Wear-resistant low-resistance impeller-pump body fit clearance optimization method and system

Abstract

The invention relates to the technical field of impeller-pump body clearance optimization, in particular to a wear-resistant low-resistance impeller-pump body fit clearance optimization method and system.The method comprises the steps that the membership degree weight of each cluster at each moment is determined by analyzing the membership degree between all flow field two-tuples and each cluster, and the membership degree weight of each cluster is calculated; determining a flow state risk index by combining the difference between the clustering centers of the high-loss low-turbulence clusters and the clustering centers of the medium-loss medium-turbulence clusters and the low-loss high-turbulence clusters; obtaining the total number of optimal solutions at each moment by adopting a particle swarm optimization algorithm; and based on the theoretical life prediction value and the optimal solution total number of the impeller-pump body at each moment, the adjustable diversity of the impeller-pump body at each moment is determined, so that the grid density of the impeller-pump body gap is adjusted. According to the method, the problem that a structured grid is difficult to adapt to complex geometric deformation is solved, and the prediction precision of impeller-pump body gap particle induced local wear and gap flow state evolution is improved.

Description

Technical Field

[0001] The present application relates to the technical field of impeller-pump body clearance optimization, and specifically to a wear-resistant low-resistance impeller-pump body clearance optimization method and system. Background Art

[0002] With the urgent demand for high efficiency, energy saving, and long life in industrial equipment, traditional impeller pumps often face a dilemma in the design of the clearance between the impeller and the pump body during operation. A clearance that is too large can lead to fluid leakage and reduced efficiency; a clearance that is too small can easily cause friction, wear, and increased energy consumption. This is especially true when conveying media containing particles, where wear and failure are more prominent. With advances in fluid dynamics theory, materials science, and precision manufacturing technology, optimizing the clearance structure can reduce turbulent losses within the pump, improve overall energy efficiency, and achieve the requirements of high efficiency and energy saving equipment. Wear-resistant coatings combined with reasonable clearances can extend the life of the impeller and reduce downtime and maintenance costs.

[0003] Existing technologies mainly alleviate the contradiction of grid division through multi-scale grid division strategies. For example, boundary layer grids are used in the gap area to improve the near-wall flow resolution, and unstructured grids are used in the peripheral area to reduce the amount of calculation. However, the discontinuous transition at the junction of multi-scale grids may cause flow field distortion, and structured grids are difficult to adapt to complex geometric deformations, making it difficult to accurately predict the local wear induced by particles in the impeller-pump body gap and the evolution of the gap flow state. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a wear-resistant low-resistance impeller-pump body fit clearance optimization method and system. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body, the method comprising the following steps:

[0006] Real-time acquisition of all flow field data, theoretical life data, and geometric clearance dimensions of the impeller-pump body. All flow field data include: turbulent kinetic energy, dissipation rate, wall shear stress, and particle collision frequency.

[0007] A principal component analysis algorithm is used to analyze the dominant role of all types of flow field data at each moment and within a preset time period before it, and the impeller-pump body flow field binary group at each moment is determined; the flow field binary groups at each moment and all moments within a preset time period before it are clustered, wherein all clusters are respectively high-loss and low-turbulence clusters, medium-loss and medium-turbulence clusters, and low-loss and high-turbulence clusters; by analyzing the membership between all flow field binary groups and each cluster, the membership weight of each cluster at each moment is determined, and combined with the difference in cluster centers between the high-loss and low-turbulence cluster, the medium-loss and medium-turbulence cluster, and the low-loss and high-turbulence cluster, the flow state risk index of the impeller-pump body at each moment is determined;

[0008] Based on all flow risk indices, all particle collision rates, and all theoretical life data within a preset time period before each moment, a support vector regression algorithm is used to obtain the theoretical life prediction value of the impeller-pump body at each moment; based on all flow risk indices, all geometric gap dimensions, and all theoretical life data within a preset time period before each moment, a particle swarm optimization algorithm is used to obtain the total number of optimal solutions at each moment; based on the theoretical life prediction value and the total number of optimal solutions of the impeller-pump body at each moment, the adjustable diversity of the impeller-pump body at each moment is determined;

[0009] The grid density of the impeller-pump body clearance is adjusted based on the average distribution of the adjustable diversity at all moments in a preset time period before the current moment.

