Pump head flow field optimization correction method and system based on visual analysis

By optimizing the pump head design through visualization analysis and flow field simulation technology, the problem of low pump head design accuracy was solved, and the fluid performance and stability were improved, making it adaptable to complex flow environments and demanding working conditions.

CN118886366BActive Publication Date: 2024-11-26POWERCHINA SPEM CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411312427.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-20
Publication Date
2024-11-26
Estimated Expiration
2044-09-20

AI Technical Summary

Technical Problem

Existing technologies suffer from low design precision in pump head design, making it difficult to adapt to complex flow environments and demanding operating conditions, resulting in insufficient pump head performance.

Method used

By collecting the initial specifications of the impeller and guide shell inside the pump head, and using visualization 3D modeling and flow field simulation technology, a visualization model of the pump head is constructed and the flow field grid is divided. The flow field information and stability are analyzed, the specifications are adjusted, the flow field distribution is optimized, and finally the design is corrected and the flow field is verified.

Benefits of technology

It improves the fluid performance and stability of the pump head under actual working conditions, solves the problem of low design accuracy, and achieves more efficient and reliable fluid delivery.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118886366B_ABST
    Figure CN118886366B_ABST
Patent Text Reader

Abstract

The present invention discloses a pump head flow field optimization correction method and system based on visual analysis, which relates to the field of data processing technology, including: collecting initial specification information of the pump head, constructing a pump head visualization model; performing flow field simulation to obtain an initial simulation flow field, performing flow field grid division to obtain a flow field grid, and dividing the pump head visualization model to obtain a pump head model grid; performing specification indexing within the pump head visualization model to obtain multiple pump head grid specification information, performing pump head stability analysis in combination with the grid flow field information, and obtaining initial flow field stability information; adjusting the initial specification information, performing visual three-dimensional modeling and flow field simulation, optimizing the pump head flow field, obtaining optimal specification information and optimal simulation flow field, and performing pump head design correction and flow field verification. The present invention solves the technical problem of low design accuracy in the prior art when designing a pump head, and achieves the technical effect of improving the fluid performance and stability of the pump head under actual working conditions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for optimizing and correcting a pump head flow field based on visual analysis. Background Art

[0002] As the core component of the fluid transmission system, the design of the pump head has a direct impact on the performance, efficiency and stability of the entire system. In fluid dynamics design, the traditional pump head design mainly relies on engineering experience and experimental verification. However, this design method is often difficult to effectively improve the performance of the pump head in the face of complex flow environments and high-demand working conditions, and it is difficult to adapt to the modern industrial production requirements for efficient, precise and reliable fluid delivery, and there is a problem of low design accuracy. Summary of the invention

[0003] The present application provides a pump head flow field optimization correction method and system based on visual analysis, which is used to solve the technical problem of low design accuracy in the prior art when designing pump heads.

[0004] In view of the above problems, the present application provides a pump head flow field optimization correction method and system based on visual analysis.

[0005] The first aspect of the present application provides a pump head flow field optimization correction method based on visual analysis, the method comprising:

[0006] The initial specification information of the impeller and the guide shell in the pump head of the current molten salt pump is collected, and a pump head visualization model is constructed based on visual three-dimensional modeling; based on the pump head visualization model, flow field simulation is performed to obtain an initial simulated flow field, and flow field grid division is performed to obtain a plurality of flow field grids; and the pump head visualization model is divided according to the flow field grids to obtain a plurality of pump head model grids, wherein the greater the flow velocity in the flow field division grid, the greater the number of grids; based on the plurality of pump head model grids, specification indexing is performed in the pump head visualization model to obtain a plurality of pump head grids. Specification information is combined with the grid flow field information in the multiple flow field grids to perform pump head stability analysis to obtain initial flow field stability information; the initial specification information of the impeller and the guide shell is adjusted, and visual three-dimensional modeling and flow field simulation are performed to optimize the pump head flow field to obtain optimal specification information and optimal simulated flow field, wherein, based on the initial flow field stability information, combined with the rationality of the distribution of the multiple flow field grids divided in the optimization, optimization judgment is performed; according to the optimal specification information and the optimal simulated flow field, the design correction and flow field verification of the molten salt pump head are performed.

[0007] The second aspect of the present application provides a pump head flow field optimization correction system based on visual analysis, the system comprising:

[0008] A visualization model building module, the visualization model building module collects the initial specification information of the impeller and the guide shell in the pump head of the current molten salt pump, and builds a pump head visualization model based on visualization three-dimensional modeling; a flow field simulation module, the flow field simulation module performs flow field simulation based on the pump head visualization model, obtains the initial simulation flow field, performs flow field grid division, obtains multiple flow field grids, and divides the pump head visualization model according to the flow field grids to obtain multiple pump head model grids, wherein the greater the flow velocity in the flow field division grid, the greater the number of grids; a specification indexing module, the specification indexing module performs specification indexing in the pump head visualization model based on the multiple pump head model grids A plurality of pump head grid specification information is obtained, and the pump head stability analysis is performed in combination with the grid flow field information in the plurality of flow field grids to obtain initial flow field stability information; an optimization judgment module, wherein the optimization judgment module adjusts the initial specification information of the impeller and the guide shell, and performs visual three-dimensional modeling and flow field simulation to optimize the pump head flow field to obtain the optimal specification information and the optimal simulated flow field, wherein, based on the initial flow field stability information, an optimization judgment is performed in combination with the rationality of the distribution of the plurality of flow field grids divided in the optimization; a correction verification module, wherein the correction verification module performs design correction and flow field verification of the molten salt pump head according to the optimal specification information and the optimal simulated flow field.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] The present application collects the initial specification information of the impeller and the guide shell in the pump head of the current molten salt pump, and constructs a pump head visualization model based on visual three-dimensional modeling; based on the pump head visualization model, flow field simulation is performed to obtain the initial simulated flow field, flow field grid division is performed to obtain multiple flow field grids, and the pump head visualization model is divided according to the flow field grids to obtain multiple pump head model grids, wherein the greater the flow velocity in the flow field division grid, the greater the number of grids; specification indexing is performed in the pump head visualization model based on the multiple pump head model grids to obtain multiple pump head grids. The grid specification information is combined with the grid flow field information in multiple flow field grids to perform pump head stability analysis and obtain initial flow field stability information; the initial specification information of the impeller and the guide shell is adjusted, and visual three-dimensional modeling and flow field simulation are performed to optimize the pump head flow field and obtain the optimal specification information and the optimal simulated flow field, wherein, based on the initial flow field stability information and combined with the rationality of the distribution of multiple flow field grids divided in the optimization, optimization judgment is performed; according to the optimal specification information and the optimal simulated flow field, the design correction and flow field verification of the molten salt pump head are carried out. The present invention solves the technical problem of low design accuracy in the prior art when designing pump heads. By collecting the initial specification information of the impeller and the guide shell in the pump head of the molten salt pump, and utilizing visual three-dimensional modeling and flow field simulation technology, a pump head visualization model is constructed and the flow field grid is divided. The flow field information and stability of the pump head model grid are analyzed, the specification information is adjusted, the flow field distribution of the pump head is optimized, the optimal specifications and simulated flow field are obtained, and finally the optimization results are combined to perform design correction and flow field verification, so as to achieve the technical effect of improving the fluid performance and stability of the pump head under actual working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic flow chart of a pump head flow field optimization correction method based on visual analysis provided in an embodiment of the present application;

