basic conditions for constructing a dynamic dryness prediction model.

[0065] S200: Based on the initial dryness estimate X0 and the structural parameters of the evaporation section, construct the dryness evolution relationship model SX of the two-phase region, and obtain the dryness evolution coefficient S1 based on the model.

[0066] During implementation, to construct a model of the dryness evolution relationship in the two-phase region, it is necessary to first obtain the structural parameters of the evaporation section. These parameters include the evaporation section length L1, the cross-sectional area A2 of the flow channel, the wetted perimeter U1 of the flow channel, and the surface roughness Q1 of the evaporation section wall. The evaporation section length L1 is obtained by measuring the axial distance from the inlet to the outlet of the evaporation section; the cross-sectional area A2 is calculated by measuring the geometry of the internal flow channel of the evaporation section and applying the corresponding geometric formula; the wetted perimeter U1 is determined based on the circumference of the inner wall profile of the flow channel; and the surface roughness Q1 of the evaporation section wall is measured using a surface profilometer.

[0067] After obtaining the above parameters, the equivalent heat transfer area parameter A1 of the evaporation section is calculated based on heat transfer theory. The equivalent heat transfer area parameter A1 is determined by the following formula: A1=L1×U1; the equivalent heat transfer area parameter A1 is used to describe the effective area of ​​the evaporation section wall that can exchange vaporization heat with the working fluid, and this parameter will directly affect the dryness response to heat flow disturbances.

[0068] After obtaining the initial dryness fraction estimate X0 and the equivalent heat transfer area parameter A1, a dryness fraction evolution model SX for the two-phase region is constructed. This dryness fraction evolution model SX adopts a time-continuous differential form, and the model is defined as follows: Where: dX / dt represents the rate of change of dryness; q(t) is the instantaneous heat flux density of the evaporation section at time t; m(t) is the instantaneous liquid supply mass flow rate at time t; C1 and C2 are the dryness change coefficients, used to describe the influence of heat flux and liquid supply on the rate of change of dryness.

[0069] The specific process of constructing the dryness evolution relationship model SX is as follows:

[0070] Use the initial dryness estimate X0 as the initial value for the model;

[0071] The heat flux density term is normalized using the equivalent heat transfer area parameter A1, making the model applicable to evaporation sections with different structural parameters.

[0072] The model uses a differential form to express the dynamic characteristics of dryness over time, enabling it to accurately reflect the evolution of dryness under transient conditions.

[0073] After constructing the aridity evolution model SX, it is necessary to identify the model parameters to determine the aridity variation coefficients C1 and C2 in the model. The parameter identification process is based on the principle of minimizing the error between the rate of change of the evaporation section inlet temperature and the aridity predicted by the model.

[0074] First, calculate the rate of change of the inlet temperature of the evaporation section, R2, based on the aforementioned temperature sequence T. R2 is defined as: Secondly, an error function E is established by relating the output dryness Xmodel(t) of the dryness evolution model SX to the inlet temperature change rate R2: Where k3 is the proportionality coefficient between the temperature change rate and the dryness change. The least squares method is used to minimize the error function E, and then the model parameter set P1 is solved, where P1 = {C1, C2, k3}. After parameter identification, the dryness evolution model SX can adaptively adjust according to temperature changes under real working conditions, resulting in high accuracy in the model's output dryness.

[0075] After obtaining the set of dryness evolution parameters P1, input the set of dryness evolution parameters P1 into the dryness evolution relationship model SX, and linearize the output of the model to obtain the dryness evolution coefficient S1.

[0076] The linearization process is based on the first-order expression of the Taylor expansion, as shown in the following formula: The dryness fraction evolution coefficient S1 reflects the slope of the response of dryness fraction to changes in instantaneous heat flux density and is a key input parameter for subsequent branch impedance calculation steps. In practical applications, the dryness fraction evolution coefficient S1 is updated as the dryness fraction parameter set P1 changes, thereby ensuring that the dryness fraction evolution relationship model can adapt to changes in the heat load of the evaporation section in real time.

[0077] S300: Based on the dryness evolution coefficient S1, the inlet pressure and mass flow rate of each branch in the two-phase liquid cooling system are jointly analyzed to obtain the equivalent impedance Zi of each branch.

[0078] To calculate the equivalent impedance of each branch, it is first necessary to obtain the pressure difference parameter ΔPi, which reflects the distribution of flow resistance. Inlet pressure sensors are installed at the inlet location of each branch, and outlet pressure sensors are installed at the corresponding outlet location. The inlet pressure sensors measure the branch inlet pressure Piin, and the outlet pressure sensors measure the branch outlet pressure Piout.