[0010] Preferably, the method for determining the impeller-pump body flow field binary at each moment is:

[0011] All kinds of flow field data at each moment and within the preset time period before it are used as the input of the principal component analysis algorithm, and the principal component scores PC1 and PC2 at each moment and all moments before it are output. The tuple composed of the principal component scores PC1 and PC2 at each moment is recorded as the impeller-pump body flow field tuple at each moment.

[0012] Preferably, the method for obtaining the high-loss low-turbulence cluster, the medium-loss medium-turbulence cluster, and the low-loss high-turbulence cluster is:

[0013] The mean of the particle collision frequency of all flow field binary groups in each cluster at the corresponding time is calculated, and the clusters corresponding to each mean in the results of descending order are recorded as high-loss low-turbulence cluster, medium-loss medium-turbulence cluster and low-loss high-turbulence cluster.

[0014] Preferably, the membership weight of each cluster at each moment is the result of taking the average of the membership between the flow field binary and each cluster at each moment and all previous moments.

[0015] Preferably, the method for determining the impeller-pump flow risk index at each moment is:

[0016] Calculate the distances from the cluster centers of the high-loss, low-turbulence cluster, the medium-loss, medium-turbulence cluster, and the low-loss, high-turbulence cluster to the cluster center of the low-loss, high-turbulence cluster at each moment;

[0017] The multiplication result of the membership weight of each cluster and the corresponding distance at each moment is calculated, and the mean of the multiplication results of all clusters at each moment is used as the impeller-pump body flow risk index at each moment.

[0018] Preferably, the step of obtaining the theoretical life prediction value of the impeller-pump body at each moment includes:

[0019] All flow risk indices and all particle collision rates within the preset time before each moment are used as independent variables in the support vector regression algorithm, and all theoretical life data within the preset time before each moment are used as dependent variables in the support vector regression algorithm, and the theoretical life prediction value of the impeller-pump body at each moment is output.

[0020] Preferably, the particle swarm optimization algorithm is used to obtain the total number of optimal solutions at each moment, including:

[0021] All flow risk indices, all geometric gap sizes, and all theoretical life data within a preset time period before each moment are used as input to the particle swarm optimization algorithm, and all optimal solutions are output.

[0022] Preferably, the adjustable diversity of the impeller-pump body at each moment is the product of the theoretical life prediction value of the impeller-pump body at each moment and the total number of optimal solutions.

[0023] Preferably, the adjusting the grid density of the impeller-pump body clearance at the current moment includes:

[0024] The grid density correction value ρ of the impeller-pump body clearance at the current moment ′ The expression is: Where, ρ base represents the initial mesh density obtained by the ANSYS Meshing tool; It represents the average value of the adjustable diversity of the impeller-pump body at the current moment and all moments in the previous preset period; exp() represents an exponential function with a natural constant as the base.

[0025] In the second aspect, an embodiment of the present application also provides a wear-resistant low-resistance impeller-pump body fitting clearance optimization system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0026] This application has at least the following beneficial effects:

[0027] In order to solve the problems of high dimensionality, multicollinearity and dimensional difference in gap flow field data, this application adopts principal component analysis algorithm to reduce the dimension, combines clustering algorithm to divide the flow field into three typical working modes, quantifies the coupling hazard intensity of turbulent loss and wear, solves the problem that traditional methods are difficult to dynamically associate flow field characteristics with equipment performance degradation, and provides a quantitative basis for optimizing target weight allocation; further, in order to solve the imbalance problem between efficiency and life in single-objective optimization, this application adopts support vector regression to predict wear life, combines multi-objective particle swarm optimization algorithm to generate Pareto front optimal solution set, dynamically evaluates the flexibility and wear resistance of the design space, eliminates the limitation of local optimal solution dominance in traditional optimization, and ensures a global dynamic balance between minimizing turbulent loss and maximizing wear resistance; further, this application designs a segmented grid adjustment strategy based on real-time feedback of adjustable diversity, dynamically adjusts the grid density in the gap area, so that when the adjustable diversity decreases, the grid is encrypted to capture subtle flow field changes, breaking through the rigid constraints of traditional fixed grids, and improving the prediction accuracy of local wear induced by particles in the impeller-pump body gap and gap flow evolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0029] Figure 1 A flowchart of a method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body provided in one embodiment of the present application;

[0030] Figure 2 A schematic diagram of an adjustable diversity extraction process provided in one embodiment of the present application. DETAILED DESCRIPTION

[0031] To further illustrate the technical means and effectiveness of this application to achieve the intended invention objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effectiveness of a wear-resistant, low-resistance impeller-pump body clearance optimization method and system proposed in this application. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0032] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0033] The following describes in detail a method and system for optimizing the clearance between a wear-resistant and low-resistance impeller and a pump body provided by the present application with reference to the accompanying drawings.