[0013] Figure 2 Schematic diagram of the structure of the pump head flow field optimization and correction system based on visual analysis provided in an embodiment of the present application.

[0014] Explanation of the reference numerals: visualization model building module 11, flow field simulation module 12, specification indexing module 13, optimization judgment module 14, correction verification module 15. DETAILED DESCRIPTION

[0015] The present application provides a pump head flow field optimization and correction method and system based on visual analysis, aiming to solve the technical problem of low design accuracy in the prior art when designing pump heads. By collecting the initial specification information of the impeller and the guide shell in the molten salt pump head, the visual three-dimensional modeling and flow field simulation technology are used to construct a pump head visualization model and divide the flow field grid, analyze the flow field information and stability of the pump head model grid, adjust the specification information, optimize the flow field distribution of the pump head, obtain the optimal specifications and simulated flow field, and finally combine the optimization results to perform design correction and flow field verification, so as to achieve the technical effect of improving the fluid performance and stability of the pump head under actual working conditions.

[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0017] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0018] Embodiment 1, as Figure 1 As shown, the present application provides a pump head flow field optimization correction method based on visual analysis, the method comprising:

[0019] Step S100: collecting initial specification information of the impeller and guide shell in the pump head of the current molten salt pump, and constructing a pump head visualization model based on visualized three-dimensional modeling.

[0020] In the embodiment of the present application, in order to collect the initial specification information of the impeller and guide shell in the current molten salt pump head, firstly, a high-precision measuring device, such as a laser rangefinder, is used to collect the specification data of the key components inside the pump head, such as the impeller and the guide shell. These key data include the diameter of the impeller, the number of blades, the inclination angle of the blades, the shape and inner diameter of the guide shell, etc. Next, based on the collected initial specification information, a three-dimensional modeling software, such as SolidWorks or AutoCAD, is used to perform a visual three-dimensional modeling of the pump head to complete the construction of the visual model of the pump head.

[0021] Furthermore, in the method provided in the embodiment of the application, the initial specification information of the impeller and the guide shell in the pump head of the current molten salt pump is collected, and a pump head visualization model is constructed based on the visualization three-dimensional modeling, which also includes:

[0022] The initial specification information of the impeller and the initial specification information of the guide shell in the pump head of the current molten salt pump are collected; and a pump head visualization model is constructed based on the visualization 3D modeling according to the initial specification information of the impeller and the initial specification information of the guide shell.

[0023] In the embodiment of the present application, high-precision measuring tools are first used, such as laser rangefinders, three-dimensional coordinate measuring machines, or non-contact scanners. Through these devices, key dimensional information of the impeller is obtained, such as the diameter, thickness, blade angle, and number of blades of the impeller. Similarly, information such as the inner and outer diameters, wall thickness, and shape characteristics of the guide shell are also collected through measuring equipment. These initial specification information defines the geometric characteristics of the impeller and guide shell in their current state.

[0024] Next, the collected initial specification information is input into 3D modeling software, such as SolidWorks, ANSYS or AutoCAD. In the software, based on the initial specification information of the impeller and the initial specification information of the guide shell, the visual model of the molten salt pump head is gradually constructed using 3D modeling tools. The 3D modeling process includes creating basic geometric shapes, such as cylinders, cones, etc., and then combining the geometric shapes into a complete model of the pump head through operations such as extrusion, rotation and cutting.

[0025] Through the above process, the construction of the pump head visualization model is completed.

[0026] Step S200: Based on the pump head visualization model, flow field simulation is performed to obtain an initial simulated flow field, flow field grid division is performed to obtain a plurality of flow field grids, and the pump head visualization model is divided according to the flow field grids to obtain a plurality of pump head model grids, wherein the greater the flow velocity in the flow field division grid, the greater the number of grids in the area.

[0027] In the embodiment of the present application, the pump head visualization model is first imported into the fluid simulation software, such as ANSYS Fluent or CFX. In the simulation, predetermined boundary conditions and fluid properties, such as the density and viscosity of the molten salt, are set, and the initial flow field data inside the pump head is obtained by solving the Navier-Stokes equation, including information such as flow velocity and pressure distribution in different areas.

[0028] After obtaining the initial simulation flow field, the next step is to divide the simulation space into flow field grids according to the flow velocity information of different areas. Specifically, flow field grid division is to divide the entire flow field space into multiple small areas, each of which represents a grid. The principle of dividing the flow field grid is that the larger the flow velocity area, the more detailed the grid division is required in order to more accurately capture the flow changes in these areas. Therefore, according to the flow velocity information of each area, the size of the grid is set. Generally, areas with higher flow velocities are divided into smaller grids, while areas with lower flow velocities use larger grids.

[0029] After the flow field grid division is completed, the flow field grids are then mapped to the three-dimensional visualization model of the pump head. This process is achieved through spatial coordinate indexing technology to ensure that the physical position of each flow field grid corresponds to the corresponding position in the pump head model. Through this process, multiple pump head model grids are obtained, among which the greater the flow velocity in the flow field division grid, the greater the number of grids.