[0079] Within each sampling period, the inlet pressure Piin and the outlet pressure Piout are differentially calculated to obtain the pressure difference parameter ΔPi, specifically calculated as follows: ΔPi = Piin - Piout; where: Piin: branch inlet pressure, unit Pa; Piout: branch outlet pressure, unit Pa; ΔPi: pressure difference parameter, representing the pressure drop across the branch. The pressure difference parameter ΔPi characterizes the flow resistance generated by frictional pressure drop, phase change acceleration pressure drop, and local resistance pressure drop within the branch, and is the core input for subsequent equivalent impedance calculation.

[0080] To enhance the sensitivity of mass flow rate to changes in gas-liquid ratio, it is necessary to normalize the real-time measured mass flow rate Mi of each branch based on the dryness evolution coefficient S1.

[0081] The real-time measured mass flow rate is denoted as Mi, and this mass flow rate is also affected by changes in dryness fraction. To eliminate the deviation in mass flow rate under different dryness fractions, a normalization model is introduced: Where: Mi: Mass flow rate measured in real time by the branch, in kg / s; S1: Dryness evolution coefficient, used to characterize the degree of response of dryness to thermal flow disturbance; Mi1: Normalized mass flow rate, so that the mass flow rate can reflect the true flow state after the change in dryness.

[0082] In order to incorporate the interfacial friction effect, accelerated pressure drop and vapor-liquid distribution disturbance unique to two-phase flow into the impedance calculation, it is necessary to construct a two-phase flow correction factor K1 based on the dryness evolution coefficient S1.

[0083] The two-phase flow correction factor K1 is constructed as follows: Where: K1: two-phase flow correction factor, a value greater than 1 indicates enhanced two-phase flow resistance; S1: dryness evolution coefficient; α: first-order dryness sensitivity coefficient, which needs to be determined through experimental calibration and is used to reflect the influence of vapor-liquid friction in the low dryness region; β: second-order dryness sensitivity coefficient, used to increase the nonlinear response capability of the model in the high dryness region.

[0084] The process of constructing the correction factor K1 includes:

[0085] Determine the influence trend of the dryness evolution coefficient S1 on the change of interfacial friction;

[0086] First-order and second-order terms are used to enhance the modeling capability of flow disturbances under high dryness conditions;

[0087] By calibrating α and β, the theoretical model is made consistent with the actual two-phase flow characteristics.

[0088] After obtaining the pressure difference parameter ΔPi, the normalized mass flow rate Mi1, and the two-phase flow correction factor K1, the equivalent impedance Zi of each branch is obtained by using the impedance calculation model based on these three parameters. The impedance calculation model is as follows: Where: Zi: equivalent impedance of the branch, in Pa·s / kg, used to quantify the magnitude of the branch flow resistance; ΔPi: branch pressure difference, representing the pressure drop experienced by the fluid through this branch; K1: two-phase flow correction factor, used to introduce the influence of dryness fraction on resistance; Mi1: normalized mass flow rate, representing the actual effective flow rate after dryness fraction correction. By calculating the equivalent impedance Zi, the flow resistance of the branch under different dryness fraction states can be quantified, providing a precise basis for subsequent valve regulation strategies.

[0089] S400: Based on the equivalent impedance Zi, determine the target opening Vi of the liquid supply control valve so that the branch with high impedance gets high liquid supply priority; and input the target opening Vi into the dryness evolution relationship model SX to calculate the dryness prediction value X1.

[0090] After calculating the equivalent impedance Zi of each branch, the branches need to be sorted according to their resistance magnitude. First, the equivalent impedance values ​​of each branch are collected and formed into an impedance set {Zi}. Then, the impedance sets are sorted in descending order to obtain the sorted impedance sequence.

[0091] During the sorting process, if the equivalent impedance of branch A is greater than that of branch B, then branch A is ranked before branch B. After sorting, the sorting result is defined as an impedance priority sequence. The impedance priority sequence is used to directly indicate the priority of each branch in liquid supply regulation. The larger the impedance, the more significant the flow resistance of that branch, therefore, the liquid supply should be increased first to reduce dryness fluctuations. Where, Zi: the equivalent impedance of the i-th branch, in Pa·s / kg.

