Valve runner optimization method based on simulation model

Abstract

The invention belongs to the technical field of valve manufacturing, and particularly discloses a valve flow channel optimization method based on a simulation model, which comprises the following steps: acquiring flow channel optimization parameters and determining an optimization threshold value, constructing a parameterized geometric model, simulating flow property parameters through a dynamic simulation model, and screening an optimal flow channel. Wherein the flow channel optimization parameters comprise the curvature radius and the sectional area change rate, and the curvature radius optimization comprises the steps that a key curvature radius variable threshold value is set, and a curvature radius optimization model is combined with a four-quadrant method to analyze fluid pressure drop and turbulence intensity so as to compensate the optimization parameters; in the sectional area change rate optimization, a time attribute is added to a flow parameter mutation point, a neural network model is constructed to train a mapping relation between a flow parameter and a sectional area change rate, and the mapping relation is dynamically adjusted. According to the method, the problems of insufficient flow property optimization, poor dynamic working condition adaptability, low design efficiency and the like in the traditional design are solved, multi-parameter collaborative optimization of the flow channel is realized, and the stability and reliability of the valve under the complex working condition are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of valve manufacturing, and particularly relates to a method for optimizing a valve flow channel based on a simulation model. Background Art

[0002] In the field of valve manufacturing, the design of the flow channel structure is the core link determining the fluid control performance of the valve. The geometric parameters of the flow channel directly affect the flow resistance of the fluid, the turbulent distribution, and the energy loss. The adaptability of the flow channel under dynamic conditions determines the stability of the valve in complex scenarios. With the development of industrial fluid systems towards high-pressure and high-frequency dynamic regulation, higher requirements are put forward for the refined design and multi-parameter collaborative optimization of the valve flow channel.

[0003] In the prior art, the traditional flow channel design mainly relies on engineers' experience for parameter setting and structure adjustment. It is difficult to quantitatively analyze the mapping relationship between parameters such as the curvature radius and the cross-sectional area change rate and the flow performance, resulting in the optimization of the flow channel structure remaining at the empirical level, and problems such as excessive pressure drop or flow field disorder are likely to occur; moreover, the multi-parameter optimization process relies on physical test trial and error, with a long design cycle and high cost, and it is easy to cause parameter optimization conflicts.

[0004] Therefore, it is urgent to propose a method for optimizing a valve flow channel based on a simulation model to solve the above technical problems. Summary of the Invention

[0005] The purpose of the present invention is to overcome the defects existing in the prior art and provide a method for optimizing a valve flow channel based on a simulation model.

[0006] The present invention provides a method for optimizing a valve flow channel based on a simulation model, including: Obtaining flow channel optimization parameters, determining a flow channel optimization threshold based on the flow channel optimization parameters, constructing a parametric geometric model, and generating multiple groups of flow channel schemes based on the flow channel optimization threshold; Simulating and analyzing the flow performance parameters of different flow channels through a dynamic simulation model, and screening out the valve flow channel with the optimal flow performance parameters.

[0007] A further solution is that the flow channel optimization parameters at least include the curvature radius and the cross-sectional area change rate; The optimization process of the curvature radius is: setting the key curvature radius as a variable and determining the variable threshold of each key curvature radius based on the design requirements, and outputting a curvature radius compensation optimization strategy through a curvature radius optimization model; The optimization process of the cross-sectional area change rate is as follows: Measuring the flow parameters of the fluid at several cross-sections of the flow channel as the input of the dynamic simulation model to obtain the instantaneous process of the fluid flow in the flow channel; and measuring the cross-sectional area change rate of the cross-section corresponding to the flow parameters to obtain the mapping relationship between the flow parameters and the cross-sectional area change rate, and dynamically adjusting the cross-sectional area change rate based on the mapping relationship and the preset flow parameters; the cross-sectional area change rate is the ratio of the cross-sectional area of the flow channel at any moment when the fluid flows through the flow channel to the cross-sectional area at the valve body inlet.

