Design method and device for sealing structure of large-diameter slurry valve suitable for high-wear working conditions

Abstract

The present invention relates to the technical field of slurry valve design, and discloses a design method and device for a sealing structure of a large-diameter slurry valve suitable for high-wear working conditions. The method first obtains key geometric parameters of the sealing structure and constructs a three-dimensional topological base model and a mechanical performance simulation model, which are then integrated to form a digital twin model; then a wear morphology time series data set is constructed; at the same time, three-level warning thresholds are set for key working condition parameters, and working condition risk level division rules are established, and real-time working condition risk levels are obtained accordingly; finally, based on the wear morphology time series data set and the real-time working condition risk level, a wear evolution prediction model is constructed, and the key geometric parameters of the sealing structure are optimized in combination with the digital twin model to obtain the optimal geometric parameter combination; the present invention can realize the accurate prediction and optimal design of the wear state of the sealing structure of a large-diameter slurry valve, and significantly improve the service life and reliability of the sealing structure under high-wear working conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of slurry valve design, and more specifically, to a design method and device for a sealing structure of a large-diameter slurry valve suitable for high-wear working conditions. Background Art

[0002] Under high-wear conditions, the stable operation of large-diameter slurry valves is crucial to ensure the safety and continuity of industrial production. However, due to the long-term erosion of high-speed, high-pressure, and highly abrasive slurries, the sealing structure of slurry valves is prone to wear and failure, leading to serious consequences such as leakage and shutdown, which has always been a key technical problem that needs to be solved in the industrial field.

[0003] A Chinese patent with publication number CN117421831A discloses a method for reducing the wear of the valve body of a Y-type slurry valve and a method for verifying its effectiveness. The method modifies the shape of the valve seat sealing ring and changes the inner angle of the valve seat sealing ring to 45°<a≤90°, preferably a=90°, thereby changing the internal flow field characteristics of the Y-type slurry valve, reducing the wear distribution area and reducing the relative wear rate. The method also provides a method for verifying its effectiveness by comparing the wear characteristic distribution diagrams and relative wear rates of different models. However, the method mainly targets the valve body wear problem of the Y-type slurry valve and does not involve the wear prediction and optimization design of the sealing structure. At the same time, the method only relies on changing the geometric shape of the valve seat sealing ring to improve the flow field. For complex and changeable actual working conditions, its adaptability and robustness need to be further improved. In addition, the method lacks dynamic monitoring and prediction of the wear process, and cannot adjust the optimization plan in real time according to changes in actual working conditions.

[0004] The Chinese patent with the publication number CN117786887A proposes a method for wear prediction and structural optimization design of pumping S valves. The method establishes an initial S valve three-dimensional model, performs parameterization on it, determines the structural parameters of the S valve as the design variables, takes the minimum S valve wall wear as the optimization target, and generates the sample space of the S valve model based on the sampling scheme within the preset constraint range. Then, a preset fresh concrete pumping numerical simulation model is constructed, the optimization target value of the S valve under different design variables is calculated, and an agent model based on the sample and the optimization target value is established. Finally, it is determined whether the error of the prediction accuracy of the agent model meets the preset value. If not, the sample size is increased; if it is satisfied, the design variables are optimized by genetic algorithm based on the agent model to obtain the optimal solution of the geometric shape of the S valve. This method optimizes the S valve structure with the goal of reducing the wear of the S valve. The wear degree of the S valve can be reduced through structural improvement, and the wear life of the S valve of the concrete pump truck can be increased. However, this method is mainly aimed at the concrete pumping S valve, and its application object is quite different from the large-diameter slurry valve under high wear conditions targeted by the present invention. The proxy model constructed by this method mainly relies on the results of numerical simulation. However, the accuracy of the numerical simulation model is limited by the understanding of complex wear mechanisms and the limitations of modeling methods, and it is difficult to accurately reflect the wear evolution law under actual working conditions. In addition, this method lacks real-time monitoring and feedback of the wear state during actual operation, and cannot achieve dynamic tracking and prediction of the wear state.