[0030] Furthermore, in the method provided in the embodiment of the application, based on the pump head visualization model, flow field simulation is performed to obtain an initial simulated flow field, flow field grid division is performed to obtain a plurality of flow field grids, and further includes:

[0031] Based on the pump head visualization model, flow field simulation is performed to obtain an initial simulation flow field, wherein the initial simulation flow field includes initial flow field velocity information of all areas within the pump head visualization model; according to the initial flow field velocity information of all areas, the initial simulation flow field is divided into multiple flow velocity areas, wherein each flow velocity area includes all areas with the same flow field velocity information; according to the flow field velocity information of the multiple flow velocity areas, the multiple flow velocity areas of the initial simulation flow field are divided into flow field grids to obtain multiple flow field grids, wherein according to the flow field velocity information in each flow velocity area, the grid size of the divided flow field grids is set, and flow field grid division is performed, and the size of the flow field velocity information is negatively correlated with the grid size; according to the multiple flow field grids, the area corresponding to the pump head visualization model is indexed and divided to obtain the multiple pump head model grids.

[0032] In an embodiment of the present application, the pump head visualization model previously constructed by three-dimensional modeling software is first imported into the CFD simulation software. Then, based on the actual working conditions of the pump head, the fluid properties are set, and boundary conditions, such as the flow rate at the pump head inlet and the pressure at the outlet, are set to ensure that the simulation environment is as close to the actual working conditions as possible. Then, a suitable numerical solution is selected, such as the finite volume method FVM and a turbulence model (such as the standard k-ε model) to simulate the turbulent state of the fluid. By solving the Navier-Stokes equations and performing simulation calculations, the initial simulation flow field inside the pump head is obtained, and the simulation results include physical quantities such as the flow rate and pressure in each area inside the pump head.

[0033] After obtaining the initial simulation flow field, the flow field inside the entire pump head is divided into multiple flow rate regions according to the flow rate based on the simulation results. This step is achieved through regional segmentation technology, which relies on CFD post-processing tools for automatic segmentation. First, the flow rate data is extracted from the simulation results; then, the flow field is segmented using a regional segmentation algorithm, such as a threshold-based segmentation method, to classify areas with similar flow rates into the same flow rate region, and all positions in each flow rate region have the same or similar flow rate. Subsequently, a unique identifier is assigned to each flow rate region to distinguish between areas with different flow rates, and multiple flow rate regions covering different positions are obtained through the above process.

[0034] After the velocity region division is completed, the flow field grid division is performed according to the velocity information of each velocity region. Specifically, the grid size of each velocity region is determined according to the velocity size in the region, forming a grid division strategy that combines high and low precision. The size of the velocity information is negatively correlated with the grid size, that is, in areas with higher velocity, the grid size will be reduced in order to more accurately capture the details of fluid movement in these areas; on the contrary, in areas with lower velocity, the grid size will increase to reduce the waste of computing resources. For example, in high velocity areas, a small grid of 1 square centimeter may be used, while in low velocity areas, a larger grid of 2 square centimeters may be used. Through the above process, multiple flow field grids are finally generated, and each grid contains the velocity information of its corresponding area.

[0035] Finally, according to multiple flow field grids, the area corresponding to the pump head visualization model is indexed and divided, and the spatial position of each grid is mapped one by one with the corresponding area in the three-dimensional model of the pump head, that is, the coordinate information of the flow field grid is matched with the coordinate system of the pump head model. Through this mapping, it is ensured that each flow field grid can be accurately located in the corresponding area of ​​the pump head model, and multiple pump head model grids corresponding to the flow field grids are obtained.

[0036] Furthermore, in the method provided in the embodiment of the application, based on the flow field velocity information of the multiple flow velocity areas, the multiple flow velocity areas of the initial simulated flow field are divided into flow field grids to obtain multiple flow field grids, and further includes:

[0037] Obtain a set of sample flow field velocity information, set and mark corresponding sample grid size information according to different sample flow field velocity information, and obtain a set of sample grid size information, wherein the size of the sample flow field velocity information is negatively correlated with the sample grid size information; construct a grid division size classifier using the sample flow field velocity information set and the sample grid size information set; input the flow field velocity information of the multiple flow velocity areas into the grid division size classifier to obtain multiple grid size information; perform grid division on the multiple flow velocity areas according to the multiple grid size information to obtain multiple flow field grids.

[0038] In the embodiment of the present application, a sample flow field velocity information set is first acquired from historical data, which are from previous flow field simulation results or actual flow velocity measurement data, and record the fluid flow state of the pump head under different working conditions.

[0039] Next, based on the obtained sample flow field velocity information set, the adaptive grid generation algorithm is used to set and annotate the sample grid size information corresponding to each sample. The algorithm automatically adjusts the grid size according to the changes in the flow velocity area. Specifically, the velocity information of each area is first extracted from the sample flow field; then the density of the grid is set using the existing adaptive grid generation tool to ensure that areas with higher flow velocities use finer grids, such as 1 square centimeter. Areas with lower flow velocities use larger grids, such as 2 square centimeters. Through this negative correlation, the larger the flow velocity, the smaller the grid, and the smaller the flow velocity, the larger the grid, and finally a sample grid size information set is formed.

[0040] Based on the sample flow field velocity information set and the sample grid size information set, a grid partitioning size classifier is trained based on the sample flow field velocity information set and the sample grid size information set by using machine learning techniques such as the K-nearest neighbor algorithm or the random forest algorithm. The classifier automatically learns the relationship between velocity and grid size, and predicts the appropriate grid size based on the input velocity information in actual use. Specifically, the sample velocity information extracted from the historical data and the annotated grid size information are input into the machine learning model as training data; the data is trained using classification algorithms such as KNN or random forest so that the classifier can learn how to select the grid size based on the velocity information; the classifier performance is optimized by methods such as cross-validation to ensure that the prediction results of the classifier are highly consistent with the actual situation. After the classifier training is completed, the flow field velocity information of multiple velocity areas is input into the grid partitioning size classifier. The classifier predicts the most appropriate grid size for each velocity area based on the input velocity information and obtains multiple grid size information.

[0041] Finally, the grid size information generated by the grid size classifier is imported into the grid generation tool, and the adaptive grid generation tool is used to automatically grid the flow field to generate multiple flow field grids, each grid accurately matches its corresponding flow velocity area.

