Wellhead valve and wellhead valve parameter optimization method and system based on intelligent algorithm

Abstract

The invention provides a wellhead valve and a wellhead valve parameter optimization method and system based on an intelligent algorithm, and relates to the field of wellhead valves and valve parameter optimization. The pressure relief hole is additionally formed in the back face of the valve plate, so that the overflowing area of the well mouth valve under each opening degree is reduced, the throttling effect is improved, the temperature of a flow field in the well mouth valve is located at a higher level in the whole CO2 discharging process, and the risk that the valve is frozen and blocked is reduced. According to the parameter optimization method, scientific sampling is carried out through Latin hypercube, structure parameters corresponding to sample points are determined, influence factors of the parameters on an optimization target are identified by constructing a radial basis function proxy model and carrying out sensitivity analysis, and finally optimization is carried out through a genetic algorithm, so that a final optimization result is obtained. And a valve structure corresponding to the final optimized structure is established, so that the valve is better prevented from being frozen and blocked. The method is suitable for pressure release valve parameter optimization in the dense-phase CO2 release process.

Description

Technical Field

[0001] The present invention relates to the field of wellhead valves and valve parameter optimization, and particularly to a wellhead valve optimization method based on intelligent algorithms. Background Art

[0002] For the transportation of large-scale carbon dioxide, high-temperature and normal-pressure liquid-phase carbon dioxide pipeline transportation, also known as "dense-phase" transportation, is adopted. However, compared with the pipeline transportation of hydrocarbons such as natural gas, the industry experience in transporting carbon dioxide through pipelines is much less, and the relevant risks should be fully understood and effectively managed.

[0003] There is a need to control the single-phase state during the transportation of carbon dioxide. Near the critical region, small differences in pressure and temperature will cause sharp fluctuations in the state characteristics of carbon dioxide. Therefore, the pressure during transportation needs to be maintained above the critical pressure to avoid two-phase transportation. Currently, the operating pressure during transportation is 8.5 MPa - 12 MPa, which is in the "smooth" region of the carbon dioxide phase diagram, that is, the state of carbon dioxide in this region is stable and there will be no mutations. However, during the pressure relief process, the characteristics of carbon dioxide will inevitably change sharply. Due to accidental failures (third-party construction damage or corrosion) or planned maintenance, pipelines often need to be depressurized and discharged. During the depressurization process, the escaping gas will cool the pipeline. If the temperature drops too much and is lower than the ductile-brittle transition temperature of the pipeline material, the pipeline material will become embrittled, resulting in brittle fracture and serious pipeline damage. The formation of dry ice during the carbon dioxide pressure relief process also poses a certain risk of freezing and blocking to the pipeline or valve. Therefore, the pressure relief of the carbon dioxide pipeline needs to be controlled to ensure that the pipeline temperature is not lower than the design temperature.

[0004] Therefore, to prevent the risk of freezing and blocking of the wellhead valve, it is necessary to optimize the structure of the wellhead valve. Summary of the Invention

[0005] The present invention provides a wellhead valve. By adding a pressure relief hole on the back of the valve plate, the flow area of the wellhead valve at each opening is reduced, and the throttling effect is increased, so that the temperature of the internal flow field of the wellhead valve is at a higher level during the entire CO 2 discharge process, thereby reducing the risk of valve freezing and blocking. At the same time, a wellhead valve parameter optimization method based on intelligent algorithms is also proposed to optimize the size of the pressure relief hole and better prevent the risk of valve freezing and blocking.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention provides a wellhead valve, which includes a valve plate, and a pressure relief hole is provided on the back of the valve plate.

[0008] Furthermore, there is also a preferred embodiment, where the diameter a of the pressure relief hole ranges from 55 mm to 65 mm;

[0009] The position b of the pressure relief hole ranges from 40 mm to 60 mm;

[0010] The width c of the pressure relief hole ranges from 14 mm to 22 mm.

[0011] The present invention also provides an optimization method for wellhead valve parameters based on an intelligent algorithm. The optimization method is used to optimize the above-mentioned pressure relief hole parameters, and the method is as follows:

[0012] Step S1: According to the diameter a, position b, and width c of the pressure relief hole, sampling is performed through Latin hypercube to obtain each sample point;

[0013] Step S2: Construct a radial basis function surrogate model according to each sample point;

[0014] Step S3: Adopt a global sensitivity analysis method based on Monte Carlo to sample the diameter a, position b, and width c of the pressure relief hole, obtain different input combinations, calculate the output results of the radial basis function surrogate model under different input combinations, and obtain the contribution degree of each input parameter to the output result;

[0015] Step S4: According to the contribution degree of each input parameter to the output result, use a genetic algorithm to optimize the diameter a, position b, and width c of the pressure relief hole to obtain the optimal parameters of the diameter a, position b, and width c of the pressure relief hole.

[0016] Furthermore, each of the above sample points corresponds to the diameter a, position b, and width c of the pressure relief hole.

[0017] Furthermore, the output of the above global sensitivity analysis method based on Monte Carlo is the temperature of the wellhead valve.

[0018] Furthermore, by statistically analyzing the variance of the output results, the total variance of the temperature change is decomposed into the contributions of different input parameters to obtain the contribution degree of each input parameter to the output result.

[0019] Furthermore, there is also a preferred embodiment, where the optimal diameter a of the pressure relief hole is 55.10 mm, the position b is 55.72 mm, and the width c is 18.89 mm.

[0020] The optimization method for wellhead valve parameters based on an intelligent algorithm described in the present invention can be fully implemented by computer software. Therefore, correspondingly, the present invention also provides an optimization system for wellhead valve parameters based on an intelligent algorithm. The system includes a storage device, and the storage device is used to execute the above-mentioned optimization method and steps for wellhead valve parameters based on an intelligent algorithm.