[0008] A further solution is that the key curvature radii are: the inlet transition curvature radius, at the connection between the pipe before the valve and the valve body; the throat contraction curvature radius, which affects the peak flow velocity; the outlet diffusion curvature radius, which affects the pressure drop recovery.

[0009] A further solution is that the curvature radius optimization model includes an input layer, a curvature radius library, a conditional selection gate, a fluid simulation compensation unit, and an output layer; The input layer includes at least three independent input paths, which are respectively used to input the variable threshold of the inlet transition curvature radius, the variable threshold of the throat contraction curvature radius, and the variable threshold of the outlet diffusion curvature radius; The curvature radius library is respectively connected to the three independent input paths and is used to temporarily store the variable thresholds; The conditional selection gate is connected to the curvature radius library and serves as the only data transmission channel between the curvature radius library and the fluid simulation compensation unit. The conditional selection gate selects the optimal variable threshold of the key curvature radius in the curvature radius library based on the application scenario and parameter requirements of the valve, and clears the curvature radius library; The fluid simulation compensation unit performs fluid dynamic simulation based on the optimal variable threshold and compensates the optimal variable threshold based on the fluid dynamic simulation results; The output layer is used to output the compensated optimal inlet transition curvature radius, the optimal throat contraction curvature radius, and the optimal outlet diffusion curvature radius.

[0010] A further solution is that the compensation process of the fluid simulation compensation unit for the optimal variable threshold is as follows: Quantify the flow performance parameters into fluid pressure drop variables and fluid turbulence intensity variables and set the pressure drop threshold and turbulence intensity threshold; Respectively obtain the flow performance parameters of the fluid dynamic simulation at the connection between the pipe before the valve and the valve body, at the throat contraction, and at the connection between the pipe after the valve and the valve body, and compare them with the pressure drop threshold and the turbulence intensity threshold; The four - quadrant method is used to analyze the fluid pressure drop and fluid turbulence intensity. If the fluid pressure drop and fluid turbulence intensity are in the first quadrant, the optimal variable threshold is increased geometrically based on the increasing ratio of the fluid pressure drop and fluid turbulence intensity; if the fluid pressure drop and fluid turbulence intensity are in the third quadrant, the optimal variable threshold is decreased geometrically based on the decreasing ratio of the fluid pressure drop and fluid turbulence intensity; if the fluid pressure drop and fluid turbulence intensity are in the second quadrant or the fourth quadrant, it is considered abnormal, and the condition selection gate re - selects the optimal variable threshold.

[0011] A further solution is that the calculation formula for increasing or decreasing the optimal variable threshold geometrically is: ; where, is the increase or decrease amount of the optimal variable threshold, is the fluid pressure drop, is the pressure drop threshold, is the fluid turbulence intensity, is the turbulence intensity threshold, is the optimal critical curvature radius.

[0012] A further solution is to add a time attribute to the mutation point of the flow parameters and synchronize the time attribute to the cross - sectional area change rate, so that the cross - sectional area change rates with different time attributes correspond to the mutation points of the flow parameters at that time attribute.

[0013] A further solution is that the preset flow parameters are set based on the application scenario and parameter requirements of the condition selection gate; the time interval of the preset flow parameters is set based on the time attribute of the mutation point, so that any time attribute of the cross - sectional area change rate is within the time interval and at that time attribute, the cross - sectional area change rate corresponds uniquely to the preset flow parameters.

[0014] A further solution is that the process of dynamically adjusting the cross - sectional area change rate is as follows: collect a large amount of pressure data and flow velocity data with time tags, and calculate the cross - sectional area change rate of the flow channel at the moment of the time tag, and manually mark the mutation time attributes of the pressure data and flow velocity data by an expert; The pressure data , the flow velocity data , the cross - sectional area change rate and the mutation point time attribute are manually marked by an expert to form several groups of mapped pairs with time tags as the training set; The mapped pairs are input into the neural network unit for iterative training to obtain the cross - sectional area change rate based on the pressure data , the flow velocity data , the mutation point time attribute Output cross-sectional area change rate Optimization model of cross-sectional area change rate; Dynamically adjust the cross-sectional area change rate based on the optimization model of cross-sectional area change rate .