[0042] Step S300: Based on the multiple pump head model grids, specification indexing is performed in the pump head visualization model to obtain multiple pump head grid specification information, and pump head stability analysis is performed in combination with the grid flow field information in the multiple flow field grids to obtain initial flow field stability information.

[0043] In the embodiment of the present application, firstly, based on the existing three-dimensional visualization model of the pump head, the spatial coordinates of each pump head model grid are aligned and matched with the geometric structure of the pump head model, and the specification index is performed. The function of the specification index is to obtain the pump head grid specification information of each pump head grid through the three-dimensional coordinate system, including the thickness of the guide shell, the shape and position of the impeller, etc. The specification information of multiple pump head grids is obtained through the above process.

[0044] Next, the grid flow field information in multiple flow field grids is combined, and the pump head stability analysis is performed through the pump head stability analyzer to obtain the initial flow field stability information.

[0045] Furthermore, in the method provided in the embodiment of the application, specification indexing is performed in the pump head visualization model based on the multiple pump head model grids to obtain multiple pump head grid specification information, and the pump head stability analysis is performed in combination with the grid flow field information in the multiple flow field grids, which also includes:

[0046] Based on the multiple pump head model grids, specification indexing is performed in the pump head visualization model to obtain multiple pump head grid specification information; based on the initial simulation flow field, multiple grid flow field information within the multiple flow field grids is obtained, and each grid flow field information includes grid flow field velocity information; a pump head stability analyzer is trained and obtained; using the pump head stability analyzer, stability analysis of the multiple pump head model grids is performed on the multiple pump head grid specification information and the multiple grid flow field information to obtain multiple grid flow field stability information, and initial flow field stability information is calculated.

[0047] In the embodiment of the present application, the specification index is first performed in the three-dimensional visualization model based on the pump head model grid. Specifically, each grid is matched with the geometric structure in the model through three-dimensional coordinate mapping technology. An automatic mapping tool is used in this step to ensure that the spatial position of each grid is consistent with the geometric structure of the pump head. Finally, relevant specification information is extracted from each grid, such as the thickness, shape, volume, etc. of the guide shell, to generate multiple pump head grid specification information.

[0048] Next, the flow field inside the pump head is simulated and analyzed by computational fluid dynamics simulation technology. The CFD tool is used to simulate the internal fluid movement of the pump head and obtain the physical properties of each grid, such as flow rate, pressure and temperature. Through the simulation settings, the pump head model is imported into the CFD simulation software, the boundary conditions and fluid properties are set, and then the Navier-Stokes equations are solved to simulate the fluid movement inside the pump head and calculate the flow rate, pressure, turbulence and other physical parameters. Finally, the physical properties of each flow field grid, such as flow rate, pressure, etc., are extracted from the simulation results to generate multiple grid flow field information, each of which includes grid flow field flow rate information.

[0049] In order to automatically evaluate the stability of the pump head, a pump head stability analyzer is trained using a machine learning algorithm. The function of this analyzer is to predict the stability of the pump head based on the specification information and flow field information of the grid. The specific training process includes collecting historical data or simulation data, including the specification information, flow field information and corresponding stability results of the pump head grid, selecting a suitable algorithm, such as random forest, and inputting the data into the model for training. The training process is a supervised learning mode, that is, the model learns the mapping relationship between the grid attributes and its stability through the input data. During the training process, the model is trained using historical data, the parameters are optimized and cross-validated to ensure the accuracy and reliability of the model. Finally, a trained pump head stability analyzer is obtained, which can automatically predict the stability of the pump head grid based on the input data.

[0050] The trained stability analyzer is used to automatically analyze the stability of the pump head grid. Specifically, the pump head grid specification information and flow field grid information are input into the analyzer, and the analyzer will predict the stability of each grid during operation by evaluating its geometric features and flow field properties. Finally, the flow field stability information of multiple grids is obtained through the evaluation results of the analyzer, that is, the stability of each grid under the current flow field.

[0051] By analyzing the stability of each grid, summarizing the stability results of all grids, calculating the mean of the flow field stability information of multiple grids, and obtaining the initial flow field stability information.

[0052] Furthermore, in the method provided in the embodiment of the application, training and obtaining a pump head stability analyzer further includes:

[0053] According to the historical operation data of the molten salt pump head, the sample grid flow field information set and the sample pump head grid specification information set are processed and collected, and the wear rate of the pump head grid under different sample grid flow field information and different pump head grid specification information is obtained, and the sample grid flow field stability information set is marked and obtained; the sample grid flow field information set, the sample pump head grid specification information set and the sample grid flow field stability information set are used to train and obtain a pump head stability analyzer through supervised training.

[0054] In an embodiment of the present application, the historical operation data of the molten salt pump head is first collected from a historical database, and the historical operation data includes the fluid dynamic parameters of the pump head under different working conditions and the geometric specification information of the pump head. After formatting and cleaning, the flow field data extracted from the historical database forms a complete sample grid flow field information set. Each grid flow field information includes physical parameters such as flow velocity, pressure, and temperature in the area. At the same time, the pump head geometry and material information extracted from the historical database are processed to generate a sample pump head grid specification information set. These specification information describe the geometric structure and material properties of the pump head, including the guide shell thickness, impeller shape, and material properties.

[0055] After processing the sample information, the next step is to obtain the wear rate of each pump head grid under different flow field and geometric conditions. The wear rate is determined based on historical experimental data, which records the wear of the guide shell and impeller under different flow rates and pressure conditions. The obtained wear rate is matched one by one with the sample flow field information and geometric specification information to generate a sample grid flow field stability information set containing the wear rate, where the value obtained by subtracting the wear rate of the pump head in the grid from 1 is used as the sample grid flow field stability information.

[0056] To achieve automated stability analysis, a pump head stability analyzer is trained through supervised learning. Specifically, the sample grid flow field information set, the sample pump head grid specification information set, and the sample grid flow field stability information set with wear rate annotated are first integrated to form a training data set. Next, a machine learning algorithm such as random forest or support vector machine is used to supervise the training data. Through multiple iterations, the model can learn the relationship between wear rate and stability under different flow fields and geometric conditions. Finally, cross-validation is used to optimize the model performance to ensure that the model can accurately predict the stability of the pump head under actual working conditions. Through this process, the training of the pump head stability analyzer is completed.