[0092] After establishing the impedance priority sequence, the target valve opening Vi for each branch needs to be determined through the mapping relationship between impedance and valve opening. The valve opening mapping function adopts a monotonically increasing function form, such that the larger the equivalent impedance, the larger the valve opening. The mapping function is defined as follows: Where: Vi: Target valve opening of the i-th branch, in percentage; Vmin: Minimum valve opening, used to ensure basic liquid supply capacity, generally set to 5%–10% according to the structure; γ: Valve opening gain coefficient, obtained through experimental calibration, used to control the intensity of the influence of impedance on valve opening; Zi: Equivalent impedance of the i-th branch.

[0093] The method for constructing a mapping function includes the following steps:

[0094] Determine the minimum valve opening Vmin based on the actual flow characteristic curve of the valve;

[0095] The valve opening gain coefficient γ is determined based on the impedance variation range so that the valve opening variation range can cover the actual requirements.

[0096] The equivalent impedance Zi is input into the mapping function to calculate the target opening Vi.

[0097] This mapping relationship ensures that branches with high impedance can obtain a larger opening during liquid supply regulation, thereby reducing their dryness more quickly.

[0098] After calculating the target valve opening Vi, it is necessary to input Vi into the dryness fraction evolution model SX to update the instantaneous liquid supply function m(t) in the model. The relationship between the liquid supply rate m(t) and the valve opening Vi is defined by the liquid supply flow characteristic curve as follows: Where: m(t): the mass flow rate of the liquid supply at time t, in kg / s; Vi: the target valve opening; k1: the slope of the liquid supply curve, calibrated experimentally, used to characterize the flow rate increment corresponding to a unit opening; k2: the initial offset of the liquid supply, used to ensure that the valve still has basic liquid supply capacity at the minimum opening.

[0099] The fluid supply update process is as follows:

[0100] Input the calculated target opening degree Vi into the liquid supply calculation formula;

[0101] Update the liquid supply mass flow rate m(t) based on parameters k1 and k2;

[0102] The updated m(t) is written into the dryness evolution model SX, so that the model can change in real time with the liquid supply adjustment.

[0103] This process tightly couples valve regulation with dryness prediction to achieve real-time dynamic dryness prediction.

[0104] After updating the liquid supply mass flow rate m(t), m(t) and heat flux density q(t) are substituted together as input variables into the dryness fraction evolution model SX. The model expression is as follows: Where: X: dryness fraction variable; dX / dt: rate of change of dryness fraction; q(t): instantaneous heat flux density, in units of... A1: Equivalent heat transfer area parameter; m(t): Updated liquid supply mass flow rate; C1, C2: Dryness fraction evolution coefficients, obtained through parameter identification. By discretizing the above differential equation, the predicted dryness fraction value X1 can be obtained: ; where Δt is the calculation time step. The dryness prediction value X1 is used for subsequent dryness risk calculation and impedance model correction, and is an important parameter for two-phase liquid cooling supply control.

[0105] S500: Calculate the drying risk parameter D based on the predicted dryness value X1 and the temperature difference between the evaporation section wall and the wall surface. To accurately determine the degree of overheating in local areas of the evaporation section, multiple wall temperature sensors are installed along the flow direction on the evaporation section wall. Each wall temperature sensor collects the wall temperature Tw(i) at its location in real time. Simultaneously, an inlet temperature sensor is installed at the evaporation section inlet to collect the inlet temperature Tin. For each wall location, the wall temperature data is differentiated from the inlet temperature to obtain the wall temperature difference parameter ΔT1(i), which is expressed as: ΔT1(i) = Tw(i) - Tin; where: Tw(i): temperature at the i-th position on the wall (°C); Tin: evaporation section inlet temperature (°C); ΔT1(i): wall temperature difference parameter, used to characterize the local wall temperature rise. The larger the wall temperature difference parameter, the thinner the liquid film, the stronger the local vaporization, and the higher the drying risk.

[0106] To evaluate the impact of the predicted dryness fraction X1 on liquid film stability, the predicted dryness fraction X1 was input into the dryness-sensitive model. The dryness-sensitive model is constructed based on the influence of dryness on vapor-liquid interface fluctuations and liquid film thickness changes, and the model adopts an exponential response form. Where: X1: predicted dryness fraction, ranging from 0 to 1; λ: dryness sensitivity coefficient, used to adjust the degree of influence of dryness on the liquid film rupture trend; F1: dryness response factor, the larger the value, the more unstable the liquid film. The dryness sensitivity coefficient λ is obtained through experimental calibration based on different working fluids and flow channel structures, ensuring the model has physical consistency.