[0005] In summary, the existing technology lacks multi-scale, all-round monitoring methods for the wear process of the slurry valve sealing structure, making it difficult to obtain comprehensive and accurate wear status information, resulting in the inability to accurately grasp the wear evolution law of the sealing structure; the existing design methods mainly rely on experience or numerical simulation, and lack wear prediction and optimization design methods that deeply integrate actual operating conditions, multi-scale wear data and digital twin models, making it difficult to achieve accurate prediction of the wear state of the sealing structure and dynamic optimization of key geometric parameters. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a design method and device for a large-diameter slurry valve sealing structure suitable for high-wear conditions. It integrates a digital twin model, multi-scale wear morphology data and real-time operating parameters, constructs a wear evolution prediction model, and realizes intelligent optimization of key geometric parameters of the sealing structure; it improves the service life and reliability of the slurry valve sealing structure under high-wear conditions.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] The design method of the sealing structure of a large-diameter slurry valve suitable for high-wear conditions includes:

[0009] Obtain key geometric parameters of the slurry valve sealing structure, and construct a three-dimensional topological base model of the slurry valve sealing structure based on the key geometric parameters; construct a mechanical performance simulation model of the slurry valve sealing structure based on the three-dimensional topological base model; fuse and associate the three-dimensional topological base model of the slurry valve sealing structure with the mechanical performance simulation model to construct a digital twin model of the slurry valve sealing structure;

[0010] Obtain a time series of multi-scale wear morphology images of the sealing structure of the slurry valve, and obtain a wear morphology time series data set based on the time series of multi-scale wear morphology images; obtain key operating parameters of the slurry valve, set three-level warning thresholds for the key operating parameters, and establish operating risk level classification rules based on the three-level warning thresholds; collect real-time operating parameters of the sealing structure of the slurry valve, and obtain the real-time operating risk level based on the operating risk level classification rules;

[0011] Based on the wear morphology time series dataset and real-time operating condition risk level, a wear evolution prediction model is constructed. Combined with the digital twin model of the slurry valve sealing structure, the key geometric parameters of the sealing structure are optimized to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

[0012] Furthermore, the digital twin model of the slurry valve sealing structure is constructed including:

[0013] The finite element method is used to solve the mechanical performance simulation model of the slurry valve sealing structure to obtain the stress and strain distribution results of the slurry valve sealing structure under the current working conditions;

[0014] The stress-strain distribution results obtained by finite element solution and calculation are mapped to the three-dimensional topological base model of the slurry valve sealing structure to obtain a digital twin model of the slurry valve sealing structure.

[0015] Further, the multi-scale wear morphology image time series includes a three-dimensional morphology image time series and a nanoscale morphology image time series;

[0016] The method of obtaining a time series of multi-scale wear morphology images of a slurry valve sealing structure comprises:

[0017] Select the critical area of ​​the slurry valve sealing surface;

[0018] In-situ characterization tests were carried out on key areas of the slurry valve sealing surface at different time points to obtain a time series of multi-scale wear morphology images;

[0019] The step of obtaining a wear morphology time series data set according to the multi-scale wear morphology image time series includes:

[0020] Calculate the surface roughness parameters of the three-dimensional topography image time series to obtain a three-dimensional topography feature parameter sequence;

[0021] Perform difference calculation on the time series of nanoscale topography images to obtain a series of geometric profile changes;

[0022] The three-dimensional morphological feature parameter sequence and the geometric profile variation sequence are fused to obtain the wear morphology time series data set.

[0023] Further, the three-level warning threshold includes a pressure three-level warning threshold, a temperature three-level warning threshold and a vibration three-level warning threshold;

[0024] The three-level warning thresholds for key operating parameters are set as follows:

[0025] Obtain the mechanical performance parameters and the first thermal safety margin parameters of the slurry valve sealing material, and determine the three-level pressure warning threshold according to the mechanical performance parameters and the first thermal safety margin parameters; the three-level pressure warning threshold includes the pressure warning threshold and the pressure danger threshold; the mechanical performance parameters include the yield strength σ s , stress fluctuation amplitude of the sealing structure under historical load , Critical failure pressure of sealing structure and the maximum pressure under historical loads The first thermal safety margin parameter includes the maximum temperature that the sealing structure can withstand and critical failure temperature of sealing materials ;

[0026] Obtain the second thermal safety margin parameter of the sealing material, and determine the temperature three-level warning threshold according to the first thermal safety margin parameter and the second thermal safety margin parameter; the temperature three-level warning threshold includes the temperature warning threshold and the temperature danger threshold; the second thermal safety margin parameter includes the safe working temperature T0 of the sealing material, the allowable amplitude ΔT of the sealing material temperature fluctuation w and the safety margin temperature difference ΔT of the sealing material d ;

[0027] The first vibration data of the slurry valve sealing structure is obtained, and a three-level vibration warning threshold is determined according to the first vibration data; the three-level vibration warning threshold includes a vibration warning threshold and a vibration danger threshold; the first vibration data includes an average vibration level V of the slurry valve sealing structure under rated working conditions. a and critical resonance vibration velocity V cr .