[0021] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, it executes the method for optimizing wellhead valve parameters based on an intelligent algorithm described in any one of the above.

[0022] The present invention also provides a computer device, which includes a memory and a processor. A computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes the method for optimizing wellhead valve parameters based on an intelligent algorithm described in any one of the above.

[0023] The beneficial effects of the present invention are as follows:

[0024] 1. The present invention proposes a wellhead valve. By adding a pressure relief hole on the back of the valve plate, the flow area of the wellhead valve at each opening is reduced, the throttling effect is increased, so that the temperature of the internal flow field of the wellhead valve is at a higher level during the entire CO 2 venting process, thereby reducing the risk of valve freezing.

[0025] 2. The present invention proposes a method for optimizing wellhead valve parameters based on an intelligent algorithm. Through Latin hypercube sampling, the structural parameters corresponding to each sample point and the corresponding results under each working condition are determined. By constructing a radial basis function surrogate model and performing sensitivity analysis, the influencing factors of each parameter on the optimization target are identified, and finally, through genetic algorithm optimization, the final optimization result is obtained, and the valve structure corresponding to the final optimized structure is established, which better prevents the risk of valve freezing.

[0026] The present invention is applicable to the parameter optimization of pressure relief valves during the dense-phase CO 2 venting process. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0028] Figure 1 It is a structural diagram of the wellhead valve and the pressure relief hole according to the embodiment of the present invention;

[0029] Figure 2 It is a 1 / 23D flow domain diagram of the wellhead gate valve according to the embodiment of the present invention;

[0030] Figure 3 It is a schematic diagram of the mesh division of the wellhead gate valve according to the embodiment of the present invention;

[0031] Figure 4 is the curve graph of the corresponding relationship between the saturation temperature and the saturation pressure of CO described in the embodiment of the present invention 2 ;

[0032] Figure 5 is the Latin hypercube sampling distribution graph described in the embodiment of the present invention, where (a) is the width sampling distribution graph, (b) is the diameter sampling distribution graph, and (c) is the diameter sampling distribution graph

[0033] Figure 6 is the pressure nephogram of the wellhead gate valve pipeline at 25% opening described in the embodiment of the present invention

[0034] Figure 7 is the velocity nephogram of the wellhead gate valve pipeline at 25% opening described in the embodiment of the present invention

[0035] Figure 8 is the volume fraction nephogram of liquid CO in the wellhead gate valve pipeline at 25% opening described in the embodiment of the present invention 2 ;

[0036] Figure 9 is the volume fraction nephogram of gaseous CO in the wellhead gate valve pipeline at 25% opening described in the embodiment of the present invention 2 ;

[0037] Figure 10 is the temperature nephogram of the wellhead gate valve pipeline at 25% opening described in the embodiment of the present invention

[0038] Figure 11 is the area where the valve described in the embodiment of the present invention may have the risk of freezing and blocking

[0039] Figure 12 is the flow chart of the genetic algorithm described in the embodiment of the present invention Specific Embodiments

[0040] The following further describes in detail the specific embodiments of the present invention with reference to the accompanying drawings. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several changes and improvements can still be made, and these all belong to the protection scope of the present invention

[0041] Embodiment 1. Refer to Figures 1 to 4 To illustrate this embodiment, this embodiment provides a wellhead valve. By adding a pressure reducing hole on the back of the valve plate, the flow area of the wellhead valve at each opening is reduced, and the throttling effect is increased, so that the temperature of the internal flow field of the wellhead valve is at a higher level during the entire CO 2 venting process to reduce the risk of valve freezing and blocking

[0042] The specific structure of the pressure relief hole is as follows: Figure 1 As shown in Table 1, a is the diameter of the pressure reducing hole, the original size is 65mm, and the value range is 55mm-65mm; b is the relative position of the pressure reducing hole relative to the bottom of the valve plate, the original parameter is 50mm, and the value range is 40-60mm; c is the width of the pressure reducing hole, the original size is 18mm, and the value range is 14-18mm. The specific situation is shown in Table 1.

[0043] Table 1

[0044] Design parameters Original dimensions / mm Lower boundary / mm Upper boundary / mm Diameter a of the pressure relief hole 65 55 65 Position b of the pressure relief hole 50 40 60 Width c of the pressure relief hole 18 14 22

[0045] Temperature is an important parameter to measure the risk of wellhead valve freezing. In order to ensure the stability of the data and the convergence of the optimization, the area of ​​the valve symmetry surface where the temperature is lower than the triple point is selected as the optimization target S, that is,

[0046]

[0047] In practical application, this embodiment constructs a CO 2 A method for analyzing the freezing risk of wellhead valves during decompression, and optimizing the valve structure through the analysis method;

[0048] The method is:

[0049] First, a 1 / 23-D wellhead gate valve model was built, and watershed extraction, grid division, and solver settings were performed. Then, the CO 2 Physical properties were set and model verification was performed to obtain CO 2 The freezing risk numerical simulation model is used. Finally, the CO 2 The freezing risk numerical simulation model is used for CFD simulation calculation, and the freezing risk of the valve is analyzed according to the calculation results, so as to optimize the valve structure, that is, to add a pressure relief hole on the back of the valve plate as mentioned above.

[0050] Furthermore, the above-mentioned 1 / 23-D wellhead gate valve model is built, and the watershed extraction, meshing and solver settings are specifically as follows:

[0051] Wellhead valve is a kind of valve used to release CO 2 The plate gate valve mainly includes valve stem, valve cover, valve plate, valve seat, valve body and other structures. During the CO2 release process, by turning the hand wheel of the gate valve, the valve stem and valve plate are gradually opened under the push of the lead screw. The whole valve opening process lasts about 60 seconds, and then the valve reaches the maximum opening and performs a stable release process.