[0034] See also Figure 1 , which shows a flowchart of a method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body provided by one embodiment of the present application, the method comprising the following steps:

[0035] Step S1: Acquire all flow field data, theoretical life data and geometric gap size of the impeller-pump body in real time, wherein all flow field data include: turbulent kinetic energy, dissipation rate, wall shear stress, and particle collision frequency.

[0036] First, a detailed modeling of the impeller-pump body gap flow field was performed through fluid dynamics simulation, with a focus on optimizing the gap structure to achieve high efficiency and energy saving. Specifically, ANSYS Fluent was used as the fluid dynamics (CFD) modeling software. Its built-in multi-physics solver and discrete phase model (DPM) can effectively simulate the complex flow and wear behavior of particle-containing media, helping to accurately simulate the impact of particle-containing media flow on high efficiency and energy saving performance. The specific modeling process is as follows:

[0037] A 3D geometric model of the impeller and pump body was constructed based on SolidWorks, with a focus on refining the geometric features of the gap area. The geometry cleanup tool of ANSYS Workbench was used to eliminate small chamfers and interference surfaces. The parametric modeling tool of ANSYS Workbench was used to record the geometric gap size between the impeller and the pump body in real time. ANSYS Meshing was then used for unstructured meshing. A boundary layer mesh generation strategy was adopted for the gap area between the impeller and the pump body to ensure precise capture of the near-wall flow. A sparse hexahedral mesh was used in the peripheral area to reduce the amount of calculation. The initial mesh density ρ at the gap between the impeller and the pump body was output using the ANSYS Meshing tool. base The physical model selected was the SSTk-ω turbulence model to balance the accuracy of near-wall flow and free shear flow. The DPM model within the Euler-Lagrangian framework was used to simulate particle motion. The particle size distribution was set according to the Rosin-Rammler model. The inlet boundary was defined as a velocity inlet, the outlet as a static pressure of 0 Pa, and a no-slip boundary condition was used on the wall to achieve three-dimensional geometric modeling of the impeller and pump body. The construction of these three-dimensional geometric models is well known to those skilled in the art and will not be elaborated on here.

[0038] During the simulation, the flow field parameters are iteratively calculated using a second-order implicit format using a transient solver, and all flow field data are monitored in real time. All flow field data includes turbulent kinetic energy, dissipation rate, wall shear stress, and particle collision frequency. In this embodiment, the data acquisition frequency is uniformly set to 10Hz through the monitoring point (Probe) function of ANSYS Fluent, and the real-time acquisition of all parameters is triggered synchronously based on the simulation time to ensure the turbulent kinetic energy, dissipation rate, wall shear stress, and particle collision frequency data at each timestamp. The theoretical life data at each acquisition moment is obtained through the Archard wear model to quantify the key parameters of high-efficiency energy-saving performance. Among them, the acquisition and processing of the above four flow field parameter data are technologies well known to people in this field and will not be elaborated here.

[0039] Furthermore, in order to eliminate the influence of data dimension, various flow field data, theoretical life data and geometric gap dimensions are normalized respectively. In this embodiment, the z-score normalization method is used to normalize the collected data. In actual application, as other implementation methods, the implementer can also set them according to the specific situation. This embodiment does not impose any special restrictions.

[0040] Among them, the z-score normalization method is a well-known technology, and the specific process of using it to normalize the data will not be described in detail.

[0041] It should be noted that all data in this embodiment are collected synchronously in real time.