[0107] The wall heat load factor is used to describe the effect of wall temperature difference on the liquid film evaporation rate and drying tendency, and is determined by the correspondence between wall temperature difference parameters and wall heat absorption capacity.

[0108] The wall heat load factor H1 is constructed as follows: Where: ΔT1: the average or maximum value of the wall temperature difference parameter, used to characterize the temperature shift of the whole or the most dangerous location; μ: the temperature difference amplification factor, determined based on the thermal conductivity and surface morphology of the evaporation section material; H1: the wall heat load factor, representing the enhancement of the liquid film evaporation capacity due to the increase in wall temperature. The physical function of the temperature difference amplification factor μ is to convert the temperature difference into a heat load index, enabling this parameter to participate in subsequent risk synthesis.

[0109] To comprehensively reflect the combined influence of dryness and wall heating conditions on the drying trend, the dryness response factor F1 and the wall heat load factor H1 are linearly synthesized according to their weights to obtain the drying risk parameter D: Wherein: s1: Dryness influence weighting coefficient, determining the contribution of dryness changes to the formation of drying; s2: Wall heat load influence weighting coefficient, determining the proportion of thermal instability caused by temperature difference; D: Drying risk parameter, used to assess the combined risk of liquid film rupture and drying in the evaporation section. The weighting coefficients s1 and s2 are calibrated according to the flow channel structure, working fluid characteristics, and experimental conditions, so that the drying risk parameter D can accurately reflect the overall drying trend of the evaporation section. The larger the value of the drying risk parameter D, the closer the evaporation section is to the drying boundary, which can be used as the basis for judgment in the subsequent impedance correction step S600.

[0110] S600: If the drying risk parameter D exceeds the preset threshold, the branch impedance model SZ is corrected based on the dryness prediction value X1 and the equivalent impedance Zi to obtain a new branch impedance Zi', and the process returns to step S400; otherwise, proceed to step S700.

[0111] When the drying risk parameter D exceeds the preset threshold Dth, it indicates that the liquid film in the evaporation section is close to instability, requiring correction of the branch impedance model SZ. To determine the correction direction and magnitude, the predicted dryness value X1 is compared with the initial dryness estimate X0 to obtain the dryness deviation ΔX, expressed as follows: ΔX = X1 - X0; where: X1: predicted dryness value, used to characterize the gas-liquid ratio in the evaporation section at the next moment; X0: initial dryness estimate, characterizing the gas-liquid ratio at the initial state of the model; ΔX: dryness deviation, used to measure the trend of dryness change. The sign of the dryness deviation ΔX is used to indicate the correction direction of the impedance model SZ:

[0112] When ΔX is positive, it indicates that the dryness is increasing and the proportion of gas phase is increasing, so the flow resistance estimate in the impedance model should be improved.

[0113] When ΔX is negative, it indicates that the dryness fraction decreases and the liquid phase ratio increases, so the flow resistance estimate in the impedance model should be reduced.

[0114] The absolute value of the dryness deviation is used to determine the correction magnitude. The larger the deviation, the greater the correction force of the branch impedance model SZ.

[0115] After obtaining the dryness deviation ΔX and the equivalent impedance Zi, an impedance response factor K2 is constructed to reflect the coupling relationship between the dryness variation and the flow resistance variation. The impedance response factor K2 is constructed using the following functional form: Where: K2: impedance response factor, used to amplify or reduce the influence of dryness on impedance; ΔX: dryness deviation; a1: primary dryness sensitivity coefficient, calibrated based on experimental or simulation data; a2: secondary dryness sensitivity coefficient, used to enhance the model's nonlinear response capability to high dryness regions.

[0116] The steps for constructing the impedance response factor K2 include:

[0117] Analyze the effects of dryness variation on gas-liquid interface friction and accelerated pressure drop;

[0118] Select linear and quadratic terms according to the model requirements to enhance the sensitivity of high-dryness areas;

[0119] Parameters a1 and a2 are calibrated based on the structural dimensions of the evaporation section, the properties of the working fluid, and the flow characteristics.

[0120] The physical meaning of K2 is to reflect the magnitude of impedance increase or decrease caused by changes in dryness, so that the impedance model can present the true dynamic behavior of two-phase flow.