[0052] Based on the above model of the wellhead gate valve, watershed extraction was carried out in the model processing software. The structure of the wellhead gate valve is a plane-symmetric structure. Therefore, for the simulations at different opening degrees of the wellhead gate valve, 1 / 23-D models were used for simulation calculations. Considering the actual working environment of the valve, its calculation watershed was extracted and simplified as shown in Figure 2. As Figure 2 shown, the most important throttling area of the wellhead gate valve is near the gate plate. During the opening process of the valve, the throttling area of the gate plate region changes from small to large. At this time, liquid CO 2 flows through the valve under the drive of high pressure, and the temperature, pressure, and flow velocity change violently, resulting in a phase change. Therefore, the implementation analyzes the possible freezing and blocking problems during the valve opening and working processes by calculating the detailed changes in the valve flow field at different opening degrees.

[0053] Furthermore, to ensure the simulation accuracy and convergence, it is necessary to add watersheds with diameters 5 times and 10 times the valve inlet diameter at the inlet and outlet of the valve, so as to fully develop the flow fields before and after the valve. The inlet and outlet boundary conditions of the watershed are specified according to the specific working conditions.

[0054] After the calculation watershed extraction was completed, the computational grid of this embodiment was divided according to the structural characteristics of the wellhead gate valve. Although the simplified three-dimensional cavity model structure of the wellhead gate valve is simple, the internal flow characteristics, phase change heat transfer and mass transfer characteristics of the valve cavity are quite complex. Therefore, a hybrid grid division method was adopted to generate high-quality grids.

[0055] Furthermore, considering the complex flow, heat transfer and mass transfer phenomena inside the wellhead gate valve, it is necessary to process different regions when dividing the grid. In the region near the pressure relief slit of the wellhead gate valve, the flow situation is relatively complex, and local grid refinement is required. While in the region far from the pressure relief slit (which can be considered as most of the region inside the wellhead gate valve cavity), the fluid flow is relatively gentle, so relatively sparse grids can be divided in this region. Dividing the grid in this way by region can not only ensure the accuracy of the simulated results of the internal flow field of the pipe, but also appropriately reduce the number of grids, save computing resources and improve the computing speed of the simulation.

[0056] Furthermore, for CFD analysis, grid density is one of the key factors to ensure the accuracy and convergence of simulation calculations. To obtain the optimal computational grid model, this embodiment carried out simulation calculations on the wellhead gate valve using grids with different density levels, and explored the influence of grid density on simulation accuracy by comparing the temperatures at the valve plate outlet after different grid simulations. When the number of grids is greater than 2,673,022, the simulation results are independent of the grid density, that is, the calculation results converge to the computational grid. Considering the computational efficiency and cost comprehensively, this embodiment selected the grid generation scheme with 2,673,022 grids for subsequent simulation calculations. The final grid scheme is as Figure 3 shown.

[0057] After the mesh generation is completed, the solver parameters also need to be set in the CFD simulation, including: solver parameters (Pressure-based, Transient Absolute velocity formulation 3-D symmetric); working medium (gas-liquid two-phase CO 2 ); solution method (Coupled); spatial discretization format (second-order upwind); convergence criterion (all convergence residuals are less than 1e-4); turbulence model (Realizable k-ε); time step (3e-6 s).

[0058] In addition to the basic solver parameter settings, the two-phase flow parameters of CO 2 also need to be set before the calculation, including: liquid-phase CO2 parameters (density, viscosity, thermal conductivity); gas-phase CO 2 parameters (density, viscosity, thermal conductivity, enthalpy at standard state, reference temperature).

[0059] The above-mentioned 1 / 23-D wellhead gate valve model after setting is subjected to CO 2 physical property setting and model verification to obtain the specific numerical simulation model of the CO 2 freezing blockage risk;

[0060] The working pressure of the dense-phase carbon dioxide pipeline for normal-temperature transportation is maintained at about several tens of megapascals. After being discharged through the wellhead gate valve, the pressure drops suddenly, and the corresponding saturation temperature will decrease rapidly. Therefore, the dense-phase carbon dioxide will undergo a phase transition from the liquid phase to the gas phase (similar to the flashing phenomenon). The physical property parameters of the dense-phase carbon dioxide will fluctuate within a large range during the phase transition process. The correct carbon dioxide physical property parameters can ensure the accuracy of the simulation results.

[0061] The phase diagram and pressure-temperature relationship of carbon dioxide are as Figure 4 shown. The triple critical point is obtained by Refprop, T lim＝ 30.92 °C, P lim = 7.37 MPa. The inlet pressure of the wellhead gate valve is 30 MPa, the temperature is 300 K, and the outlet back pressure is 0.1 MPa. According to the inlet and outlet pressures of the wellhead gate valve, the carbon dioxide physical properties are divided into three sections: 0.1 - 0.6 MPa, 0.6 - 7.37 MPa, 7.37 - 30 MPa.

[0062] Among them, in the 7.37 - 30 MPa section, that is, the inlet pressure of the wellhead gate valve drops from 30 MPa to the critical pressure of carbon dioxide 7.37 MPa. Within this pressure drop range, the carbon dioxide does not undergo a phase transition, and the saturation temperature at different pressures is set to a constant value of 35 °C (the temperature in the calculation domain at the initial moment of the simulation is 25 °C).