[0028] Further, determining the third-level pressure warning threshold according to the mechanical performance parameter and the first thermal safety margin parameter includes:

[0029] Get the number of cycles of the sealing structure under historical loads and reference value of fatigue life of sealing materials ,according to , , and σ s , calculate the pressure warning threshold coefficient C w ; According to the yield strength σ s and pressure warning threshold coefficient C w , get the pressure warning threshold P w ;

[0030] according to , , and , calculate the pressure danger threshold coefficient C d ; According to the yield strength σ s and pressure warning threshold coefficient C d , get the pressure warning threshold P d .

[0031] Further, determining the temperature level 3 warning threshold according to the first thermal safety margin parameter and the second thermal safety margin parameter includes:

[0032] According to the safe working temperature T0 and the allowable amplitude of temperature fluctuation ΔT s , calculate the temperature warning threshold T w ; According to the safety margin temperature difference ΔT d and critical failure temperature , calculate the temperature danger threshold T d .

[0033] Further, determining the vibration level 3 warning threshold according to the first vibration data includes:

[0034] Set the warning margin coefficient R w and the risk margin factor R d , according to the warning margin coefficient R w and the average vibration level V under rated conditions a , calculate the vibration warning threshold V w ;

[0035] According to the risk margin factor R d and critical resonance vibration velocity V cr , calculate the vibration danger threshold V d .

[0036] Furthermore, the establishment of operating condition risk level classification rules according to the three-level warning thresholds includes:

[0037] According to different key working condition parameters, corresponding working condition parameter fuzzy linguistic variables, working condition parameter fuzzy subsets and working condition parameter domains are set;

[0038] According to the working condition parameter fuzzy linguistic variables, working condition parameter fuzzy subsets and working condition parameter domains corresponding to each key working condition parameter, the working condition parameter fuzzy subset membership function of the corresponding key working condition parameter is set;

[0039] Define the operating condition risk fuzzy linguistic variables {normal, abnormal, dangerous}, the operating condition risk fuzzy subset {normal level, abnormal level, dangerous level} and the operating condition risk domain [0,1];

[0040] Based on the fuzzy subset membership function of the working condition parameters, the working condition risk value is calculated; the membership function of the corresponding working condition risk fuzzy subset is defined in the working condition risk domain;

[0041] According to the operating condition risk value and the membership function of the operating condition risk fuzzy subset, the operating condition risk level is obtained;

[0042] According to the IF-THEN format, fuzzy rules are established and a fuzzy rule base is generated; each fuzzy rule contains an antecedent part and a consequent part, and the format of the fuzzy rule is: IF antecedent part THEN consequent part, and the operating condition risk level is written into the consequent part.

[0043] Furthermore, the construction of the wear evolution prediction model includes:

[0044] Obtain a historical wear morphology time series dataset from the wear morphology time series dataset;

[0045] A time window is selected, and according to the selected time window, a continuous statistical feature parameter subsequence is extracted from the historical wear morphology time series data set as an input feature vector;

[0046] The statistical characteristic parameters of the next moment corresponding to the input feature vector are used as the output target value to construct input-output pairs to form training samples;

[0047] Through the sliding time window, the entire historical wear morphology time series data set is traversed to generate a complete training sample set; the real-time working condition risk level is used as the attribute label of the training sample and integrated into the training sample set;

[0048] A long short-term memory network (LSTM) model was selected, and multiple wear evolution prediction sub-models were trained using a training sample set incorporating attribute labels according to different real-time operating condition risk levels.

[0049] A large-diameter slurry valve sealing structure design device suitable for high-wear working conditions, which is used to implement the above-mentioned large-diameter slurry valve sealing structure design method suitable for high-wear working conditions, and the device includes:

[0050] Digital twin model construction module: used to obtain the key geometric parameters of the slurry valve sealing structure, and construct a three-dimensional topological base model of the slurry valve sealing structure based on the key geometric parameters; construct a mechanical performance simulation model of the slurry valve sealing structure based on the three-dimensional topological base model; fuse and associate the three-dimensional topological base model of the slurry valve sealing structure with the mechanical performance simulation model to construct a digital twin model of the slurry valve sealing structure;

[0051] Multi-scale data acquisition module: used to obtain the multi-scale wear morphology image time series of the slurry valve sealing structure, and obtain the wear morphology time series data set according to the multi-scale wear morphology image time series;

[0052] Working condition risk classification module: used to obtain key working condition parameters of the slurry valve, set three-level warning thresholds for the key working condition parameters, and establish working condition risk level classification rules based on the three-level warning thresholds; collect the real-time working condition parameters of the slurry valve sealing structure, and obtain the real-time working condition risk level based on the working condition risk level classification rules;