[0121] To correlate the change in dryness with the change in equivalent impedance, the impedance response factor K2 and the equivalent impedance Zi are input into the impedance correction function to construct the impedance correction amount ΔZi. The impedance correction amount is determined as follows: Where: ΔZi: impedance correction amount, used to update branch impedance; Zi: current branch equivalent impedance; K2−1: used to convert the response factor into the actual magnitude of its impact on impedance. The above model reflects the coupling relationship between impedance and interference degree.

[0122] The more drastic the change in dryness, the greater the impedance correction.

[0123] If the dryness increases and ΔX is positive, then the impedance correction ΔZi is positive, indicating that the impedance estimate needs to be improved.

[0124] If the dryness decreases and ΔX is negative, then the impedance correction ΔZi is negative, indicating that the impedance estimate needs to be reduced.

[0125] This coupling relationship can improve the response capability of the impedance model SZ to changes in the actual state of the two-phase flow.

[0126] After obtaining the impedance correction ΔZi, it is input into the branch impedance model SZ to update the impedance parameters in the model and generate the updated impedance Zi'. The calculation method is: Zi' = Zi + ΔZi; where: Zi': the updated branch impedance, used for subsequent liquid supply adjustment steps; Zi: the equivalent impedance of the branch before correction; ΔZi: the impedance correction value calculated based on the dryness deviation and impedance response factor.

[0127] The updated branch impedance Zi' can more accurately reflect the impact of changes in dryness in the evaporation section on the branch flow resistance. After obtaining Zi', the updated impedance is input into the liquid supply adjustment step S400 to recalculate the valve opening, so that the liquid supply distribution can adapt to the latest gas-liquid distribution changes.

[0128] If several risk parameters D do not exceed the threshold Dth, then there is no need to perform impedance correction, and the process directly proceeds to step S700 to determine the liquid supply balance.

[0129] S700: Based on the dryness prediction value X1 and the equivalent impedance Zi or Zi', the overall liquid supply distribution of the two-phase liquid cooling system is simulated to obtain the liquid supply balance parameter W.

[0130] In order to accurately evaluate the real-time balance of multi-branch liquid supply distribution in a two-phase liquid cooling environment, it is necessary to integrate the dryness prediction value X1 with the equivalent impedance Zi or updated impedance Zi' determined in the previous step to form the liquid supply simulation input set Sinput.

[0131] The dryness fraction prediction value X1 is used to characterize the current gas-liquid ratio in the evaporation section and reflects the sensitivity of each branch to changes in the liquid supply rate; the equivalent impedance Zi or the updated impedance Zi' is used to characterize the overall resistance to the flow within each branch. The purpose of integrating X1 with Zi or Zi' is to simultaneously reflect the thermodynamic and flow states of the branch, so that the simulation process can reflect the impact of changes in gas-liquid distribution in the actual two-phase flow on the liquid supply channel.

[0132] The liquid supply simulation input set Sinput is composed as follows: Sinput={X1,Zi or Zi′}; where: X1: dryness fraction prediction value; Zi: equivalent impedance, representing the branch flow resistance under the uncorrected condition; Zi': updated impedance, representing the branch resistance recalibrated under the influence of dryness fraction deviation; Sinput: liquid supply simulation input set, providing the input data basis for subsequent flow distribution solution.

[0133] After establishing the liquid supply simulation input set Sinput, the two-phase flow distribution solution model is called to calculate the achievable mass flow rate of multiple branches under given impedance conditions. The two-phase flow distribution solution model is constructed based on the resistance distribution law and mass conservation relationship in the series-parallel structure of multiple branches. The relationship between the achievable mass flow rate Mi(t) of the branch and the impedance Zi or Zi' in the model is defined by the following expression: Where: Mi(t): the achievable mass flow rate of the i-th branch at time t, in kg / s; Zi or Zi': the impedance parameter of the i-th branch, used to determine the proportion of flow allocated to this branch; Zk or Zk': the impedance of all branches participating in the flow allocation; Mtotal: the total supply mass flow rate, provided by the supply pump or pressure stabilizing device.

[0134] The purpose of the above allocation model is to achieve a natural flow allocation mechanism based on resistance distribution, where the higher the impedance, the smaller the flow share that the branch can obtain; and the lower the impedance, the larger the flow that the branch can obtain.