[0057] After training is completed, the pump head stability analyzer can automatically predict the stability of the pump head based on the new pump head flow field information and geometric specification information.

[0058] Step S400: Adjust the initial specification information of the impeller and guide shell, perform visual three-dimensional modeling and flow field simulation, optimize the pump head flow field, and obtain optimal specification information and optimal simulated flow field, wherein optimization judgment is performed based on the initial flow field stability information and the rationality of the distribution of multiple flow field grids divided in the optimization.

[0059] In the embodiment of the present application, the geometric parameters of the impeller and the guide shell are first adjusted. For example, the number of blades, angle and radius of curvature of the impeller may be optimized to affect the direction and speed of the fluid flow, thereby improving the flow field distribution. The adjusted geometric structure is then visualized using a three-dimensional modeling tool. After adjusting the specification information of the impeller and the guide shell, the new model is simulated using computational fluid dynamics simulation technology to generate detailed data on parameters such as flow velocity, pressure, and turbulence.

[0060] After the simulation results come out, a comparative analysis is performed based on the stability information of the initial flow field to determine whether the adjustment is effective. By comparing the new simulation results with the initial data, it is evaluated whether the new pump head design reduces these instabilities. If the new design can improve the stability of the flow field, it means that the adjustment is successful.

[0061] After multiple simulations and adjustment iterations, the optimal specification information of the pump head and the optimal simulation flow field are finally obtained.

[0062] Furthermore, in the method provided in the embodiment of the application, the initial specification information of the impeller and the guide shell is adjusted, and visual three-dimensional modeling and flow field simulation are performed to optimize the flow field of the pump head, and the method also includes:

[0063] The impeller specification information space and the guide shell specification information space of the impeller and the guide shell are obtained, and the initial specification information of the impeller and the guide shell is randomly adjusted in the impeller specification information space and the guide shell specification information space to obtain the first impeller specification information and the first guide shell specification information; according to the first impeller specification information and the first guide shell specification information, visual three-dimensional modeling and flow field simulation are performed to obtain the first simulated flow field, and a plurality of first flow field grids are obtained by division, and the first flow field stability information is obtained by analysis; according to the plurality of first flow field grids, the first distribution rationality information is obtained by analysis; according to the ratio of the first flow field stability information to the initial flow field stability information, the first flow field fitness is obtained by weighted calculation in combination with the first distribution rationality information; the initial specification information of the impeller and the guide shell is continued to be adjusted and optimized until convergence, and the impeller specification information and the guide shell specification information with the maximum flow field fitness are output to obtain the optimal specification information and the optimal simulated flow field.

[0064] In the embodiment of the present application, the technical experts first define the impeller specification information space and the guide shell specification information space. These specification information spaces represent the value ranges of various geometric parameters of the impeller and the guide shell, such as the number, angle, and curvature of the impeller blades, and the thickness and shape of the guide shell.

[0065] After defining the specification information space, the initial specification information of the impeller and the guide shell needs to be randomly adjusted. This means that in the impeller specification information space and the guide shell specification information space, different impeller and guide shell geometric parameter combinations are selected through random algorithms, such as the Monte Carlo method, to generate new specification combinations. Specifically, a new geometric parameter combination is generated in the impeller specification information space and the guide shell specification information space using a random algorithm to obtain the first impeller specification information and the first guide shell specification information.

[0066] After obtaining the new specification information, the 3D modeling software is used to perform visual 3D modeling of the adjusted impeller and guide shell to generate a new 3D model of the pump head. Next, the flow field of the new model is simulated by computational fluid dynamics simulation to generate a new first simulated flow field. At the same time, the flow field grid is divided in the same way as above to obtain multiple first flow field grids, and the new model is divided according to the first flow field grid to obtain multiple first pump head model grids. Then, the first pump head grid specification information and the first grid flow field information are obtained by the same method as above, and input into the pump head stability analyzer for analysis to obtain the first flow field stability information.

[0067] After rasterizing the flow field and analyzing its stability, it is necessary to further evaluate the rationality of the flow field distribution and generate first distribution rationality information based on the analysis results.

[0068] Next, a weighted calculation is performed based on the ratio of the first flow field stability information to the initial flow field stability information in combination with the first distribution rationality information, that is, the first flow field fitness is obtained by multiplying the ratio of the first flow field stability information to the initial flow field stability information by a predetermined weight and adding the first distribution rationality information multiplied by a predetermined weight. The weight of the ratio of the first flow field stability information to the initial flow field stability information and the weight of the first distribution rationality information are predetermined, and the sum of these two weights is 1.

[0069] Then continue to adjust the specifications of the impeller and guide shell, and gradually adjust the design through multiple simulations and optimization iterations until the fitness is no longer significantly improved and convergence is reached. Finally, output the specifications of the impeller and guide shell with the highest flow field fitness, and obtain the optimal specifications of the pump head and the optimal simulation flow field.

[0070] Furthermore, in the method provided in the embodiment of the application, analyzing and obtaining the first distribution rationality information according to the plurality of first flow field grids further includes:

[0071] Acquire a standard simulation flow field, divide it into multiple standard flow field grids, and construct a standard flow field grid distribution, wherein the standard flow field grid distribution includes the grid sizes and distribution positions of the multiple standard flow field grids; construct a first flow field grid distribution based on the multiple first flow field grids; analyze the similarity between the standard flow field grid distribution and the first flow field grid distribution to obtain the first distribution rationality information.

[0072] In an embodiment of the present application, a reference flow field is first obtained as a reference, which is the standard simulation flow field. The standard flow field is obtained by computational fluid dynamics simulation technology and is generated based on the existing best design or industry standard. CFD simulation simulates the flow behavior of the fluid inside the pump head based on the Navier-Stokes equation. Specifically, the standard design geometry model of the pump head is first imported into the CFD simulation software, the fluid boundary conditions are set, and an appropriate turbulence model is selected. Subsequently, the simulation is run and the Navier-Stokes equation is solved to obtain the standard simulation flow field inside the pump head.

[0073] After completing the simulation, the next step is to process the standard flow field and divide it into multiple standard flow field grids using the same method as above. After dividing the standard flow field grids, the next step is to build a standard flow field grid distribution model. This model records the size, position, and fluid dynamics characteristics of the standard flow field grid. Specifically, the flow velocity and pressure information are extracted from the CFD simulation data through the simulation post-processing tool, and these data are combined with the spatial position to generate a spatial distribution map of the grid, forming a standard flow field grid distribution, where the standard flow field grid distribution includes the grid size and distribution position of multiple standard flow field grids.