[0053] Geometric parameter optimization module: Based on the wear morphology time series data set and the real-time operating condition risk level, a wear evolution prediction model is constructed. Combined with the digital twin model of the slurry valve sealing structure, the key geometric parameters of the sealing structure are optimized to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] Aiming at the problem that the sealing structure of large-caliber slurry valve is prone to failure under high-wear conditions, the present invention proposes a new design method based on digital twin and multi-scale data fusion, realizing the transformation from traditional empirical design to data-driven intelligent design. The method deeply integrates physical models and data models, and realizes the accurate mapping and virtual simulation of the physical world of the sealing structure by constructing a high-precision digital twin model, providing a reliable virtual test platform for the performance analysis and optimization of the sealing structure. At the same time, the method innovatively introduces multi-scale wear morphology image time series data, and combines it with real-time working condition parameters to construct a wear evolution prediction model, realizing the accurate prediction of the wear state of the sealing structure, making up for the deficiency that traditional methods are difficult to accurately capture complex wear behaviors. More importantly, the method combines the wear prediction results with the digital twin model, realizes the intelligent optimization of the key geometric parameters of the sealing structure, and can dynamically adjust the design of the sealing structure according to the actual working conditions and wear conditions, thereby significantly improving the adaptability and durability of the sealing structure under high-wear conditions, extending its service life, reducing maintenance costs, and ensuring the safe and reliable operation of the production process. This data-driven intelligent design method not only improves design efficiency and accuracy, but also provides new ideas and technical means for the design and operation of similar complex engineering equipment, and has important promotion and application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0057] Figure 1 It is a principle flow chart of the design method of the sealing structure of a large-diameter slurry valve suitable for high-wear conditions in the present invention;

[0058] Figure 2 It is a flow chart of a method for constructing a mechanical performance simulation model of a slurry valve sealing structure based on a three-dimensional topological base model in a design method of a large-diameter slurry valve sealing structure suitable for high-wear conditions of the present invention;

[0059] Figure 3 A flow chart of a method for constructing a digital twin model of a slurry valve sealing structure in a large-caliber slurry valve sealing structure design method suitable for high-wear working conditions of the present invention;

[0060] Figure 4 It is a flow chart of a method for digitally processing a time series of multi-scale wear morphology images in a design method of a large-caliber slurry valve sealing structure suitable for high-wear conditions of the present invention;

[0061] Figure 5 A flow chart of a method for determining a three-level vibration warning threshold in a design method for a large-diameter slurry valve sealing structure suitable for high-wear conditions of the present invention;

[0062] Figure 6 A flow chart of a method for obtaining a real-time operating condition risk level in a large-diameter slurry valve sealing structure design method suitable for high-wear operating conditions of the present invention;

[0063] Figure 7 A flow chart of a method for constructing a training sample set of a wear evolution prediction model in a design method of a large-diameter slurry valve sealing structure suitable for high-wear conditions of the present invention;

[0064] Figure 8 It is a flow chart of a method for obtaining an optimal geometric parameter combination of a slurry valve sealing structure in a design method of a large-diameter slurry valve sealing structure suitable for high-wear working conditions of the present invention;

[0065] Fig. 9 This is a functional module diagram of the sealing structure design device for a large-diameter slurry valve suitable for high-wear conditions in the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Example 1

[0068] See also Figure 1 As shown, this embodiment provides a design method for a large-diameter slurry valve sealing structure suitable for high-wear working conditions, including:

[0069] Step S1000, obtaining key geometric parameters of the slurry valve sealing structure, and constructing a three-dimensional topological base model of the slurry valve sealing structure according to the key geometric parameters; constructing a mechanical property simulation model of the slurry valve sealing structure based on the three-dimensional topological base model; fusing and associating the three-dimensional topological base model of the slurry valve sealing structure and the mechanical property simulation model to construct a digital twin model of the slurry valve sealing structure;

[0070] Further, step S1000 includes:

[0071] Step S1100, obtaining key geometric parameters of the slurry valve sealing structure, and constructing a three-dimensional topological base model of the slurry valve sealing structure according to the key geometric parameters;

[0072] Specifically, first of all, it is necessary to comprehensively and carefully analyze the overall structural design drawings of the slurry valve, especially the parts directly related to the seal. This process requires designers to have a solid foundation in mechanical drawing and rich experience in valve design, to be able to accurately understand the geometric meaning of each view, section view and partial enlargement in the drawing, and to have a deep understanding of the valve assembly structure and sealing principle.