[0063] In the range of 0.6 to 7.37 MPa, during the process of decreasing the carbon dioxide pressure, the saturation temperature corresponding to the pressure drops sharply, so the phase change process of carbon dioxide occurs in this section. The physical properties of saturated carbon dioxide at the pressure in this section are obtained by using Refprop software, and the results are shown in Table 2 below.

[0064] Table 2

[0065]

[0066]

[0067]

[0068] The discrete point data shown in Table 2 above are polynomially fitted to obtain the relationship between the saturation temperature and the corresponding saturation pressure.

[0069] T = 0.241P 3 - 4.2P 2 + 31.46P - 68.15

[0070] In the formula, T is the saturation temperature of carbon dioxide in °C; P is the saturation pressure of carbon dioxide in MPa. The correlation coefficient R of this fitting formula 2 = 0.9991, and the fitting accuracy is relatively high.

[0071] In the pressure range of 0.1 to 0.6 MPa, the fitting formula for the carbon dioxide saturation temperature in the range of 0.1 MPa to 0.6 MPa is no longer used for the range of 0.6 to 7.37 MPa. It is refitted according to the triple point and 0.1 MPa point data, and the refitted relationship formula is reloaded into the FLUENT evaporation and condensation model.

[0072] The fitting formula is as follows:

[0073] T = - 87.42P 2 + 124.96P + 176.53

[0074] In the formula, T is the saturation temperature of carbon dioxide in °C; P is the saturation pressure of carbon dioxide in MPa. The correlation coefficient R of this fitting formula 2 = 0.9995.

[0075] Embodiment 2: This embodiment provides a method for optimizing the wellhead valve parameters based on an intelligent algorithm. The optimization method is used to optimize the decompression hole parameters described in Embodiment 1 above to better prevent the risk of valve freezing. The optimization method includes the following steps:

[0076] Step S1: According to the diameter a, position b, and width c of the decompression hole, sampling is carried out through Latin hypercube to obtain each sample point;

[0077] Step S2: Construct a radial basis function surrogate model according to each sample point;

[0078] Step S3: Adopt a Monte Carlo-based global sensitivity analysis method to sample the diameter a, position b, and width c of the pressure relief hole, obtain different input combinations, calculate the output results of the radial basis function surrogate model under different input combinations, and obtain the contribution degree of each input parameter to the output results;

[0079] Step S4: Optimize the diameter a, position b, and width c of the pressure relief hole by using the genetic algorithm according to the contribution degree of each input parameter to the output result, and obtain the optimal parameters of the diameter a, position b, and width c of the pressure relief hole.

[0080] Embodiment 3. Refer to Figure 5 Describe this embodiment. This embodiment specifically describes a wellhead valve optimization method based on an intelligent algorithm described in the above Embodiment 2;

[0081] Step S1: According to the diameter a, position b, and width c of the pressure relief hole, perform sampling through Latin hypercube to obtain each sample point;

[0082] Specifically:

[0083] Select the Latin hypercube design method to sample the diameter a, position b, and width c of the pressure relief hole. The Latin hypercube design method can reasonably control the distribution of training samples in the design space, prevent repeated sampling in the local space, and ensure the space filling and uniformity of the training sample distribution to the greatest extent.

[0084] Among them, the principle of the Latin hypercube design method is: divide the design space into multiple small sub-intervals according to the design dimension, then determine several sampling sub-intervals through random permutation, and finally randomly arrange sample points in each sampling sub-interval.

[0085] (1) Suppose N sample points are generated, and the design space is normalized into a unit hypercube Ω = {X ∈ R N | 0 ≤ x i ≤ 1}.

[0086] (2) Divide the i-th dimension of the unit hypercube into n sub-intervals equally (0, 1 / N), (1 / N, 2 / N),..., (1 - 1 / N, 1). Denote the j-th sub-interval of the i-th dimension as

[0087] (3) Randomly take d different full permutations of (1, 2,... N) and form a random permutation matrix (ω ij ) d×N .

[0088] (4) Generate a sampling sub - interval matrix:

[0089]

[0090] (5) Randomly generate a point in each sub - interval of each sampling sub - interval matrix to obtain a sample matrix (x( 1 ), x( 2 ), …, x( N ))

[0091] According to the above - determined optimization structure and corresponding optimization parameters, 12 groups of sample points are sampled through Latin hypercube sampling. The sampling results are shown in Table 3 below, and the sample distribution is as Figure 5 shown.

[0092] Table 3

[0093]

[0094]

[0095] By establishing the above - mentioned sample model, simulation calculations are carried out respectively at two typical opening degrees, node data is exported, S corresponding to each working condition is calculated, and optimization data is obtained as the data basis for constructing the surrogate model.

[0096] Step S2: Construct a radial basis function surrogate model according to each sample point;

[0097] Specifically:

[0098] According to the above - extracted sample points, construct a radial basis function surrogate model and verify its accuracy.

[0099] Among them, the radial basis function surrogate model constructed in this embodiment is a radially symmetric function. It is an interpolation method with the advantages of simple form, strong adaptability, and high accuracy.

[0100] Let the function y = f(x) be an n - dimensional real - valued function. Using the experimental design method, m sample points are selected. The set composed of these sample points is expressed as: X = {x 1 , x 2 , …, x m} T , and the response value set y = {f(x 1 ), f(x 2 ), …, f(x m )} T , then the form of using the radial basis function to fit the function y is as follows:

[0101]

[0102] In the formula, λ i represents the undetermined coefficient before the i-th basis function, and ||x - x i || represents the Euclidean norm between the prediction point x and the sample point x i .

[0103] For the n-dimensional design space φ represents the form of the basis function of the radial basis. The commonly used basis function forms are shown in Table 4. Let r = ||x - x i ||.