[0015] A further solution is to use the pressure drop threshold and the turbulence intensity threshold as the global constraint conditions of the cross-sectional area change rate optimization model. Compare the fluid pressure drop and the fluid turbulence intensity of the cross-section after the cross-sectional area change rate is adjusted with the pressure drop threshold and the turbulence intensity threshold respectively, and trigger fine-tuning measurement based on the comparison results: If , , reduce the cross-sectional area change rate according to the following formula: ; If , , increase the cross-sectional area change rate according to the following formula: ; Wherein, is the reduced or increased cross-sectional area change rate.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: Through parametric geometric models and dynamic simulation analysis, the present invention quantitatively analyzes the influence of flow channel optimization parameters on flow performance parameters, can compare multiple groups of flow channel schemes at the same time, improves the optimization efficiency, ensures the selection of the optimal flow channel scheme, reduces the fluid pressure drop and turbulence intensity, and improves the energy utilization efficiency.

[0017] The present invention focuses on the radius of curvature and the cross-sectional area change rate. Through variable threshold setting and dynamic optimization, it accurately controls the flow characteristics of the fluid at key positions such as the pre-valve connection, throat contraction, and post-valve diffusion, reduces the local resistance loss, and realizes the refined design of the flow channel structure by distinguishing the different effects of the radius of curvature on the flow performance, avoiding the performance conflicts caused by parameter coupling in traditional empirical design.

[0018] The present invention constructs an optimization model of the radius of curvature, analyzes the pressure drop and turbulence intensity by the four-quadrant method, and dynamically compensates the radius of curvature threshold through a calculation formula. Among them, the condition selection gate automatically matches the radius of curvature threshold based on the working condition requirements, and combines the fluid simulation compensation unit to realize the intelligent iteration of parameters, solving the problem of insufficient adaptability to multiple working conditions in traditional design, so that the flow channel structure can quickly adapt to complex scenarios such as high pressure and high-frequency regulation; the four-quadrant method and the formula-based compensation mechanism ( Calculation formula) quantify the parameter adjustment logic, avoid the subjectivity of manual experience, ensure the scientificity and repeatability of the radius of curvature optimization process, and improve the reliability of the optimization results.

[0019] The present invention breaks through the neglect of the time dimension in traditional designs. By marking the mutation points of flow parameters with time attributes, it synchronizes the cross-sectional area change rate precisely with the dynamic working conditions, solves the problem of lag in the flow field response of the valve during the transient process, and improves the stability during working condition mutations. The neural network model of the present invention establishes a non-linear mapping relationship between pressure, flow velocity, time, and cross-sectional area change rate through big data training, realizes the adaptive adjustment of the flow channel structure, and has a stronger fitting ability for flow parameters (such as periodic load fluctuations) in complex time intervals. The present invention uses the pressure drop threshold and the turbulence intensity threshold as the global constraint conditions for the cross-sectional area change rate optimization model to ensure that the adjustment of the cross-sectional area change rate does not exceed the performance bottom line. Combining with a formula-based fine-tuning mechanism, it maintains the balance of the flow channel performance in dynamic optimization and avoids performance overrun caused by single-parameter optimization. Description of the Drawings

[0020] The following drawings only schematically illustrate and explain the present invention and are not used to limit the scope of the present invention, where: Figure 1 is the flow chart of the valve flow channel optimization method of the present invention; Figure 2 is the principle block diagram of the curvature radius optimization model; Figure 3 is the schematic diagram of the four-quadrant distribution of fluid pressure drop and fluid turbulence intensity. Detailed Embodiments

[0021] In order to make the purpose, technical solutions, design methods, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings through specific embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0022] As Figure 1 shown, the present invention provides a valve flow channel optimization method based on a simulation model, including the following steps: Step 1: Obtain the flow channel optimization parameters. In this embodiment, the flow channel optimization parameters are determined to be the curvature radius and the cross-sectional area change rate. Among them, the curvature radius includes three types of key parameters: the inlet transition curvature radius : the curvature at the connection between the pipe before the valve and the valve body, which affects the smoothness of the fluid entering the valve body; the throat contraction curvature radius : the contraction curvature of the valve body throat (the area of peak flow velocity), which determines the flow velocity distribution; the outlet diffusion curvature radius : the diffusion curvature at the connection between the pipe after the valve and the valve body, which affects the pressure drop recovery. The cross-sectional area change rate : is defined as the ratio of the cross-sectional area of the fluid flowing through the flow channel at any moment to the cross-sectional area of the valve body inlet, and is used to characterize the dynamic change of the flow channel cross-section.