[0042] Step S2: Use the principal component analysis algorithm to analyze the dominant role of all types of flow field data at each moment and within the preset time period before it, and determine the flow field binary of the impeller-pump body at each moment; cluster the flow field binary at each moment and all moments within the preset time period before it, wherein all clusters are high-loss and low-turbulence clusters, medium-loss and medium-turbulence clusters, and low-loss and high-turbulence clusters; by analyzing the membership between all flow field binary groups and each cluster, determine the membership weight of each cluster at each moment, and combine the differences in cluster centers between the high-loss and low-turbulence cluster and the medium-loss and medium-turbulence cluster and the low-loss and high-turbulence cluster to determine the flow risk index of the impeller-pump body at each moment.

[0043] Due to the difference in the order of magnitude between the millimeter to sub-millimeter scale of the impeller-pump body clearance and the overall impeller size, the flow field characteristic data has high dimensionality, multicollinearity and physical dimension difference problems, making it difficult to directly quantify the coupling relationship between wear risk and turbulent loss. Therefore, in order to quantitatively evaluate the coupling relationship between wear and turbulent loss, in response to the problems of "high dimensionality, multicollinearity and physical dimension difference", this embodiment uses the principal component analysis algorithm to analyze the dominant role of all types of flow field data at each moment and within the preset time period before, and determines the impeller-pump body flow field binary group at each moment; clusters the flow field binary groups at each moment and within the preset time period before, and determines the membership weight of each cluster at each moment by analyzing the membership between all flow field binary groups and each cluster cluster, and combines the difference in cluster centers between the high-loss low-turbulence cluster and the medium-loss medium-turbulence cluster and the low-loss high-turbulence cluster to determine the impeller-pump body flow risk index at each moment. The specific process is as follows:

[0044] (1) In this embodiment, the principal component analysis algorithm is used to analyze the dominant role of all flow field data at each moment and within a preset time period before that, and the impeller-pump body flow field binary at each moment is determined, specifically:

[0045] In this embodiment, all flow field data at each moment and within the preset time period before it are used as the input of the principal component analysis (PCA) algorithm. In this embodiment, the number of principal components is set to 2. Therefore, the principal component scores PC1 and PC2 at each moment and all moments within the preset time period before it are output to extract the dominant factors affecting high-efficiency energy saving and wear risks. The principal component scores PC1 and PC2 respectively summarize the two key dimensions that affect system performance, and together reveal the dominant mechanism of gap flow on wear and resistance.

[0046] Furthermore, the binary group consisting of the principal component score PC1 and the principal component score PC2 at each moment is recorded as the impeller-pump body flow field binary group at each moment.

[0047] It should be noted that the value of the preset time length is set manually. In this embodiment, the value of the preset time length is 1 minute. In actual application, as other implementation methods, the implementer can also set it by himself based on the specific situation. This embodiment does not impose any special restrictions.

[0048] The principal component analysis (PCA) algorithm is a well-known technology, and its specific principles and processes are not described in detail here.

[0049] (2) Furthermore, this embodiment clusters the flow field binary groups at each moment and all moments within a preset time period before it, wherein all clusters are respectively high-loss and low-turbulence clusters, medium-loss and medium-turbulence clusters, and low-loss and high-turbulence clusters; by analyzing the membership between all flow field binary groups and each cluster, the membership weight of each cluster at each moment is determined, specifically:

[0050] In this embodiment, the flow field binary groups at all moments within a preset time period before each moment are used as inputs of the clustering algorithm, and multiple clusters and the degree of membership between each flow field binary group and different clusters are output. In this embodiment, the number of clusters is set to 3, and the metric distance is the Euclidean distance between the flow field binary groups. All clusters specifically correspond to three typical working conditions: "high wear-low turbulence", "medium wear-medium turbulence", and "low wear-high turbulence". Among them, the "low wear-high turbulence" mode is directly related to the high-efficiency energy-saving goal. In this embodiment, the mean of the particle collision frequency of all flow field binary groups in each cluster at the corresponding moment is calculated, and the clusters corresponding to each mean in the results of descending order of the mean are recorded as high-loss and low-turbulence clusters, medium-loss and medium-turbulence clusters, and low-loss and high-turbulence clusters, that is, the cluster corresponding to the maximum mean is regarded as the high-loss and low-turbulence cluster, the cluster corresponding to the mean between the maximum mean and the minimum mean is regarded as the medium-loss and medium-turbulence cluster, and the cluster corresponding to the minimum mean is regarded as the low-loss and high-turbulence cluster.