[0073] Based on the analysis of the drawings, the key geometric parameters that have a significant impact on the sealing performance are extracted. These parameters include but are not limited to: sealing surface width, sealing surface inclination (for example, the cone angle of the conical sealing surface), sealing surface surface roughness (such as Ra, Rz, etc.), inner and outer diameter dimensions of the sealing ring, and assembly dimension chain of the sealing structure. Among them, the sealing surface width directly determines the size of the sealing contact area, which in turn affects the distribution of the sealing contact stress; the sealing surface inclination is closely related to the self-tightness of the sealing pair and the size of the leakage channel; the sealing surface surface roughness affects the microscopic contact state and lubrication conditions of the sealing surface; the inner and outer diameter dimensions of the sealing ring need to be strictly matched with the valve stem diameter and the inner diameter of the valve seat to ensure good coaxiality and sealing effect; the calculation and control of the assembly dimension chain are crucial to ensure the assembly quality and reliability of the sealing structure. Therefore, extracting these key geometric parameters requires designers to comprehensively consider multiple factors such as sealing principles, processing technology, and assembly requirements, and take into account representativeness and measurability when selecting parameters.

[0074] After obtaining the key geometric parameters, it is necessary to build a three-dimensional solid model of the slurry valve sealing structure in the three-dimensional CAD software. This process can use parametric modeling technology to map the extracted geometric parameters into dimension-driven parameters of the three-dimensional model. Parametric modeling is a constraint-based geometric modeling method. By defining geometric constraints (such as parallelism, coaxiality, etc.) and dimensional constraints (such as length, diameter, etc.) in the solid model, parameter associations between model features are established to achieve rapid modification and updating of the model. For example, the inner and outer diameter dimensions of the sealing ring can be set as parameters, and the shape and size of the sealing ring can be changed by modifying the parameter values ​​without redrawing the model. Parametric modeling enables designers to generate and optimize three-dimensional models flexibly and efficiently, laying a solid geometric foundation for subsequent structural analysis and optimization.

[0075] Through the above steps, a parametric three-dimensional solid model of the slurry valve sealing structure, also known as the three-dimensional topological base mold, is obtained. This model completely describes the geometric shape, topological relationship and parameter attributes of the sealing structure, and is the core data carrier for digital analysis and optimization. Compared with traditional static geometric models, parameterized three-dimensional topological base molds have stronger flexibility and adaptability. Designers can easily adjust key parameters, generate multiple design schemes, and evaluate their impact on sealing performance. In addition, the three-dimensional topological base mold can also be seamlessly integrated with digital tools such as finite element analysis and multi-physics field simulation to achieve seamless connection and closed-loop iteration of design-analysis-optimization, greatly improving design efficiency and optimization quality.

[0076] Step S1200, constructing a mechanical performance simulation model of a slurry valve sealing structure based on a three-dimensional topological base model;

[0077] Furthermore, if Figure 2As shown, step S1200 includes:

[0078] Step S1210, meshing the three-dimensional topological base model to generate a preliminary finite element model;

[0079] Step S1220, defining material property parameters of the preliminary finite element model to obtain a material property enabled finite element model;

[0080] Step S1230, setting the boundary conditions and current load conditions of the material property-enabled finite element model to obtain a mechanical property simulation model of the slurry valve sealing structure.

[0081] Specifically, after obtaining the three-dimensional topological base model of the slurry valve sealing structure, it is necessary to build a corresponding mechanical performance simulation model on this basis. The model aims to analyze and predict the mechanical performance indicators such as stress-strain distribution, contact state, fatigue life, etc. of the sealing structure under actual working conditions through numerical simulation methods, and provide quantitative basis for the optimization design and reliability evaluation of the sealing structure. Step S1210 meshes the three-dimensional topological base model, which is a pre-processing link of finite element analysis. Meshing refers to discretizing the continuous geometric domain into a mesh model composed of nodes and units to provide discrete space division for solving the control equation. The quality of the mesh directly affects the accuracy and convergence speed of the finite element calculation. In view of the geometric characteristics of the slurry valve sealing structure (such as narrow and long contact areas, sharp edges, etc.), it is necessary to adopt adaptive meshing technology to ensure the quality of the mesh while taking into account the calculation efficiency. For example, local mesh encryption can be performed in the sealing contact area to improve the calculation accuracy of the area with large stress-strain gradient; larger mesh size is used in non-critical areas to reduce the number of units and improve the calculation speed. The generated mesh model should be quality checked, such as unit size, distortion and other indicators must meet the requirements of the finite element solver. A high-quality mesh model is an important prerequisite for conducting mechanical performance simulation.