[0023] Step 2: Determine the flow channel optimization threshold based on the flow channel optimization parameters, construct a parametric geometric model through 3D software, and set , , and as adjustable variables, set the initial variable threshold range, and generate multiple groups of flow channel schemes through parametric modeling. Each group of schemes corresponds to different combinations of curvature radii and sequences of cross-sectional area change rates; Simulate the flow performance parameters of different flow channels through a dynamic simulation model, and screen the valve flow channel with the optimal flow performance parameters.

[0024] Step 3: The curvature radius optimization process is as follows: Set the key curvature radius as a variable and determine the variable threshold of each key curvature radius based on the design requirements, and output the curvature radius compensation optimization strategy through the curvature radius optimization model; among them, as Figure 2 shown, the curvature radius optimization model includes an input layer, a curvature radius library, a condition selection gate, a fluid simulation compensation unit, and an output layer; The input layer includes at least three independent input paths, and the variable threshold ranges of , , are input through the three independent paths respectively, supporting manual input or automatic loading of default thresholds based on industry standards (such as ASME B16.34); The curvature radius library is respectively connected to the three independent input paths, and is used to temporarily store the input threshold ranges to form a discretized parameter combination (such as , , are each divided into 5 intervals to generate 125 combinations); The condition selection gate is connected to the curvature radius library and serves as the only data transmission channel between the curvature radius library and the fluid simulation compensation unit. The condition selection gate calls preset rules according to the valve application scenario (such as high-pressure working conditions, high-frequency regulation working conditions) to screen out the suitable curvature radius combination. For example, in high-pressure working conditions, a smaller is preferably selected to increase the flow velocity, while in high-frequency regulation working conditions, a larger , is preferably selected to reduce turbulence.

[0025] Delete the obviously unreasonable parameter combinations (such as or ), and empty the non-preferred combinations in the curvature radius library; The fluid simulation compensation unit performs hydrodynamic simulation based on the optimal variable threshold, and performs hydrodynamic simulation on the selected curvature radius combinations using CFD simulation software (such as ANSYS Fluent) to obtain the fluid pressure drops at the connection before the valve, the throat, and the connection after the valve. and the turbulence intensity , and compensates the optimal variable threshold based on the hydrodynamic simulation results; The output layer is used to output the compensated , , , which are used to update the geometric model.

[0026] In the above, the process of the fluid simulation compensation unit compensating the optimal variable threshold is as follows: Quantify the flow performance parameters into fluid pressure drop variables and fluid turbulence intensity variables and set the pressure drop threshold and the turbulence intensity threshold; Obtain the flow performance parameters of the hydrodynamic simulation at the connection between the pipe before the valve and the valve body, the throat constriction, and the connection between the pipe after the valve and the valve body respectively, and compare them with the pressure drop threshold and the turbulence intensity threshold; As Figure 3 shown, use the four-quadrant method to analyze the fluid pressure drop and the fluid turbulence intensity. If the fluid pressure drop and the fluid turbulence intensity are in the first quadrant, it means that both the pressure drop and the turbulence exceed the standard. Increase the optimal variable threshold in proportion to the increase ratio of the fluid pressure drop and the fluid turbulence intensity to increase the smoothness of the flow channel. The calculation formula is: ; If the fluid pressure drop and the fluid turbulence intensity are in the third quadrant, it means that both the pressure drop and the turbulence are insufficient. Decrease the optimal variable threshold in proportion to the decrease ratio of the fluid pressure drop and the fluid turbulence intensity to enhance the fluid constraint; The calculation formula is: ; Among them, is the increase or decrease amount of the optimal variable threshold, is the fluid pressure drop, is the pressure drop threshold, is the fluid turbulence intensity, is the turbulence intensity threshold, is the optimal critical curvature radius.