[0051] It should be noted that there are many commonly used clustering algorithms. In this embodiment, the fuzzy C-means clustering algorithm is used to cluster the flow field binary groups. In actual application, as other implementation methods, the implementer can also set them according to the specific situation. This embodiment does not impose any special restrictions.

[0052] Among them, the calculation method of Euclidean distance and the fuzzy C-means clustering algorithm are both well-known technologies. Therefore, the specific calculation process of Euclidean distance and the specific process of clustering flow field binary groups using the fuzzy C-means clustering algorithm are not repeated here.

[0053] Furthermore, in this embodiment, the average of the membership between the flow field binary group and each cluster at each moment and all previous moments is taken as the membership weight of each cluster at each moment.

[0054] Calculate the distances from the cluster center of the high-loss, low-turbulence cluster, the cluster center of the medium-loss, medium-turbulence cluster, and the cluster center of the low-loss, high-turbulence cluster to the cluster center of the low-loss, high-turbulence cluster at each moment.

[0055] It should be noted that there are many methods to measure the distance between binary groups. In this embodiment, the Euclidean distances from the cluster centers of the high-loss and low-turbulence clusters, the cluster centers of the medium-loss and medium-turbulence clusters, and the cluster centers of the low-loss and high-turbulence clusters to the cluster centers of the low-loss and high-turbulence clusters at each moment are used as the distances from the cluster centers of the high-loss and low-turbulence clusters, the cluster centers of the medium-loss and medium-turbulence clusters, and the cluster centers of the low-loss and high-turbulence clusters to the cluster centers of the low-loss and high-turbulence clusters at each moment. In actual application, as other implementation methods, the implementers can also set them by themselves according to the specific circumstances, and this embodiment does not impose any special restrictions.

[0056] Furthermore, this embodiment calculates the product of the membership weight of each cluster and the corresponding distance at each moment, and takes the mean of the product of all clusters at each moment as the impeller-pump flow risk index at each moment.

[0057] According to the flow state risk index of the impeller-pump body at each moment, it can be understood that the flow state risk index comprehensively evaluates the flow field state in the impeller-pump body gap at each moment, especially the coupling relationship and degree of harm between its wear risk and turbulent loss, and quantifies the degree to which the current flow state deviates from the ideal high-efficiency energy-saving state, that is, the "low wear-high turbulence" mode. If the cluster center of the high-loss low-turbulence cluster, the cluster center of the medium-loss medium-turbulence cluster, and the cluster center of the low-loss high-turbulence cluster are closer to the cluster center of the low-loss high-turbulence cluster, the smaller the distance is, the closer the working mode to which the current cluster corresponds to or is close to. The deviation from the ideal high-efficiency and energy-saving state will cause the overall flow field to tend toward an inherently high-risk and low-efficiency working state. The larger the corresponding flow state risk index and the greater the membership weight of the current cluster, the more significantly the flow field state tends toward the working mode represented by the current cluster. If the cluster itself is high-risk, such as "high wear-low turbulence" or "medium wear-medium turbulence", then the high membership weight will significantly increase the flow state risk index, reflecting that the current flow field state is clearly in a high-risk or low-efficiency mode.

[0058] On the contrary, if the distance between each cluster center and the low-loss and high-turbulence cluster center is smaller, it means that the operating mode of the flow field corresponding to the current cluster belongs to or is close to the ideal high-efficiency and energy-saving state, which will cause the overall flow field to tend to a working state with lower inherent risk and higher efficiency, and the corresponding flow risk index will be smaller; and if the membership weight of the current cluster is smaller, it means that the possibility of the state of the flow field tending to the operating mode represented by the current cluster is smaller, or it is only very weakly affected by this mode. Even if this cluster itself may have a higher risk, due to the small membership weight, its contribution to the overall flow risk index will be very small, which reflects that the current flow field state has a low correlation with this specific high-risk or inefficient mode, or that the flow field is closer to other modes, which may be a mode with better risk / efficiency, such as closer to the "low wear-high turbulence" mode.