[0082] Step S1220 defines and assigns material properties to the finite element model. Material property parameters are key physical quantities that describe the mechanical behavior of materials and directly determine the response of the structure under load. For the slurry valve sealing structure, the key material parameters include: elastic modulus, Poisson's ratio, yield strength, tensile strength, density, etc. Among them, the elastic modulus and Poisson's ratio determine the stiffness characteristics of the material in the elastic deformation stage; the yield strength and tensile strength determine the ultimate stress level of the material during plastic deformation and fracture failure; the density affects the inertial characteristics and vibration characteristics of the structure. For high wear conditions, it is also necessary to focus on the tribological properties of the material, such as hardness and wear resistance. In addition, for material behaviors that exhibit obvious nonlinearity (such as hyperelastic rubber materials, etc.), it is necessary to obtain their stress-strain curves through material tests and introduce them into the finite element material library through methods such as hyperelastic constitutive models. The accurate definition and assignment of material properties are key factors to ensure the reliability of simulation results.

[0083] Step S1230 applies boundary conditions and load conditions to the finite element model. This process is the key link in mapping the actual service environment of the slurry valve sealing structure to the numerical model. According to the form and working principle of the sealing structure, it is necessary to reasonably set the displacement constraint and load distribution function. Taking the valve stem-stuffing box seal as an example, an axial displacement constraint can be applied to the lower end face of the valve stem to simulate its cooperation with the valve plate; a preload can be applied to the upper end face of the stuffing box to simulate the effect of the stuffing gland; and radial pressure can be applied to the contact surface between the valve stem and the stuffing to simulate the effect of the medium pressure. For dynamic seals, it is also necessary to apply rotational speed or speed boundary conditions to simulate the reciprocating motion of the valve stem. The setting of boundary conditions and loads needs to be as close to the actual working conditions as possible and coordinated with the deformation of the sealing structure. In addition, it is necessary to define the contact algorithm and friction model for the sealing contact surface to simulate the contact state and friction dissipation of the sealing pair. After the above processing, the mechanical properties simulation model of the slurry valve sealing structure is completed. The model comprehensively considers multiple physical field effects such as materials, geometry, loads, and contact, and can comprehensively and accurately simulate the mechanical behavior of the sealing structure during service, providing a reliable digital means for subsequent performance evaluation and optimized design.

[0084] Step S1200 makes full use of numerical simulation technology to build a high-fidelity mechanical performance simulation model based on the geometric model of the sealing structure. This model maps the physical information such as material properties, geometric characteristics, load conditions, etc. of the sealing structure into a discrete mathematical model through meshing, material property assignment, boundary condition definition, etc., so that the originally complex sealing system can be efficiently solved through finite element equations. Compared with traditional empirical design and physical experiments, numerical simulation can quantitatively analyze the stress and strain distribution inside the structure, visualize weak links, and accelerate design iterations, thereby greatly improving design efficiency and product quality, and reducing R&D costs and trial-and-error risks. At the same time, the mechanical performance simulation model is also an important part of the digital twin. Together with the geometric model, physical model, and data model, it constitutes a multidisciplinary and multi-scale mapping of the slurry valve sealing structure in the virtual space, laying a solid foundation for realizing intelligent management and optimized control throughout the life cycle.

[0085] Step S1300: fuse and associate the three-dimensional topological base model and the mechanical performance simulation model of the slurry valve sealing structure to construct a digital twin model of the slurry valve sealing structure.

[0086] Furthermore, if Figure 3 As shown, step S1300 includes:

[0087] Step S1310, performing finite element solution calculation on the mechanical performance simulation model of the slurry valve sealing structure to obtain the stress and strain distribution results of the slurry valve sealing structure under the current working condition;

[0088] Step S1320, mapping the stress-strain distribution results obtained by finite element solution calculation to the three-dimensional topological base model of the slurry valve sealing structure to obtain a digital twin model of the slurry valve sealing structure.

[0089] Specifically, step S1300 is the key to establishing a complete digital twin model, which organically combines the static geometric model and the dynamic mechanical performance simulation model to form a high-fidelity digital model that integrates structure and performance, and realizes the two-way mapping and interaction between the virtual model and the physical entity. This fusion process requires the association and synchronization of data and models at multiple levels such as geometry, physics, and behavior.

[0090] Step S1310 performs finite element solution calculation on the mechanical performance simulation model of the slurry valve sealing structure. Finite element solution is the process of approximating partial differential equations through numerical methods to obtain the mechanical response of complex structures under load. According to the force characteristics and material properties of the sealing structure, it is necessary to select appropriate solution algorithms and convergence criteria. Currently, there are two common solution algorithms: direct method and iterative method. Direct methods, such as Gaussian elimination method, are suitable for small-scale, linear problems; iterative methods, such as conjugate gradient method, are suitable for large-scale, nonlinear problems. In addition, it is necessary to select appropriate subspace methods, such as ICCG method, multi-grid method, etc., according to the mathematical characteristics of the problem (such as positive definiteness, symmetry, etc.) to improve the solution efficiency. After finite element solution, the numerical solution of physical quantities such as displacement field, stress field, strain field, etc. of the slurry valve sealing structure under the current working conditions can be obtained. These results are usually visualized in the form of cloud maps, which intuitively reflect the distribution of mechanical states inside the structure. Designers can use this information to analyze key information such as the stress concentration area and maximum deformation position of the sealing structure, determine whether the sealing structure has potential failure risks such as insufficient strength, insufficient rigidity, and poor stability, and take corresponding optimization measures. For example, the stress cloud map shows that there is a sharp stress concentration in the contact area between the valve stem and the packing. Designers can relieve stress concentration and improve sealing reliability by setting stress buffer structures (such as chamfers, shoulders, etc.) in this area.