[0104] Table 4

[0105]

[0106] Substituting m sample points and response values into the above formula, we get:

[0107]

[0108] The above equation has m equations and n unknowns. For the convenience of expression, the above formula can be written in matrix form:

[0109] Aλ = y

[0110] where λ = {λ 1 , λ 2 , …, λ m} T , y = {f(x 1 ), f(x 2 ), …, f(x m )} T . If a set of sample points and response values are given, the undetermined coefficient λ can be obtained according to the least squares method.

[0111] It can be seen that the radial basis function surrogate model describes the complex implicit functional relationship between the structural response and the structural parameters through the linear combination of the basis functions. As the number of parameters to be corrected increases, the number of sample points required to solve the undetermined coefficients is linearly related to the number of parameters to be corrected. Therefore, the radial basis function surrogate model has the advantage of saving computational cost when fitting unknown problems.

[0112] The accuracy of the radial basis function surrogate model is verified as follows:

[0113] The radial basis function surrogate model calculates the predicted value of the structural response by computing the input variables (usually using additional sample points), and then compares it with the true structural output response corresponding to the sample points. The error between them is used as the criterion for evaluating the fitting accuracy of the radial basis function surrogate model. Currently, the main criteria for testing the accuracy of the surrogate model are: coefficient of determination R 2 test, root mean square error (RMSE), maximum absolute error (MAE), normal distribution test of residuals, mean value of residuals, etc.

[0114] (1) Coefficient of determination R 2 test

[0115] Coefficient of determination R 2 test is a criterion for evaluating the overall fitting performance of the radial basis function surrogate model in the design space. Its principle is to evaluate by assessing the error between the predicted value and the true value of the radial basis function surrogate model at the test points. The value range of R 2 is from 0 to 1. The closer its value is to 1, the higher the fitting accuracy of the radial basis function surrogate model. Let represent the calculated value of the surrogate model at the i-th test sample point, and y i represent the true value at the i-th test sample point. The expression of R 2 is as follows:

[0116]

[0117] In the formula, n t represents the number of test sample points, is the average value of the true values of all test sample points.

[0118] (2) Root Mean Square Error (RMSE)

[0119] The root mean square error is the same as the coefficient of determination R 2 and is also used to evaluate the fitting performance of the radial basis function surrogate model in the entire design space. Its value is always greater than 0. The closer it is to 0, the better the fitting performance of the radial basis function surrogate model. Its expression is:

[0120]

[0121] However, for different test problems, the magnitude change of RMSE is quite different. In order to visually compare and analyze the RMSE values of multiple different test functions, a method of normalizing the RMSE results is adopted, that is, the normalized root mean square error (NRMSE):

[0122]

[0123] According to the above formula and combined with the sample data, after constructing the radial basis function surrogate model, through R 2 , RMSE, and NRMSE to verify the accuracy of the surrogate model. The specific values of each index are: R 2 is 0.9857, RMSE is 0.0012, and NRMSE is 0.0392. The results show that the R 2 , RMSE, and NRMSE of the radial basis function surrogate model are all optimal. Therefore, the constructed radial basis function surrogate model has higher accuracy.

[0124] Step S3: Adopt the Monte Carlo-based global sensitivity analysis method to sample the diameter a, position b, and width c of the pressure relief hole to obtain different input combinations, calculate the output results of the radial basis function surrogate model under different input combinations, and obtain the contribution degree of each input parameter to the output result;

[0125] In the Monte Carlo-based global sensitivity analysis, it is first necessary to understand the basic principle of this method and how it is applied to the optimization problem of the wellhead valve temperature. The Monte Carlo method is a numerical calculation method based on random sampling, which simulates and analyzes the behavior of complex systems through a large number of random samplings. In global sensitivity analysis, the Monte Carlo method samples a large number of input parameters (such as position, width, diameter), calculates the output results of the surrogate model under different input combinations, and thus evaluates the contribution degree of each input parameter to the output result. Specifically for the wellhead valve temperature target, the surrogate model, as an approximate expression of the system, takes the temperature of the wellhead valve as the output, and simulates the changes of the input parameters (position, width, diameter) to calculate the temperature response in different situations. In this process, the Monte Carlo method can not only handle the single effect of parameters, but also capture the possible complex interaction effects between input parameters, thus providing comprehensive sensitivity information for the optimization design.

[0126] Through Monte Carlo-based global sensitivity analysis, we can quantify the influence degrees of three parameters, namely position, width, and diameter, on the wellhead valve temperature. During the implementation process, first, random sampling is performed on the input space according to the surrogate model to generate a large number of input combinations, and the corresponding temperature outputs for each input combination are calculated. Then, by statistically analyzing the variances of these output results, the total variance of the temperature change is decomposed into the contributions of different input parameters. Usually, the results of the sensitivity analysis are presented in the form of the variance contribution ratios of each input parameter, thus revealing which parameters play a dominant role in the temperature change. For example, if the analysis results show that the position contributes the most to the temperature change, it indicates that the position is the most critical factor affecting the wellhead valve temperature, and the adjustment of the position should be given priority in the optimization process. If there is a significant interaction between the width and the diameter, it may be found that they jointly affect the temperature change, and optimizing only one parameter may not achieve the optimal temperature control effect. Through this global analysis, engineers can obtain more comprehensive and accurate sensitivity information, providing a basis for the subsequent optimization process.