[0059] At this point, this embodiment extracts the dominant factors of the impeller-pump flow field data through principal component analysis (PCA), and uses the fuzzy C-means clustering algorithm to divide the flow field binary groups into three operating conditions: high loss and low turbulence, medium loss and medium turbulence, and low loss and high turbulence. By analyzing the membership of the flow field binary groups and the cluster clusters at each moment, the membership weights are determined, and the flow state risk index is calculated in combination with the cluster center differences. This method effectively solves the problems of high dimensionality, multicollinearity and physical dimension differences in flow field data, quantitatively evaluates the coupling relationship between wear and turbulent loss, helps to identify flow state risks, optimize the impeller-pump operation state, and improve system efficiency and reliability.

[0060] Step S3: Based on all flow risk indices, all particle collision rates and all theoretical life data within a preset time period before each moment, a support vector regression algorithm is used to obtain the theoretical life prediction value of the impeller-pump body at each moment; based on all flow risk indices, all geometric gap dimensions and all theoretical life data within a preset time period before each moment, a particle swarm optimization algorithm is used to obtain the total number of optimal solutions at each moment; based on the theoretical life prediction value and the total number of optimal solutions of the impeller-pump body at each moment, the adjustable diversity of the impeller-pump body at each moment is determined.

[0061] Since the traditional single-objective optimization method is difficult to meet the dual requirements of high efficiency and energy saving and life extension in the optimization of the wear-resistant low-resistance impeller-pump body clearance, the design scheme is unbalanced between turbulent loss reduction and wear resistance. Therefore, in order to consider turbulent loss and wear resistance at the same time and find the best balance between the two, this embodiment adopts a support vector regression algorithm to obtain the theoretical life prediction value of the impeller-pump body at each moment based on all flow risk indices, all particle collision rates and all theoretical life data within a preset time before each moment; based on all flow risk indices, all geometric gap dimensions and all theoretical life data within a preset time before each moment, a particle swarm optimization algorithm is adopted to obtain the total number of optimal solutions at each moment; based on the theoretical life prediction value of the impeller-pump body and the total number of optimal solutions at each moment, the adjustable diversity of the impeller-pump body at each moment is determined. The specific process is as follows:

[0062] (1) This embodiment uses a support vector regression algorithm to obtain the theoretical life prediction value of the impeller-pump body at each moment based on all flow risk indices, all particle collision rates, and all theoretical life data within a preset time period before each moment, specifically:

[0063] In this embodiment, all flow risk indices and all particle collision rates within a preset time period before each moment are used as independent variables in the support vector regression algorithm, and all theoretical life data within a preset time period before each moment are used as dependent variables in the support vector regression algorithm, wherein the kernel function is set to a radial basis function. In this embodiment, the penalty coefficient C in the support vector regression algorithm is set to 1.5, and the insensitive loss is set to 0.05, and the theoretical life prediction value of the impeller-pump body at each moment is output, thereby achieving the purpose of quantifying the impeller durability under different geometric gap parameters and coating combinations, reflecting the synergistic effect of the wear resistance of the wear-resistant coating and the gap flow state.

[0064] The support vector regression algorithm is a well-known technology, and its specific principle will not be described in detail.

[0065] (2) Furthermore, this embodiment uses a particle swarm optimization algorithm to obtain all optimal solutions at each moment based on all flow risk indices, all geometric gap sizes, and all theoretical life data within a preset time period before each moment, specifically:

[0066] In this embodiment, all flow risk indices, all geometric gap dimensions, and all theoretical life data within a preset time period before each moment are used as inputs to the particle swarm optimization algorithm, and the total number of optimal solutions is output. The optimal solution represents the global optimal solution geometry under high efficiency energy saving and life constraints. Each optimal solution achieves a dynamic balance between minimizing turbulent losses and maximizing wear resistance.

[0067] It should be noted that the particle swarm optimization algorithm used in this embodiment is a multi-objective particle swarm optimization (MOPSO) algorithm. In this embodiment, the population size is set to 150, the number of iterations is 300, and the inertia weight is linearly decreased from 0.9 to 0.4 to achieve the effect of initial global exploration and later local optimization. The acceleration factor is set to c1=c2=1.8 to maintain a balance between individual and group experience, and the final output is the total number of optimal solutions in the Pareto front optimal solution set.

[0068] Among them, the multi-objective particle swarm optimization (MOPSO) algorithm is a well-known technology, and its specific principle is not repeated here.