[0074] For new designs, the first flow field grid distribution is constructed based on the first flow field grid that has been obtained. This process is similar to the construction of the standard flow field grid distribution. The physical and geometric information of each grid is extracted through the simulation post-processing tool, and this information is combined to construct a complete flow field grid distribution. Specifically, the simulation tool is used to extract the physical parameters such as flow velocity and pressure of each grid, and the position and size of the grid in three-dimensional space are recorded. The extracted physical parameters are combined with the geometric information of the grid to generate a complete first flow field grid distribution.

[0075] Next, a similarity analysis is performed on the standard flow field grid distribution and the first flow field grid distribution using mathematical statistical methods. Similarity analysis uses a variety of methods, such as Euclidean distance, cosine similarity, etc., to measure the differences between the two flow fields in physical parameters such as flow velocity and pressure. Specifically, the standard flow field grid distribution and the first flow field grid distribution are first spatially aligned to ensure that they are compared in the same coordinate system. Then, by comparing the differences in physical parameters of the two flow field grids, the similarity between the two is calculated using a similarity formula, such as Euclidean distance. Finally, the calculated similarity is used as the rationality information of the first distribution.

[0076] Optionally, the standard flow field grid distribution and the first flow field grid distribution are used as images for similarity analysis. Specifically, the physical parameters such as flow velocity and pressure in the standard flow field grid distribution and the first flow field grid distribution are converted into two-dimensional or three-dimensional images through simulation post-processing tools. These images are visualization results of the flow field, showing the velocity distribution and pressure changes of the fluid in different areas. In order to ensure the accuracy of the image similarity analysis, the generated images are preprocessed, such as normalization, filtering denoising or edge detection, so that the image can more clearly show the characteristics of the flow field. Then, an image similarity algorithm, such as the structural similarity index SSIM, the peak signal-to-noise ratio PSNR, etc., is used to compare and analyze the standard flow field grid distribution image and the first flow field grid distribution image. The similarity value between the two images is calculated by the image similarity algorithm, and the calculated similarity value is used as the first distribution rationality information.

[0077] Step S500: According to the optimal specification information and the optimal simulated flow field, the design of the molten salt pump head is modified and the flow field is verified.

[0078] In the embodiment of the present application, geometric design corrections are performed on components such as the impeller and the guide shell of the molten salt pump head according to the optimal specification information and the optimal simulation flow field.

[0079] After the design modification is completed, the flow field of the modified pump head design is verified. Through CFD simulation technology, the operation of the modified pump head in the actual fluid environment is simulated, and by comparing with the optimal simulated flow field, it is confirmed whether the modified design has achieved the expected fluid performance. Specifically, it is verified whether the flow field of the new design is consistent with the optimal simulated flow field, whether the flow path is optimized, energy loss is reduced, and the overall efficiency of the pump head is improved.

[0080] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0081] The present application collects the initial specification information of the impeller and the guide shell in the pump head of the current molten salt pump, and constructs a pump head visualization model based on visual three-dimensional modeling; based on the pump head visualization model, flow field simulation is performed to obtain the initial simulated flow field, flow field grid division is performed to obtain multiple flow field grids, and the pump head visualization model is divided according to the flow field grids to obtain multiple pump head model grids, wherein the greater the flow velocity in the flow field division grid, the greater the number of grids; specification indexing is performed in the pump head visualization model based on the multiple pump head model grids to obtain multiple pump head grids. The grid specification information is combined with the grid flow field information in multiple flow field grids to perform pump head stability analysis and obtain initial flow field stability information; the initial specification information of the impeller and the guide shell is adjusted, and visual three-dimensional modeling and flow field simulation are performed to optimize the pump head flow field and obtain the optimal specification information and the optimal simulated flow field, wherein, based on the initial flow field stability information and combined with the rationality of the distribution of multiple flow field grids divided in the optimization, optimization judgment is performed; according to the optimal specification information and the optimal simulated flow field, the design correction and flow field verification of the molten salt pump head are carried out. The present invention solves the technical problem of low design accuracy in the prior art when designing pump heads. By collecting the initial specification information of the impeller and the guide shell in the pump head of the molten salt pump, and utilizing visual three-dimensional modeling and flow field simulation technology, a pump head visualization model is constructed and the flow field grid is divided. The flow field information and stability of the pump head model grid are analyzed, the specification information is adjusted, the flow field distribution of the pump head is optimized, the optimal specifications and simulated flow field are obtained, and finally the optimization results are combined to perform design correction and flow field verification, so as to achieve the technical effect of improving the fluid performance and stability of the pump head under actual working conditions.

[0082] Embodiment 2 is based on the same inventive concept as the pump head flow field optimization correction method based on visual analysis in the previous embodiment. Figure 2 As shown, the present application provides a pump head flow field optimization and correction system based on visual analysis, and the system and method embodiments in the present application are based on the same inventive concept. The system includes:

[0083] A visualization model construction module 11, the visualization model construction module 11 collects the initial specification information of the impeller and the guide shell in the pump head of the current molten salt pump, and constructs a pump head visualization model based on visualization three-dimensional modeling; a flow field simulation module 12, the flow field simulation module 12 performs flow field simulation based on the pump head visualization model, obtains an initial simulation flow field, performs flow field grid division, obtains a plurality of flow field grids, and divides the pump head visualization model according to the flow field grids to obtain a plurality of pump head model grids, wherein the greater the flow velocity in the flow field division grid, the greater the number of grids; a specification indexing module 13, the specification indexing module 13 performs specification indexing in the pump head visualization model based on the plurality of pump head model grids A grid index is used to obtain grid specification information of multiple pump heads, and the grid flow field information in the multiple flow field grids is combined to perform pump head stability analysis to obtain initial flow field stability information; an optimization judgment module 14, wherein the optimization judgment module 14 adjusts the initial specification information of the impeller and the guide shell, and performs visual three-dimensional modeling and flow field simulation to optimize the pump head flow field to obtain the optimal specification information and the optimal simulated flow field, wherein, based on the initial flow field stability information, the optimization judgment is performed in combination with the rationality of the distribution of the multiple flow field grids divided in the optimization; a correction verification module 15, wherein the correction verification module 15 performs design correction and flow field verification of the molten salt pump head according to the optimal specification information and the optimal simulated flow field.