[0135] By calculating Mi(t) for each branch, a flow distribution result set Mset can be obtained to evaluate the degree of difference in liquid supply between branches: Mset={M1(t),M2(t),...,Mn(t)}. To quantitatively characterize the degree of liquid supply balance across multiple branches, a liquid supply balance evaluation function needs to be constructed based on the mass flow data of each branch. The liquid supply balance evaluation function uses the dispersion of mass flow rates across multiple branches as an indicator, and its definition is as follows: Where: W: Liquid supply balance parameter, the smaller the value, the closer the liquid supply distribution is to balance; Mi(t): Mass flow rate of the i-th branch at time t; Mavg: Average mass flow rate of all branches.

[0136] The physical meaning of the liquid supply balance parameter W is as follows:

[0137] When W is close to zero, it indicates that the liquid supply from multiple branches is nearly consistent, and the system temperature field has high stability.

[0138] When W is large, it indicates that the difference in liquid supply between branches is aggravated, which may lead to higher dryness in some evaporation sections and cause the risk of local drying.

[0139] After obtaining the liquid supply balance parameter W, it is used as the basis for stability determination in the next step S800 to determine whether it is necessary to re-enter the impedance update and liquid supply adjustment cycle.

[0140] S800: If the liquid supply balance parameter W does not reach the stable condition, then input W into the dryness evolution relationship model SX to obtain the updated dryness evolution coefficient S2, and return to step S300; if the stable condition is reached, then output the final valve opening of each branch and the steady-state dryness distribution of the system.

[0141] After obtaining the liquid supply balance parameter W, it needs to be compared with the preset stability condition Wth. The preset stability condition Wth is a threshold used to determine whether the two-phase liquid cooling liquid supply distribution has reached a stable state, and it is obtained through experiments or simulations based on the flow fluctuation characteristics of the two-phase flow.

[0142] The judgment rule is as follows: When W ≤ Wth, the deviation of the liquid supply flow rate of multiple branches is considered to be within an acceptable range, and the flow rate has reached a stable state; when W > Wth, it indicates that there are still significant differences in the liquid supply of multiple branches, and the dryness evolution process needs to be further corrected. Wherein, W: liquid supply balance parameter, used to characterize the degree of dispersion of the flow rate of multiple branches; Wth: liquid supply stability threshold, a reference value used to determine the degree of liquid supply balance. This judgment result directly determines whether it is necessary to continue the correction process of the dryness evolution relationship model SX.

[0143] When the liquid supply balance parameter W exceeds the preset stability condition Wth, a balance deviation E1 needs to be constructed based on the deviation between W and Wth to quantify the strength of the liquid supply imbalance. The balance deviation E1 is defined as follows: E1 = W − Wth; where: E1: balance deviation, a positive value indicates the degree of liquid supply imbalance; W: liquid supply balance parameter; Wth: stability condition threshold.

[0144] The magnitude of the equilibrium deviation E1 is used to determine the correction range of the liquid supply-related terms in the dryness-partition evolution model SX. The larger the deviation, the more obvious the dispersion of the branch flow, and the more the dryness-partition evolution model should be adjusted to strengthen the regulatory effect of liquid supply on dryness.

[0145] In order for the dryness evolution model SX to reflect the impact of unbalanced liquid supply on dryness changes, the balance deviation E1 needs to be input into the dryness evolution model SX to adjust the weight of the liquid supply mass flow rate term m(t) in the model.

[0146] The basic form of the dryness evolution relationship model SX is: The adjustment method involves introducing a liquid supply correction weight ω into the model, allowing the liquid supply term to be dynamically updated according to the degree of liquid supply imbalance. The corrected model expression is as follows: Where: ω: liquid supply correction weight, determined based on the balance deviation E1, is defined as follows: k4: Liquid supply correction weight gain coefficient, obtained through experimental calibration, to enable the model to have adjustment sensitivity under different equilibrium conditions; the remaining parameters are defined as follows: q(t): instantaneous heat flux density; A1: equivalent heat transfer area parameter; m(t): liquid supply mass flow rate; C1, C2: dryness evolution coefficients. By introducing ω, the dryness evolution relationship model SX can enhance or weaken the influence of liquid supply rate on dryness changes according to the degree of liquid supply imbalance.

[0147] After inputting the equilibrium deviation E1 into the dryness evolution model SX and adjusting the weights, the model output needs to be linearized to obtain a new dryness evolution coefficient S2.