[0069] (3) Furthermore, this embodiment determines the adjustable diversity of the impeller-pump body at each moment based on the theoretical life prediction value of the impeller-pump body at each moment and the total number of optimal solutions, specifically:

[0070] As an implementation mode, in this embodiment, the product of the theoretical life prediction value of the impeller-pump body at each moment and the total number of optimal solutions is used as the adjustable diversity of the impeller-pump body at each moment.

[0071] Preferably, the schematic diagram of the adjustable diversity extraction process provided in this embodiment is as follows: Figure 2shown.

[0072] According to the adjustable diversity of the impeller-pump body at each moment, it can be understood that the adjustable diversity takes into account the durability of the impeller and the flexibility of the design options, and achieves a comprehensive performance level of high energy efficiency and high wear resistance. The larger the theoretical life prediction value, the longer the predicted impeller wear life under the current conditions, which means that the synergistic effect of the wear-resistant coating and the gap flow pattern is better, and the impact of the flow pattern risk on wear is smaller. Even if the diversity of design options remains unchanged, this higher durability itself directly improves the adjustable diversity, which reflects that the equipment is more durable and has lower risks; and the more optimal solutions there are, it means that under the current conditions, more different gap parameter combinations have been found through the optimization algorithm, and these combinations can achieve a good balance between turbulent loss and wear life, which means that the adjustable diversity is also higher;

[0073] On the contrary, if the theoretical life prediction value is smaller, it means that under the current conditions, the predicted impeller wear life is shorter, which means that the synergistic effect of the wear-resistant coating and the gap flow state is poor, and the flow state risk has a greater impact on wear. Even if the diversity of design options remains unchanged, this lower durability itself directly reduces the adjustable diversity, which reflects the poor durability and high risk of the equipment; and the fewer the number of optimal solutions, it means that under the current conditions, the number of gap parameter combinations found by the optimization algorithm that can achieve a balance between efficiency and life has decreased, which shows that the design space has limited room for choice and reduced flexibility, which means that the adjustable diversity is also lower.

[0074] To this end, this embodiment uses support vector regression and particle swarm optimization algorithms, combined with flow risk index, particle collision rate, geometric gap size and theoretical life data, to predict the theoretical life of the impeller-pump body and determine the total number of optimal solutions, thereby obtaining the adjustable diversity of the impeller-pump body. This method effectively solves the limitations of traditional single-objective optimization methods, improves the flexibility and reliability of the impeller-pump body design, and optimizes the balance between turbulent loss and wear resistance.

[0075] Step S4: adjusting the grid density of the impeller-pump body clearance based on the average distribution of the adjustable diversity at all moments in a preset period before the current moment.

[0076] Due to the large span between the millimeter and submillimeter dimensions of the impeller-pump clearance and the overall impeller size, traditional uniform meshing faces difficulties in discretizing the computational domain in CFD simulations. Using a dense mesh to cover the gap increases the global computational workload exponentially, while local mesh refinement can easily lead to sudden changes in mesh size, resulting in aspect ratio imbalances and numerical diffusion effects, reducing the accuracy of simulations of turbulent pulsations, particle collisions, and wear resistance.

[0077] Therefore, this embodiment adjusts the grid density of the impeller-pump body clearance based on the average distribution of the adjustable diversity at all moments in a preset period before the current moment, specifically:

[0078] As an implementation method, in this embodiment, the grid density correction value ρ of the impeller-pump body clearance at the current moment is ′ The expression is: Where, ρ base represents the initial mesh density obtained by the ANSYS Meshing tool; It represents the average value of the adjustable diversity of the impeller-pump body at the current moment and all moments in the previous preset period; exp() represents an exponential function with a natural constant as the base.

[0079] Based on the same inventive concept as the above method, an embodiment of the present application also provides a wear-resistant low-resistance impeller-pump body fit clearance optimization system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-mentioned wear-resistant low-resistance impeller-pump body fit clearance optimization methods.