[0084] Furthermore, the system is also used to implement the following functions:

[0085] The initial specification information of the impeller and the initial specification information of the guide shell in the pump head of the current molten salt pump are collected; and a pump head visualization model is constructed based on the visualization 3D modeling according to the initial specification information of the impeller and the initial specification information of the guide shell.

[0086] Furthermore, the system is also used to implement the following functions:

[0087] Based on the pump head visualization model, flow field simulation is performed to obtain an initial simulation flow field, wherein the initial simulation flow field includes initial flow field velocity information of all areas within the pump head visualization model; according to the initial flow field velocity information of all areas, the initial simulation flow field is divided into multiple flow velocity areas, wherein each flow velocity area includes all areas with the same flow field velocity information; according to the flow field velocity information of the multiple flow velocity areas, the multiple flow velocity areas of the initial simulation flow field are divided into flow field grids to obtain multiple flow field grids, wherein according to the flow field velocity information in each flow velocity area, the grid size of the divided flow field grids is set, and flow field grid division is performed, and the size of the flow field velocity information is negatively correlated with the grid size; according to the multiple flow field grids, the area corresponding to the pump head visualization model is indexed and divided to obtain the multiple pump head model grids.

[0088] Furthermore, the system is also used to implement the following functions:

[0089] Obtain a set of sample flow field velocity information, set and mark corresponding sample grid size information according to different sample flow field velocity information, and obtain a set of sample grid size information, wherein the size of the sample flow field velocity information is negatively correlated with the sample grid size information; construct a grid division size classifier using the sample flow field velocity information set and the sample grid size information set; input the flow field velocity information of the multiple flow velocity areas into the grid division size classifier to obtain multiple grid size information; perform grid division on the multiple flow velocity areas according to the multiple grid size information to obtain multiple flow field grids.

[0090] Furthermore, the system is also used to implement the following functions:

[0091] Based on the multiple pump head model grids, specification indexing is performed in the pump head visualization model to obtain multiple pump head grid specification information; based on the initial simulation flow field, multiple grid flow field information within the multiple flow field grids is obtained, and each grid flow field information includes grid flow field velocity information; a pump head stability analyzer is trained and obtained; using the pump head stability analyzer, stability analysis of the multiple pump head model grids is performed on the multiple pump head grid specification information and the multiple grid flow field information to obtain multiple grid flow field stability information, and initial flow field stability information is calculated.

[0092] Furthermore, the system is also used to implement the following functions:

[0093] According to the historical operation data of the molten salt pump head, the sample grid flow field information set and the sample pump head grid specification information set are processed and collected, and the wear rate of the pump head grid under different sample grid flow field information and different pump head grid specification information is obtained, and the sample grid flow field stability information set is marked and obtained; the sample grid flow field information set, the sample pump head grid specification information set and the sample grid flow field stability information set are used to train and obtain a pump head stability analyzer through supervised training.

[0094] Furthermore, the system is also used to implement the following functions:

[0095] The impeller specification information space and the guide shell specification information space of the impeller and the guide shell are obtained, and the initial specification information of the impeller and the guide shell is randomly adjusted in the impeller specification information space and the guide shell specification information space to obtain the first impeller specification information and the first guide shell specification information; according to the first impeller specification information and the first guide shell specification information, visual three-dimensional modeling and flow field simulation are performed to obtain the first simulated flow field, and a plurality of first flow field grids are obtained by division, and the first flow field stability information is obtained by analysis; according to the plurality of first flow field grids, the first distribution rationality information is obtained by analysis; according to the ratio of the first flow field stability information to the initial flow field stability information, the first flow field fitness is obtained by weighted calculation in combination with the first distribution rationality information; the initial specification information of the impeller and the guide shell is continued to be adjusted and optimized until convergence, and the impeller specification information and the guide shell specification information with the maximum flow field fitness are output to obtain the optimal specification information and the optimal simulated flow field.

[0096] Furthermore, the system is also used to implement the following functions:

[0097] Acquire a standard simulation flow field, divide it into multiple standard flow field grids, and construct a standard flow field grid distribution, wherein the standard flow field grid distribution includes the grid sizes and distribution positions of the multiple standard flow field grids; construct a first flow field grid distribution based on the multiple first flow field grids; analyze the similarity between the standard flow field grid distribution and the first flow field grid distribution to obtain the first distribution rationality information.

[0098] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0099] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0100] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A pump head flow field optimization correction method based on visual analysis, characterized in that: The method comprises: Collect the initial specification information of the impeller and guide shell in the current molten salt pump head, and build a pump head visualization model based on visual 3D modeling; Based on the pump head visualization model, flow field simulation is performed to obtain an initial simulation flow field, flow field grid division is performed to obtain a plurality of flow field grids, and the pump head visualization model is divided according to the flow field grids to obtain a plurality of pump head model grids, wherein the area with a greater flow velocity in the flow field division grid has a greater number of grids; Based on the specification indexing of the multiple pump head model grids in the pump head visualization model, multiple pump head grid specification information is obtained, and the pump head stability analysis is performed in combination with the grid flow field information in the multiple flow field grids to obtain initial flow field stability information; The initial specification information of the impeller and the guide shell is adjusted, and visual three-dimensional modeling and flow field simulation are performed to optimize the pump head flow field to obtain the optimal specification information and the optimal simulation flow field, wherein the optimization judgment is performed based on the initial flow field stability information and the distribution rationality of the multiple flow field grids divided in the optimization; According to the optimal specification information and the optimal simulated flow field, the design correction and flow field verification of the molten salt pump head are performed; Wherein, based on the specification indexing of the multiple pump head model grids in the pump head visualization model, multiple pump head grid specification information is obtained, and the pump head stability analysis is performed in combination with the grid flow field information in the multiple flow field grids, including: Performing specification indexing in the pump head visualization model based on the multiple pump head model grids to obtain multiple pump head grid specification information; Based on the initial simulated flow field, a plurality of grid flow field information in the plurality of flow field grids is obtained, each grid flow field information including grid flow field flow velocity information; Training to obtain pump head stability analyzer; Using the pump head stability analyzer, the stability analysis of the multiple pump head model grids is performed on the multiple pump head grid specification information and the multiple grid flow field information, the multiple grid flow field stability information is obtained, and the initial flow field stability information is calculated; The initial specification information of the impeller and the guide shell is adjusted, and visual three-dimensional modeling and flow field simulation are performed to optimize the flow field of the pump head, including: Obtaining an impeller specification information space and a guide shell specification information space of the impeller and the guide shell, and randomly adjusting the initial specification information of the impeller and the guide shell in the impeller specification information space and the guide shell specification information space to obtain first impeller specification information and first guide shell specification information; Performing visual three-dimensional modeling and flow field simulation according to the first impeller specification information and the first guide shell specification information to obtain a first simulated flow field, dividing the flow field to obtain a plurality of first flow field grids, and analyzing to obtain first flow field stability information; Analyzing and obtaining first distribution rationality information according to the plurality of first flow field grids; According to the ratio of the first flow field stability information to the initial flow field stability information and the first distribution rationality information, a weighted calculation is performed to obtain a first flow field fitness; Continue to adjust and optimize the initial specification information of the impeller and guide shell until convergence, output the impeller specification information and guide shell specification information with the maximum flow field fitness, and obtain the optimal specification information and the optimal simulation flow field.