[0148] The linearization method employs a first-order Taylor expansion, using the partial derivative of the dryness fraction change rate with respect to the heat flux density term as the update coefficient S2: Since the liquid supply term m(t) has been corrected by the weight ω, the value of S2 will change relative to the previous stage, reflecting the change in the sensitivity of dryness to heat flow disturbance under the condition of uneven liquid supply. S2: The updated dryness evolution coefficient, used for impedance calculation in the subsequent step S300; The slope of the response of dryness to heat flux density is used as a result of model linearization. After the calculation is completed, the updated dryness evolution coefficient S2 is returned to step S300 to recalculate the equivalent impedance of the branch, thereby realizing closed-loop control of dryness-impedance-liquid supply.

[0149] If the liquid supply balance parameter W meets the stability condition, there is no need to execute the above correction process. Instead, the final valve opening and the steady-state dryness distribution of the system are directly output, and the control loop ends.

[0150] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation, characterized in that: include: S100: Obtain the temperature sequence T between the inlet and outlet of the evaporation section of the two-phase liquid cooling system, and calculate the initial dryness estimate X0 of the evaporation section based on the rate of change of the temperature sequence T. S200: Based on the initial dryness estimate X0 and the structural parameters of the evaporation section, construct the dryness evolution relationship model SX of the two-phase region, and obtain the dryness evolution coefficient S1 based on the model; S300: Based on the dryness evolution coefficient S1, the inlet pressure and mass flow rate of each branch in the two-phase liquid cooling system are jointly analyzed to obtain the equivalent impedance Zi of each branch. S400: Based on the equivalent impedance Zi, determine the target opening Vi of the liquid supply control valve so that the branch with high impedance gets high liquid supply priority. The target aperture Vi is then input into the dryness evolution relationship model SX to calculate the predicted dryness value X1. S5 00: Calculate the drying risk parameter D based on the predicted dryness value X1 and the temperature difference between the evaporation section wall; S600: If the drying risk parameter D exceeds the preset threshold, the branch impedance model SZ is corrected based on the dryness prediction value X1 and the equivalent impedance Zi to obtain a new branch impedance Zi', and the process returns to step S400. Otherwise, proceed to step S700; S700: Based on the dryness prediction value X1 and the equivalent impedance Zi or Zi', the overall liquid supply distribution of the two-phase liquid cooling system is simulated to obtain the liquid supply balance parameter W. S800: If the liquid supply balance parameter W does not reach the stable condition, then input W into the dryness evolution relationship model SX to obtain the updated dryness evolution coefficient S2, and return to step S300; if the stable condition is reached, then output the final valve opening of each branch and the steady-state dryness distribution of the system.

2. The liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to claim 1, characterized in that: S100 further includes: A first temperature sensor and a second temperature sensor are installed at the inlet and outlet of the evaporation section, respectively, and multiple sets of temperature data are collected synchronously at a fixed sampling period to form a temperature sequence T ordered by time. The instantaneous temperature gradient between any adjacent sampling points is calculated based on the temperature sequence T, and a moving average is performed through a set time window to obtain the stable temperature change rate parameter R1 of the evaporation section. Input the steady temperature change rate parameter R1 into the preset dryness-sensitive response function model, and modify the model according to the structural parameters of the evaporation section to obtain the intermediate quantity X01 for dryness estimation. A linear mapping relationship is established between the intermediate dryness estimate X01 and the reference temperature difference at the inlet of the evaporation section to obtain the initial dryness estimate X0 of the evaporation section.

3. The liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to claim 1, characterized in that: S200 further includes: Obtain the structural parameters of the evaporation section, including the length of the evaporation section, the cross-sectional area of ​​the flow channel, the wetted perimeter of the flow channel, and the surface roughness of the evaporation section wall, and calculate the equivalent heat transfer area parameter A1 of the evaporation section based on the structural parameters; Using the initial dryness estimate X0 as the input variable, and combining it with the equivalent heat transfer area parameter A1, a dryness evolution relationship model SX for the two-phase region is constructed. The dryness evolution relationship model SX establishes a functional relationship between the dryness change rate dX / dt and the instantaneous liquid supply and heat flux density of the evaporation section in a time-continuous differential form. The parameters of the dryness evolution model SX are identified. The error between the rate of change of the inlet temperature of the evaporation section and the dryness predicted by the model is calculated. The dryness evolution parameter set P1 is solved by the least squares method. Input the set of dryness evolution parameters P1 into the dryness evolution relationship model SX, and linearize the model output to obtain the dryness evolution coefficient S1.