[0080] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0082] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body, characterized in that: The method comprises the following steps: Real-time acquisition of all flow field data, theoretical life data, and geometric clearance dimensions of the impeller-pump body. All flow field data include: turbulent kinetic energy, dissipation rate, wall shear stress, and particle collision frequency. A principal component analysis algorithm is used to analyze the dominant role of all types of flow field data at each moment and within a preset time period before it, and the impeller-pump body flow field binary group at each moment is determined; the flow field binary groups at each moment and all moments within a preset time period before it are clustered, wherein all clusters are respectively high-loss and low-turbulence clusters, medium-loss and medium-turbulence clusters, and low-loss and high-turbulence clusters; by analyzing the membership between all flow field binary groups and each cluster, the membership weight of each cluster at each moment is determined, and combined with the difference in cluster centers between the high-loss and low-turbulence cluster, the medium-loss and medium-turbulence cluster, and the low-loss and high-turbulence cluster, the flow state risk index of the impeller-pump body at each moment is determined; Based on all flow risk indices, all particle collision rates, and all theoretical life data within a preset time period before each moment, a support vector regression algorithm is used to obtain the theoretical life prediction value of the impeller-pump body at each moment; based on all flow risk indices, all geometric gap dimensions, and all theoretical life data within a preset time period before each moment, a particle swarm optimization algorithm is used to obtain the total number of optimal solutions at each moment; based on the theoretical life prediction value and the total number of optimal solutions of the impeller-pump body at each moment, the adjustable diversity of the impeller-pump body at each moment is determined; The grid density of the impeller-pump body clearance is adjusted based on the average distribution of the adjustable diversity at all moments in a preset time period before the current moment.

2. A wear-resistant low-resistance impeller-pump body fitting clearance optimization method according to claim 1, characterized in that: The method for determining the impeller-pump body flow field binary at each moment is: All kinds of flow field data at each moment and within the preset time period before it are used as the input of the principal component analysis algorithm, and the principal component scores PC1 and PC2 at each moment and all moments before it are output. The tuple composed of the principal component scores PC1 and PC2 at each moment is recorded as the impeller-pump body flow field tuple at each moment.

3. The method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body according to claim 1, wherein: The method for obtaining the high-loss low-turbulence cluster, the medium-loss medium-turbulence cluster, and the low-loss high-turbulence cluster is as follows: The mean of the particle collision frequency of all flow field binary groups in each cluster at the corresponding time is calculated, and the clusters corresponding to each mean in the results of descending order are recorded as high-loss low-turbulence cluster, medium-loss medium-turbulence cluster and low-loss high-turbulence cluster.

4. The method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body according to claim 1, wherein: The membership weight of each cluster at each moment is the result of taking the average of the membership between the flow field binary and each cluster at each moment and all previous moments.

5. The method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body according to claim 1, wherein: The method for determining the impeller-pump flow risk index at each moment is: Calculate the distances from the cluster centers of the high-loss, low-turbulence cluster, the medium-loss, medium-turbulence cluster, and the low-loss, high-turbulence cluster to the cluster center of the low-loss, high-turbulence cluster at each moment; The multiplication result of the membership weight of each cluster and the corresponding distance at each moment is calculated, and the mean of the multiplication results of all clusters at each moment is used as the impeller-pump body flow risk index at each moment.

6. The method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body according to claim 1, wherein: The obtaining of the theoretical life prediction value of the impeller-pump body at each moment includes: All flow risk indices and all particle collision rates within the preset time before each moment are used as independent variables in the support vector regression algorithm, and all theoretical life data within the preset time before each moment are used as dependent variables in the support vector regression algorithm, and the theoretical life prediction value of the impeller-pump body at each moment is output.

7. The method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body according to claim 1, wherein: The particle swarm optimization algorithm is used to obtain the total number of optimal solutions at each moment, including: All flow risk indices, all geometric gap sizes, and all theoretical life data within a preset time period before each moment are used as input to the particle swarm optimization algorithm, and all optimal solutions are output.

8. The method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body according to claim 1, wherein: The adjustable diversity of the impeller-pump body at each moment is the product of the theoretical life prediction value of the impeller-pump body at each moment and the total number of optimal solutions.

9. The method for optimizing the clearance between a wear-resistant low-resistance impeller and a pump body according to claim 1, wherein: The adjusting of the grid density of the impeller-pump body clearance at the current moment includes: The grid density correction value ρ of the impeller-pump body clearance at the current moment ′ The expression is: Where, ρ base represents the initial mesh density obtained by the ANSYS Meshing tool; It represents the average value of the adjustable diversity of the impeller-pump body at the current moment and all moments in the previous preset period; exp() represents an exponential function with a natural constant as the base.

10. A wear-resistant low-resistance impeller-pump body fit clearance optimization system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images