[0027] If the fluid pressure drop and the fluid turbulence intensity are in the second quadrant or the fourth quadrant, it is determined that there is a parameter conflict (such as the pressure drop meets the standard but the turbulence exceeds the standard), and the trigger condition selection gate is triggered to re-screen the combination.

[0028] Step 4, optimization of cross-sectional area change rate: Measure the flow parameters of the fluid at several cross-sections of the flow channel as the input of the dynamic simulation model to obtain the instantaneous process of fluid flow in the flow channel; and measure the cross-sectional area change rate of the cross-section corresponding to the flow parameters to obtain the mapping relationship between the flow parameters and the cross-sectional area change rate, and dynamically adjust the cross-sectional area change rate based on the mapping relationship and the preset flow parameters; the cross-sectional area change rate is the ratio of the cross-sectional area of the flow channel at any moment when the fluid flows through the flow channel to the cross-sectional area at the valve body inlet. In this embodiment, the specific implementation process of Step 4 includes: Associate the flow parameters with time attributes: Add time attributes to the mutation points of the above flow parameters, and synchronize the time attributes to the cross-sectional area change rate, so that the cross-sectional area change rates with different time attributes correspond to the mutation points of the flow parameters with the time attributes. The preset flow parameters are set based on the application scenario and parameter requirements of the conditional selection gate; Set the time interval of the preset flow parameters based on the time attributes of the mutation points of the flow parameters, so that any time attribute of the cross-sectional area change rate is within the time interval and the cross-sectional area change rate corresponding to any time attribute is uniquely corresponding to the preset flow parameters. Specifically, the pressure data during the operation of the valve is collected in real time through a pressure sensor and a flow velocity meter , flow velocity data , cross-sectional area change rate and the time attributes of the mutation points are manually marked by experts to form several groups of mapping pairs with time tags as the training set; Input the mapping pairs into the neural network unit for iterative training to obtain the cross-sectional area change rate optimization model that outputs the cross-sectional area change rate based on the pressure data , flow velocity data , time attributes of the mutation points ; Based on the cross-sectional area change rate optimization model, dynamically adjust the cross-sectional area change rate . .

[0029] In the above, the value of the preset flow parameters needs to be directly bound to the actual application scenario and functional requirements of the valve, and is dynamically determined by the conditional selection gate according to the specific scenario to achieve the accuracy and working condition adaptability of the flow channel optimization; in this embodiment, the preset flow parameters refer to the performance index thresholds preset during the flow channel optimization, which are used to measure the rationality of the flow channel structure. For example: pressure drop threshold : the maximum allowable pressure drop value; turbulence intensity threshold : The maximum allowable turbulence intensity; the peak flow velocity threshold: the desired flow velocity range under specific working conditions; the condition selection gate is the core decision-making unit in the curvature radius optimization model, and its functions include: obtaining the application scenarios of the valve (such as high-pressure pipelines, chemical fluid regulation, high-frequency start-stop scenarios, etc.), selecting the initial variable thresholds matching the working conditions from the curvature radius library, and triggering the subsequent simulation compensation process.

[0030] Therefore, when determining the values of the preset flow parameters, the value range needs to be determined according to the application scenario. For example, in high-pressure working conditions, the fluid pressure is high, and the flow channel resistance needs to be reduced to reduce energy loss. The condition selection gate preferentially selects a larger inlet transition curvature radius and the outlet diffusion curvature radius , to reduce fluid impact; the pressure drop threshold is set to a lower value (such as ), forcing the flow channel optimization to reduce the pressure drop; the turbulence intensity threshold can be appropriately relaxed (such as ), allowing a certain degree of turbulence to ensure the flow capacity. In high-frequency regulation working conditions, the valve needs to be opened or closed frequently, and the transient response stability needs to be improved. The condition selection gate preferentially selects a smaller throat contraction curvature radius , to quickly change the flow velocity; the pressure drop threshold can be appropriately increased (such as ), allowing a certain pressure drop in exchange for the flow velocity regulation sensitivity; the turbulence intensity threshold is strictly limited (such as ), to avoid component wear caused by high-frequency turbulence.