4. The liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to claim 1, characterized in that: The S300 further includes: Inlet and outlet pressure data are collected at the inlet and outlet locations of each branch. The pressure difference between the inlet and outlet pressures is calculated to obtain the pressure difference parameter ΔPi, which reflects the resistance distribution characteristics of the two-phase flow. Based on the dryness evolution coefficient S1, the mass flow rate measured in real time for each branch is normalized. After obtaining the pressure difference parameter ΔPi and the normalized mass flow rate Mi1, a two-phase flow correction factor K1 is constructed based on the dryness evolution coefficient S1. The pressure difference parameter ΔPi, the normalized mass flow rate Mi1, and the two-phase flow correction factor K1 are input into the impedance calculation model to calculate the equivalent impedance Zi corresponding to each branch.

5. The liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to claim 1, characterized in that: The S400 further includes: Sort the equivalent impedance Zi of each branch according to its numerical value, and generate an impedance priority sequence P1 according to the impedance from largest to smallest. Based on the impedance priority sequence P1, the equivalent impedance Zi of each branch is input into the preset valve opening mapping function to calculate the target opening Vi of the liquid supply control valve. The target opening degree Vi is input into the dryness evolution relationship model SX, and the liquid supply function m(t) in the model is updated by using the influence of valve opening degree on instantaneous liquid supply. Based on the updated liquid supply function m(t) and the parameter set in the dryness evolution relationship model SX, the predicted dryness value X1 is calculated.

6. The liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to claim 1, characterized in that: The S500 further includes: Wall temperature data is acquired by multiple wall temperature sensors installed at different locations on the evaporation section wall, and the wall temperature data is compared with the evaporation section inlet temperature to form a wall temperature difference parameter that can reflect the degree of local overheating in the evaporation section. Input the predicted dryness value X1 into the preset dryness-sensitive model, and the model outputs a dryness response factor that can characterize the strength of the influence of dryness changes on liquid film stability. Based on the wall temperature difference parameters, the wall heat load factor is generated by utilizing the correspondence between the temperature difference and the wall heating condition. The dryness response factor and the wall heat load factor are combined according to the set weights to obtain the dryness risk parameter D, which characterizes the severity of dryness in the evaporation section.

7. The liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to claim 1, characterized in that: The S600 further includes: The predicted dryness value X1 is compared with the initial estimated dryness value X0 to obtain the dryness deviation amount, which is used to characterize the degree of dryness deviation in the evaporation section. The correction direction and correction magnitude of the branch impedance model SZ are determined by the dryness deviation amount. Based on the real-time changing trends of dryness deviation and equivalent impedance Zi, an impedance response factor is constructed. The impedance response factor is correlated with the equivalent impedance Zi, and an impedance correction amount is generated based on the coupling relationship between impedance and dryness. The impedance correction is input into the branch impedance model SZ, and the parameters of the branch impedance model SZ are updated to obtain the updated impedance Zi' for subsequent liquid supply regulation.

8. The liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to claim 1, characterized in that: The S700 further includes: The dryness prediction value X1 is integrated with the equivalent impedance Zi or updated impedance Zi' determined by the previous steps to establish a set of liquid supply simulation inputs to describe the liquid supply status of each branch. Based on the aforementioned liquid supply simulation input set, a preset two-phase flow distribution solution model is invoked. The model calculates the achievable mass flow rate of each branch under a given impedance condition to obtain the flow distribution result used to characterize the flow distribution trend of multiple branches. Based on the flow distribution results, the mass flow data of each branch is extracted, and a liquid supply balance evaluation function is constructed according to the flow dispersion of multiple branches to obtain the liquid supply balance parameter W, which reflects the consistency of liquid supply in multiple branches.

9. The liquid supply control method for a two-phase liquid cooling system based on dryness sensing and branch impedance regulation according to claim 1, characterized in that: The S800 further includes: The liquid supply balance parameter W is compared with the preset stability condition. When the liquid supply balance parameter W is higher than the stability condition, it is determined that the liquid supply distribution of the two-phase liquid cooling has not yet reached the flow balance state, and the dryness evolution process needs to be corrected. Based on the degree of deviation between the liquid supply balance parameter W and the stability condition, a balance deviation quantity is generated to characterize the strength of the liquid supply imbalance, so that the balance deviation quantity can reflect the influence of the flow dispersion of each branch on the dryness change trend. Input the equilibrium deviation into the dryness evolution relationship model SX, and adjust the weights of the liquid supply related terms in the dryness evolution relationship model SX. Based on the dryness evolution relationship model SX after the deviation input, the model output is linearized to obtain a new dryness evolution coefficient S2.

Images