[0127] The Monte Carlo method proposed in this embodiment has strong adaptability and can handle the nonlinearity of various input parameters and complex interaction effects. In the wellhead valve temperature optimization problem, this feature is particularly important because the change of the wellhead valve temperature may be jointly affected by multiple parameters and their interaction effects. Through global sensitivity analysis, engineers and other staff can deeply explore these interaction effects to ensure that the impacts of all parameters can be comprehensively considered in the optimization process. Especially in multi-objective optimization, sensitivity analysis can help balance the contradictions and conflicts between different objectives, rationally allocate computing resources, and avoid over-optimizing in a certain local area while ignoring the global performance. Generally speaking, Monte Carlo-based global sensitivity analysis can not only provide an intuitive ranking of parameter importance but also help identify potential interaction effects between parameters, thus providing strong support for further optimization design and decision-making.

[0128] The following Table 5 shows the results of parameter sensitivity verification, where X1 is the diameter of the pressure relief hole, X2 is the width of the pressure relief hole, and X3 is the longitudinal position of the pressure relief hole. Through the sensitivity analysis of three parameters, namely diameter ((x_1)), width ((x_2)), and position ((x_3)), the results show that the position ((x_3)) has the most significant influence on the wellhead valve temperature, with the highest sensitivity index. This indicates that the position is the key parameter affecting the temperature change, and optimizing this parameter can significantly improve the performance of the wellhead valve. Therefore, in the further optimization process, the value of the diameter should be given priority to be adjusted to achieve better temperature control effects.

[0129] Table 5

[0130] X1 X2 X3 0.2786 0.2901 0.4312

[0131] Step S4: Using the genetic algorithm to optimize the diameter a, position b, and width c of the pressure relief hole according to the contribution of each input parameter to the output result, and obtaining the optimal parameters of the diameter a, position b, and width c of the pressure relief hole.

[0132] When solving the wellhead valve temperature optimization problem using the Genetic Algorithm (GA), three control parameters, namely position, width, and diameter, are mainly concerned. As an optimization method that simulates the natural selection and genetic mechanism, the genetic algorithm continuously approaches the optimal solution through the evolutionary process of the population. In this process, position, width, and diameter are regarded as important variables in the gene chromosome. Each individual represents a potential solution in the solution space, and these individuals gradually produce more advantageous offspring through operations such as crossover, mutation, and selection. Specifically, the genetic algorithm first initializes an initial population composed of multiple individuals (i.e., parameter combinations). The fitness of each individual is evaluated through a surrogate model, which can predict the wellhead valve temperature under given parameters. The establishment of the surrogate model depends on historical data or experimental data, and fits the relationship between temperature and control parameters through machine learning methods (such as neural networks or regression analysis), so as to provide temperature prediction information for the genetic algorithm.

[0133] As the genetic algorithm evolves, the individuals in the population gradually show better adaptability, that is, they can obtain a lower valve temperature under specific parameter settings. Through the crossover operation, the genetic algorithm can combine the genes of excellent individuals into new individuals, further improving the solution quality in the search space; while the mutation operation helps to avoid falling into local optimal solutions and explores new possible solutions through a certain degree of randomness. In each generation, individuals are selected through the selection mechanism to ensure that the optimal solution continues to reproduce in the population. The advantage of the genetic algorithm lies in its global search ability, which can effectively avoid local optimal solutions and handle complex and non-linear optimization problems. Therefore, in the wellhead valve temperature optimization, the genetic algorithm can search for an optimal solution that can minimize or optimize the wellhead valve temperature by adjusting the three parameters of position, width, and diameter. The final optimization results are shown in Table 6 below.

[0134] Table 6

[0135] X1 X2 X3 55.10 18.89 55.72

[0136] As can be seen from Table 5, the optimal diameter a of the pressure relief hole is 55.10 mm, the position b is 55.72 mm, and the width c is 18.89 mm.

[0137] Embodiment 4. Refer to Figures 6 to 11 In this embodiment, this embodiment is a verification explanation of the results obtained by optimizing the diameter a, position b, and width c of the pressure relief hole by using the genetic algorithm described in the above Embodiment 3;

[0138] The optimized result of the genetic algorithm corresponds to a pressure relief hole diameter of 55.10 mm, a width of 18.89 mm, and a relative position of 55.72 mm from the bottom of the valve plate. According to the above structure and calculation method, the simulation calculation results are as follows Figures 3 to 8 shown.

[0139] Among them, Figure 6 is the pressure contour map of the wellhead gate valve pipeline at 25% opening. It can be seen that at the inlet of the valve, the entire CO 2 pressure level is at the highest position. When CO 2 flows through the valve plate area of the valve, the pressure contour map changes greatly, forming a large pressure gradient. This is because the valve is not fully opened, and the small flow area at the valve plate forms a throttle. Similarly, when CO 2 flows out of the valve plate area, a relatively obvious pressure gradient is also formed. Here, CO 2 forms a secondary pressure reduction, and finally the pressure of CO 2 gradually approaches the atmospheric pressure.

[0140] Figure 7 is the velocity contour map of the wellhead gate valve pipeline at 25% opening. According to Bernoulli's principle, the pressure of the fluid will decrease and the flow velocity will increase when flowing through a slit. Therefore, it can be seen from the velocity contour map that when CO 2 flows through the valve plate area of the valve, the flow velocity of the fluid increases significantly, which is consistent with the change of the pressure contour map. At this time, the velocity reaches nearly 400 m / s, which is already in the supersonic state. Due to the compressibility of CO 2 gas, in this supersonic state, a strong compression wave similar to a Mach disk will be formed in the flow area after passing through the valve plate area, and a velocity gradient change will be formed before and after the compression wave. Specifically, the velocity will decrease slightly before the compression wave and increase to about three times the speed of sound after the compression wave.