[0031] In the above, the neural network unit uses LSTM (Long Short-Term Memory Network) to construct the cross-sectional area change rate optimization model. The input layer is , , , and the output layer is . The training process is: input the training set into the model, optimize the network weights through the backpropagation algorithm, and the goal is to minimize the mean square error between the predicted and the actual value. The pressure drop threshold and the turbulence intensity threshold are introduced as global constraints to ensure that the simulation results corresponding to the output by the model do not exceed the performance boundary. Obtain the , , of the current working condition in real time, input them into the trained model, and output the real-time cross-sectional area change rate . If the simulation results show , , reduce the cross-sectional area change rate according to the following formula: ; If , , increase the cross-sectional area change rate according to the following formula: ; wherein, is the cross-sectional area change rate after reduction or increase.

[0032] Step 5, iterative optimization of the flow channel scheme, combine the optimization result of the curvature radius with the dynamic adjustment strategy of the cross-sectional area change rate to generate the final flow channel structure scheme; perform full-condition simulations (including steady-state and transient conditions) on all schemes, and screen out the scheme with the optimal flow performance parameters as the final design scheme.

[0033] The above has described the embodiments of the present invention. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other ordinary skill in the art in the technical field to understand the disclosed embodiments.

Claims

1. A valve flow path optimization method based on a simulation model, characterized in that Including: Obtain runner optimization parameters, determine a runner optimization threshold based on the runner optimization parameters, construct a parametric geometric model, and generate multiple groups of runner schemes based on the runner optimization threshold; Simulate the flow performance parameters of different runners through a dynamic simulation model, and screen the valve runner with the optimal flow performance parameters.

2. The valve flow path optimization method based on a simulation model according to claim 1, wherein The runner optimization parameters at least include the curvature radius and the cross-sectional area change rate; The optimization process of the curvature radius is as follows: Set the key curvature radii as variables and determine the variable thresholds of each key curvature radius based on design requirements, and output a curvature radius compensation optimization strategy through a curvature radius optimization model; The optimization process of the cross-sectional area change rate is as follows: Measure the flow parameters of the fluid at several cross-sections of the runner as the input of the dynamic simulation model to obtain the instantaneous process of the fluid flow in the runner; and measure the cross-sectional area change rate of the cross-section corresponding to the flow parameters to obtain the mapping relationship between the flow parameters and the cross-sectional area change rate, and dynamically adjust the cross-sectional area change rate based on the mapping relationship and preset flow parameters; The cross-sectional area change rate is the ratio of the cross-sectional area of the runner at any moment when the fluid flows through the runner to the cross-sectional area at the valve body inlet.

3. The valve flow channel optimization method based on a simulation model according to claim 2, wherein The key curvature radii are: the inlet transition curvature radius, the throat contraction curvature radius, and the outlet diffusion curvature radius.

4. The valve flow path optimization method based on a simulation model according to claim 3, wherein, The curvature radius optimization model includes an input layer, a curvature radius library, a conditional selection gate, a fluid simulation compensation unit, and an output layer; The input layer at least includes three independent input paths, which are respectively used to input the variable threshold of the inlet transition curvature radius, the variable threshold of the throat contraction curvature radius, and the variable threshold of the outlet diffusion curvature radius; The curvature radius library is respectively connected to the three independent input paths and is used to temporarily store the variable thresholds; The conditional selection gate is connected to the curvature radius library and serves as the only data transmission channel between the curvature radius library and the fluid simulation compensation unit. The conditional selection gate selects the optimal variable thresholds of the key curvature radii in the curvature radius library based on the application scenario and parameter requirements of the valve, and clears the curvature radius library; The fluid simulation compensation unit performs fluid dynamic simulation based on the optimal variable thresholds and compensates the optimal variable thresholds based on the fluid dynamic simulation results; The output layer is used to output the compensated optimal inlet transition curvature radius, optimal throat contraction curvature radius, and optimal outlet diffusion curvature radius.