[0091] Step S1320 maps the stress-strain distribution results obtained by the finite element solution calculation to the three-dimensional topological base model of the slurry valve sealing structure. This mapping process is a key step in realizing the fusion of the geometric model and the physical model. Since the finite element model is a discretized simplification of the geometric model, its grid nodes and units do not completely correspond to the features of the geometric model. In order to synchronize the analysis results with the design model, it is necessary to establish a mapping relationship between the two models. There are two commonly used mapping methods: interpolation mapping and projection mapping. Interpolation mapping is to calculate the corresponding physical quantities on the vertices or surfaces of the geometric model through interpolation functions (such as linear interpolation, spline interpolation, etc.) based on the finite element node results. This method is simple to calculate, but the accuracy is relatively low. Projection mapping is to project the finite element results onto the characteristic surface or characteristic edge of the geometric model, and then transfer it to the relevant area according to the topological relationship of the feature to obtain a continuous and smooth distribution of physical quantities. Although this method is complex in calculation, it can better maintain the continuity and consistency of the results. Through the above mapping method, the results of the mechanical properties simulation can be integrated with the geometric model to generate a digital twin model integrating geometry and physics. In the visual interface, designers can directly view the distribution of physical quantities such as stress and strain on the three-dimensional geometric model, and perform interactive operations such as rotation, scaling, and sectioning, so as to analyze the mechanical properties of the sealing structure from different angles and scales. The advantage of this fusion model is that while designers are modifying geometric parameters, they can evaluate the impact of geometric changes on mechanical properties in real time, accelerating design iteration and optimization. For example, when designers increase the taper angle of the sealing surface, they can observe the changing trend of the contact stress in real time in the fusion model to determine whether the selected parameters are reasonable.

[0092] Through step S1300, the construction of the digital twin model is basically completed. The model not only contains the precise three-dimensional geometric information of the slurry valve sealing structure, but also integrates the simulation results of mechanical properties under actual working conditions. It is a high-fidelity digital mapping of the physical sealing system. This highly integrated virtual model provides a powerful digital tool for performance evaluation, fault diagnosis, predictive maintenance, etc. of the sealing structure. For example, through the digital twin model, the operating conditions of the slurry valve can be monitored in real time, and the remaining life of the sealing structure can be predicted by combining historical data and machine learning algorithms, and the optimal maintenance strategy can be formulated to avoid losses caused by unplanned downtime. For another example, when the slurry valve has abnormal conditions such as leakage, the digital twin model can be used to restore the working conditions when the accident occurred, analyze the stress state of the seal failure, infer the failure mechanism, and provide digital evidence for accident investigation and improved design.

[0093] In short, the digital twin model is an important foundation for realizing intelligent management of the entire life cycle of sealing structures. Step S1300 builds a digital system of structure, performance, and working conditions by fusing and associating multi-source heterogeneous data such as geometric models, physical models, and behavioral models, enabling engineers to reproduce, analyze, and predict the actual behavior of the sealing system in a virtual space, greatly improving the intelligence level and decision-making efficiency of design, production, operation, and maintenance.

[0094] Step S2000, obtaining a multi-scale wear morphology image time series of the slurry valve sealing structure, and obtaining a wear morphology time series data set based on the multi-scale wear morphology image time series; obtaining key operating parameters of the slurry valve, setting three-level warning thresholds for the key operating parameters, and establishing operating risk level classification rules based on the three-level warning thresholds; collecting real-time operating parameters of the slurry valve sealing structure, and obtaining the real-time operating risk level based on the operating risk level classification rules;

[0095] Furthermore, step S2000 includes:

[0096] Step S2100, obtaining a multi-scale wear morphology image time series of the slurry valve sealing structure, and obtaining a wear morphology time series data set according to the multi-scale wear morphology image time series; the multi-scale wear morphology image time series includes a three-dimensional morphology image time series and a nanoscale morphology image time series;

[0097] Step S2100 aims to obtain the wear morphology data of the slurry valve sealing structure during actual operation, and provide the original experimental basis for the subsequent wear mechanism analysis and life prediction. Since wear is a complex process across scales and time domains, the generation of macroscopic wear morphology usually originates from the material removal and deformation mechanism at the micro and nano scales. Therefore, comprehensive collection of wear morphology images at different scales and time nodes is crucial for a deep understanding of the wear law and evolution mechanism.