[0141] Figure 8 and Figure 9 are the volume fraction contour maps of liquid and gaseous CO 2 in the wellhead gate valve pipeline at 25% opening, respectively. It can be seen that in the pipeline at the inlet, the volume fraction of liquid CO 2 is relatively large, about 20%, and the gas holdup here is 80%. Near the valve plate, due to the huge change in the temperature and pressure of CO 2 , CO 2 undergoes a violent phase change, and the gas holdup here increases significantly compared with the inlet pipeline, reaching a maximum of 100%. In a short section of the pipeline after CO 2 flows out of the valve area, due to the temperature drop of CO 2 , the above-mentioned potential freezing blockage area is formed. At this time, CO 2The gas holdup decreases to some extent compared with that inside the valve. In the latter half of the outlet pipeline, due to the pressure drop, CO 2 is no longer in the liquid state, and the gas holdup at this time is about 100%.

[0142] Figure 10 Figure 4 is the temperature contour map of the wellhead gate valve pipeline at 25% opening. It can be seen that at the inlet pipeline of the valve, the temperature of CO 2 is relatively high, remaining at about 288K. When CO 2 flows out of the valve plate area of the valve, the temperature of CO 2 drops suddenly, forming the lowest temperature at this time. According to the standard phase diagram of CO 2 , the triple point temperature of CO 2 is 217K. It can be seen that after CO 2 flows out of the valve plate area of the valve, the CO 2 in the pipeline is lower than 217K in some areas. At this time, CO 2 will undergo a large phase change under the drastic change of pressure and temperature. At the same time, since the temperature is lower than the triple point, there is a possibility of CO 2 freezing, that is, there is a risk of freezing and blockage in the wellhead gate valve at this time.

[0143] Figure 11 Figure 5 shows the area where the valve may be at risk of freezing and blockage. By using the area of the region where the temperature contour map on the symmetry plane is lower than the triple point as a parameter to measure the risk of valve freezing and blockage, it can be known that the larger this area is, the greater the risk of freezing and blockage of the wellhead gate valve during use. After calculation, at 25% opening, the area ratio of the region lower than the triple point is 0.3108.

[0144] According to the simulation results, it is calculated that under the optimization of the genetic algorithm, the S of the area at risk of freezing and blockage is 0.3108, which is reduced to a certain extent compared with S = 0.3249 in the original structure. That is, the structure optimized by the genetic algorithm has a certain effect on reducing the risk of freezing and blockage of the wellhead valve.

[0145] Embodiment 6: This embodiment specifically describes the principle of the Monte Carlo-based global sensitivity analysis method described in the above embodiment;

[0146] Consider a square-integrable function y = f(x) that has n-dimensional independent input variables defined in the unit hypercube [0,1] n The Sobol decomposition of f(x) is expressed as the sum of a set of functions with increasing dimensions:

[0147]

[0148] where f 0 is a constant, and each summation term The integral over any of its independent variables is zero, i.e.:

[0149]

[0150] The number of terms in the summation in the above formula is:

[0151]

[0152] The constant f 0 is the average value of the function:

[0153]

[0154] where, for simplicity, dx represents dx 1 , …, dx n . In the following sense, the terms in the summation are orthogonal to each other:

[0155]

[0156] According to the above assumptions, as long as f(x) is integrable on , the decomposition in the equation is unique. In addition, the terms in the decomposition can be analyzed. In fact, the single-variable term is:

[0157]

[0158] In this expression, denotes the integration over all variables except x i . Similarly, the two-variable term is:

[0159]

[0160] where, denotes the integration over all variables except x i and x j . Generally, the symbol "~" in the formula represents "complementary". According to this structure, any term can be written as the difference between a multi-dimensional integral and a lower-order term.

[0161] Now consider that the input parameters are independent random variables uniformly distributed in the interval [0, 1]:

[0162] Therefore, the model response Y = f(X) is a random variable, and its variance D (also called the total variance) is:

[0163]

[0164] By integrating the square of the equation

[0165]

[0166] Combining the above formulas, the total variance can be decomposed as follows:

[0167]

[0168] Among them, the partial variances appearing in the above expansion are as follows:

[0169]

[0170] The Sobol global sensitivity index is defined as follows:

[0171]

[0172] According to the definition, combining the above formulas, they satisfy:

[0173]

[0174] Therefore, each index is a sensitivity measure that describes how much of the total variance is due to the uncertainty of the set of input parameters {i 1 ,…,i s}. The first-order sensitivity index S i gives the effect of each parameter acting alone on the output, while the higher-order sensitivity index illustrates the effect of the possible synergistic effects of various parameters on the output.

[0175] The total sensitivity index is further defined to evaluate the total effect of the input parameters. They are defined as the sum of all sensitivity indices containing parameter i :

[0176]

[0177] It is easy to see that:

[0178]

[0179] Among them, S ～i represents the sum of all those that do not contain index i .

[0180] The Sobol global sensitivity index is usually calculated using Monte Carlo simulation. According to the equations and, the mean value of the response, and the estimated values of the total variance and partial variance can be obtained using N sim samples:

[0181]

[0182] In the last equation, x m =(x1m , x 2m , …, x nm ), and

[0183]

[0184] In addition, the superscripts (1) and (2) in the equation respectively indicate that two different samples are generated and mixed. Similar expressions allow the overall sensitivity index to be estimated in one go

[0185]

[0186]

[0187] Embodiment Seven. Refer to Figure 12 To illustrate this embodiment, this embodiment specifically explains the principle of the genetic algorithm described in the above embodiments;

[0188] The core part of the standard genetic algorithm consists of five parts: individual coding, population initialization, fitness function, genetic operations, and control parameters. The genetic operations of the genetic algorithm include selection, crossover, and mutation. The specific workflow of the standard genetic algorithm will be introduced in this section. Before that, a brief introduction to individual coding, fitness function, and genetic operations will be given first.