5. The valve flow path optimization method based on a simulation model according to claim 4, wherein The compensation process of the fluid simulation compensation unit for the optimal variable thresholds is as follows: Quantify the flow performance parameters into a fluid pressure drop variable and a fluid turbulence intensity variable, and set a pressure drop threshold and a turbulence intensity threshold; Respectively obtain the flow performance parameters of the fluid dynamic simulation at the connection between the pipeline before the valve and the valve body, at the throat contraction, and at the connection between the pipeline after the valve and the valve body, and compare them with the pressure drop threshold and the turbulence intensity threshold; The four - quadrant method is used to analyze the fluid pressure drop and fluid turbulence intensity. If the fluid pressure drop and fluid turbulence intensity are in the first quadrant, the optimal variable threshold is increased geometrically based on the increasing ratio of the fluid pressure drop and fluid turbulence intensity; if the fluid pressure drop and fluid turbulence intensity are in the third quadrant, the optimal variable threshold is decreased geometrically based on the decreasing ratio of the fluid pressure drop and fluid turbulence intensity; if the fluid pressure drop and fluid turbulence intensity are in the second quadrant or the fourth quadrant, it is considered abnormal, and the condition selection gate re - selects the optimal variable threshold.

6. The valve flow path optimization method based on a simulation model according to claim 5, characterized in that The calculation formula for increasing or decreasing the optimal variable threshold geometrically is: ； Among them, is the increase or decrease amount of the optimal variable threshold, is the fluid pressure drop, is the pressure drop threshold, is the fluid turbulence intensity, is the turbulence intensity threshold, is the optimal critical curvature radius.

7. The valve flow path optimization method based on a simulation model according to claim 6, characterized in that Add a time attribute to the mutation point of the flow parameter, and synchronize the time attribute to the cross - sectional area change rate, so that the cross - sectional area change rates with different time attributes correspond to the flow parameter mutation points of the corresponding time attributes.

8. The valve flow path optimization method based on a simulation model according to claim 7, characterized in that, The preset flow parameter is set based on the application scenario and parameter requirements of the condition selection gate; the time interval of the preset flow parameter is set based on the time attribute of the mutation point of the flow parameter, so that any time attribute of the cross - sectional area change rate is within the time interval and the cross - sectional area change rate corresponding to any time attribute corresponds uniquely to the preset flow parameter.

9. The valve flow channel optimization method based on a simulation model according to claim 8, wherein The process of dynamically adjusting the cross - sectional area change rate is as follows: collect a large number of pressure data and flow velocity data with time tags, calculate the cross - sectional area change rate of the flow channel at the moment of the time tag, and manually mark the mutation time attributes of the pressure data and flow velocity data by experts; The pressure data , the flow rate data , the cross-sectional area change rate and the time attribute of the mutation point are manually marked by experts to form several groups of mapped pairs with time tags as the training set; Input the mapping number pairs into the neural network unit for iterative training to obtain a cross-sectional area change rate optimization model based on pressure data , flow rate data , mutation point time attribute and output cross-sectional area change rate ; Dynamically adjust the cross-sectional area change rate based on the cross-sectional area change rate optimization model .

10. The valve flow path optimization method based on a simulation model according to claim 9, characterized in that, Take the pressure drop threshold and the turbulence intensity threshold as the global constraint conditions of the cross - sectional area change rate optimization model, compare the fluid pressure drop and fluid turbulence intensity of the cross - section after the cross - sectional area change rate is adjusted with the pressure drop threshold and the turbulence intensity threshold respectively, and trigger fine - tuning measurement based on the comparison results: If , , reduce the cross-sectional area change rate according to the following formula: ； If , , increase the cross-sectional area change rate according to the following formula: ； wherein, is the cross-sectional area change rate after reduction or increase.

Images