2. The pump head flow field optimization correction method based on visual analysis according to claim 1 is characterized in that: Collect the initial specification information of the impeller and guide shell in the current molten salt pump head, and build a pump head visualization model based on visual 3D modeling, including: Collect the initial specification information of the impeller and the guide shell in the pump head of the current molten salt pump; According to the initial specification information of the impeller and the initial specification information of the guide casing, a pump head visualization model is constructed based on visualized three-dimensional modeling.

3. The pump head flow field optimization correction method based on visual analysis according to claim 1 is characterized in that: Based on the pump head visualization model, flow field simulation is performed to obtain an initial simulated flow field, and flow field grid division is performed to obtain multiple flow field grids, including: Based on the pump head visualization model, flow field simulation is performed to obtain an initial simulation flow field, wherein the initial simulation flow field includes initial flow field velocity information of all areas within the pump head visualization model; According to the initial flow field velocity information of all the regions, the initial simulated flow field is divided into a plurality of flow velocity regions, wherein each flow velocity region includes all regions having the same flow field velocity information; According to the flow field velocity information of the multiple flow velocity areas, the multiple flow velocity areas of the initial simulation flow field are divided into flow field grids to obtain multiple flow field grids, wherein the grid size of the flow field grid is set according to the flow field velocity information in each flow velocity area, and the flow field grid division is performed, and the size of the flow field velocity information is negatively correlated with the grid size; According to the multiple flow field grids, the area corresponding to the pump head visualization model is indexed and divided to obtain the multiple pump head model grids.

4. The pump head flow field optimization correction method based on visual analysis according to claim 3 is characterized in that: According to the flow field velocity information of the plurality of flow velocity areas, the plurality of flow velocity areas of the initial simulation flow field are divided into flow field grids to obtain a plurality of flow field grids, including: Obtaining a sample flow field velocity information set, setting and marking corresponding sample grid size information according to different sample flow field velocity information, and obtaining a sample grid size information set, wherein the size of the sample flow field velocity information is negatively correlated with the sample grid size information; Using the sample flow field velocity information set and the sample grid size information set, constructing a grid partition size classifier; Inputting the flow field velocity information of the plurality of flow velocity areas into the grid division size classifier to obtain a plurality of grid size information; According to the plurality of grid size information, the plurality of flow velocity regions are grid-divided to obtain a plurality of flow field grids.

5. The pump head flow field optimization correction method based on visual analysis according to claim 1 is characterized in that: Training to obtain the Pump Head Stability Analyzer, including: According to the historical operation data of the molten salt pump head, the sample grid flow field information set and the sample pump head grid specification information set are processed and collected, and the wear rate of the pump head grid under different sample grid flow field information and different pump head grid specification information is obtained, and the sample grid flow field stability information set is marked and obtained; The sample grid flow field information set, the sample pump head grid specification information set and the sample grid flow field stability information set are used to train and obtain a pump head stability analyzer through supervised training.

6. The pump head flow field optimization correction method based on visual analysis according to claim 1 is characterized in that: Analyzing and obtaining first distribution rationality information according to the plurality of first flow field grids includes: Acquire a standard simulation flow field, divide it into multiple standard flow field grids, and construct a standard flow field grid distribution, wherein the standard flow field grid distribution includes grid sizes and distribution positions of the multiple standard flow field grids; constructing a first flow field grid distribution according to the plurality of first flow field grids; The similarity between the standard flow field grid distribution and the first flow field grid distribution is analyzed to obtain first distribution rationality information.

7. Pump head flow field optimization correction system based on visual analysis, characterized in that: The system is used to execute the pump head flow field optimization correction method based on visual analysis according to any one of claims 1 to 6, and the system comprises: A visualization model building module, wherein the visualization model building module collects initial specification information of an impeller and a guide shell in a pump head of a current molten salt pump, and builds a visualization model of the pump head based on visualization three-dimensional modeling; A flow field simulation module, wherein the flow field simulation module performs flow field simulation based on the pump head visualization model to obtain an initial simulation flow field, performs flow field grid division to obtain a plurality of flow field grids, and divides the pump head visualization model according to the flow field grids to obtain a plurality of pump head model grids, wherein the greater the flow velocity in the flow field division grid, the greater the number of grids; A specification indexing module, wherein the specification indexing module performs specification indexing in the pump head visualization model based on the multiple pump head model grids to obtain multiple pump head grid specification information, and performs pump head stability analysis in combination with grid flow field information in the multiple flow field grids to obtain initial flow field stability information; An optimization judgment module, which adjusts the initial specification information of the impeller and the guide shell, performs visual three-dimensional modeling and flow field simulation, optimizes the pump head flow field, and obtains the optimal specification information and the optimal simulation flow field, wherein the optimization judgment is performed based on the initial flow field stability information and the distribution rationality of the multiple flow field grids divided in the optimization; A correction and verification module is used to perform design correction and flow field verification on the pump head of the molten salt pump according to the optimal specification information and the optimal simulated flow field.