[0189] Individual coding: In the genetic algorithm, an "individual" or "chromosome" represents a solution to the optimization problem. The genetic algorithm cannot directly search the set of feasible solutions to the actual optimization problem, so it is necessary to convert the feasible solutions to the actual optimization problem into a type that the genetic algorithm can recognize through coding. Common coding methods mainly include binary coding, permutation coding, integer coding, etc. The individual coding method mainly depends on the actual problem to be solved, so it is also possible to define the coding method based on the actual optimization problem to be solved.

[0190] Fitness function: In the genetic algorithm, the fitness function is used to evaluate the quality of individuals (solutions to the optimization problem) in the population, that is, the fitness function is designed based on the objective function to be optimized in the optimization problem. In the selection stage of the genetic algorithm, the fitness function is used to evaluate the advantages and disadvantages of individuals, so as to select excellent individuals to participate in evolution.

[0191] The genetic operations of the standard genetic algorithm consist of three operations: selection, crossover, and mutation:

[0192] Selection: The selection operation simulates the "survival of the fittest" law in nature and is an important part of the genetic algorithm. The commonly used selection strategies mainly include roulette wheel selection and tournament selection. The idea behind roulette wheel selection is that the probability of an individual being selected is positively correlated with the magnitude of its fitness value, that is, the larger the fitness value of an individual, the greater the probability of being selected. The calculation formula for the probability of an individual being selected is shown in Equation (6.60), where f(a i ) represents the fitness value of the i-th individual, and " represents the total number of individuals in the population.

[0193]

[0194] The above roulette wheel selection strategy is based on probability, which is different from the tournament selection strategy. Tournament selection directly selects by comparing the fitness function values. The specific steps are as follows: First, randomly select a certain number of individuals from the population (with replacement sampling), and then select the individual with the largest fitness value among them to join the offspring population. Repeat the above steps until the size of the offspring population reaches the size of the original population.

[0195] Crossover and Mutation: In the genetic algorithm, the crossover operation simulates the crossover of chromosomes during biological reproduction, and its purpose is to generate new individuals. Through the crossover operation, the excellent genes of the parent individuals can be inherited to the offspring individuals, which is an important way for the genetic algorithm to obtain excellent individuals. In the crossover operation, two parent individuals exchange gene segments to generate offspring individuals. Specific crossover strategies include single-point crossover, two-point crossover, multi-point crossover, etc., and appropriate crossover strategies can be selected according to different types of coding methods. The mutation operation simulates gene mutations in the process of biological evolution, and randomly changes some gene segments of an individual to form a new individual. The mutation operation can bring new gene segments to the population, maintaining the diversity of individuals in the population to a certain extent and preventing the genetic algorithm from falling into a local optimum during the search process. In summary, the crossover and mutation operations are effective guarantees for the genetic algorithm to search for the optimal solution. To sum up, the specific workflow framework of the standard genetic algorithm is as Figure 12 shown.

[0196] The above is only the implementation mode of the present invention and does not limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the scope of the claims of the present invention.

Claims

1. A wellhead valve, comprising a valve plate, characterized in that: A pressure relief hole is provided on the back of the valve plate.

2. The wellhead valve according to claim 1, characterized in that: The diameter a of the pressure relief hole ranges from 55mm to 65mm; The pressure relief hole position b ranges from 40mm to 60mm; The width c of the pressure reducing hole ranges from 14 mm to 22 mm.

3. A method for optimizing wellhead valve parameters based on intelligent algorithms, characterized in that: The optimization method is to optimize the parameters of the pressure relief hole in the wellhead valve according to claim 2, and the method is: S1: According to the diameter a, position b and width c of the pressure relief hole, sampling is performed through Latin hypercube to obtain each sample point; S2: construct a radial basis function proxy model based on each sample point; S3: A global sensitivity analysis method based on Monte Carlo is used to sample the diameter a, position b and width c of the pressure relief hole to obtain different input combinations, calculate the output results of the radial basis function proxy model under different input combinations, and obtain the contribution of each input parameter to the output result; S4: According to the contribution of each input parameter to the output result, the diameter a, position b, and width c of the pressure relief hole are optimized by using a genetic algorithm to obtain the optimal parameters of the diameter a, position b, and width c of the pressure relief hole.

4. The method for optimizing wellhead valve parameters based on intelligent algorithm according to claim 3 is characterized in that: Each sample point corresponds to the diameter a, position b and width c of the pressure relief hole.

5. The method for optimizing wellhead valve parameters based on intelligent algorithm according to claim 3 is characterized in that: The output of the Monte Carlo-based global sensitivity analysis method is the temperature of the wellhead valve.

6. The method for optimizing wellhead valve parameters based on intelligent algorithm according to claim 4 is characterized in that: By statistically analyzing the variance of the output results, the total variance of the temperature change is decomposed into the contribution of different input parameters, and the contribution degree of each input parameter to the output result is obtained.

7. The method for optimizing wellhead valve parameters based on intelligent algorithm according to claim 3, characterized in that: The optimal pressure relief hole diameter a is 55.10 mm, position b is 55.72 mm, and width c is 18.89 mm.

8. Wellhead valve optimization system based on intelligent algorithm, characterized by: The optimization system includes a storage device, which is used to execute the wellhead valve parameter optimization method and steps based on the intelligent algorithm as described in claim 2.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for optimizing wellhead valve parameters based on an intelligent algorithm as described in any one of claims 2 to 7 is executed.

10. A computer device, characterized in that: The device includes a memory and a processor, wherein a computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes the wellhead valve parameter optimization method based on an intelligent algorithm as described in any one of claims 2-7.

Images