[0098] Furthermore, step S2100 includes:

[0099] Step S2110, selecting a key area of ​​the sealing surface of the slurry valve;

[0100] Specifically, step S2110 selects the key area of ​​the sealing surface of the slurry valve. Since the force and contact state of the sealing surface are not uniform along the circumference and radial direction, the wear condition of the local area has different effects on the overall sealing performance. In order to take into account the pertinence and representativeness of data collection, it is necessary to identify the areas that are most critical to the sealing performance based on the simulation results of the mechanical properties of the sealing structure. These areas usually have characteristics such as large contact stress, high relative sliding speed, and strong stress gradient. Engineers can use the digital twin model to visualize the stress cloud map, wear power density distribution, etc. of the sealing surface to intuitively judge the weak links of the sealing surface. At the same time, the accessibility and measurability of the key areas must also be considered. For rotary seals, the key area can be selected as the contact area between the valve stem and the packing; for reciprocating seals, the key area can be selected as the telescopic end face of the sealing ring.

[0101] Step S2120, performing in-situ characterization tests on key areas of the slurry valve sealing surface at different time points to obtain a multi-scale wear morphology image time series;

[0102] Specifically, step S2120 carries out in-situ characterization tests on key areas of the slurry valve sealing surface at different time points. Wear is a gradual dynamic process, and its morphological features evolve over time. In order to accurately characterize the dynamic behavior of wear, it is necessary to measure the sealing surface at different stages of the slurry valve operation, and minimize the interference of the measurement process on the normal operation of the system. There are two commonly used in-situ measurement methods: embedded and penetrating. The embedded method is to place a micro sensor (such as a thin film piezoresistive sensor, an eddy current sensor, etc.) between the secondary surfaces of the sealing pair to monitor the contact state and wear of the secondary surfaces in real time. The advantages of this method are high measurement accuracy and high frequency, and the disadvantage is that it may change the contact characteristics of the secondary surfaces and accelerate wear. The penetrating method is to use highly penetrating media such as ultrasound, X-rays, and neutrons to perform non-destructive detection of the inside of the sealing pair from the outside. The advantages of this method are no interference and no damage, and the disadvantage is that the measurement accuracy is relatively low. In view of the characteristics of the slurry valve seal, a strategy combining regular sampling and online monitoring can be adopted. For example, for packing seals, a micro pressure sensor array can be arranged in the contact area between the valve stem and the packing to monitor the distribution changes of the contact pressure in real time and estimate the wear volume. At the same time, after a certain period of operation, the packing gland is disassembled, the outermost graphite ring of the packing is taken out, and the three-dimensional wear morphology of its end face is measured using a laser confocal microscope, and the nano-morphology of its surface is measured using an atomic force microscope. Combining the online monitoring data with the offline characterization results, the wear morphology evolution information at different time scales and different spatial resolutions can be obtained. On the one hand, this information can be used to reveal the wear mechanism of the packing seal, and on the other hand, it is also an important input of the digital twin model, which can be used to correct the model parameters and improve the prediction accuracy.

[0103] Step S2130, digitally process the multi-scale wear morphology image time series to obtain a wear morphology time series data set.

[0104] Furthermore, if Figure 4 As shown, step S2130 includes:

[0105] Step S2131, calculating the surface roughness parameters of the three-dimensional topography image time series to obtain a three-dimensional topography feature parameter sequence;

[0106] Step S2132, performing difference calculation on the nanoscale topography image time series to obtain a geometric profile variation sequence;

[0107] Step S2133, the three-dimensional morphological feature parameter sequence and the geometric profile variation sequence are merged to obtain a wear morphology time series data set.

[0108] Specifically, after obtaining the multi-scale wear morphology image time series, these image data need to be digitized and feature extracted to facilitate the subsequent wear evolution analysis and prediction. This process involves multiple disciplines such as image processing and data mining, and requires the comprehensive use of a variety of mathematical tools and calculation methods.

[0109] First, the original wear morphology image needs to be preprocessed, including denoising, enhancement, segmentation and other operations. Due to the complexity of the in-situ observation environment, the collected image data will inevitably be affected by various noises and distortions, such as optical noise, blur caused by mechanical vibration, etc. Therefore, it is necessary to use digital image processing technologies such as median filtering and wavelet transform to remove high-frequency noise and interference information in the image and improve the signal-to-noise ratio and clarity of the image. At the same time, it is also necessary to perform operations such as contrast enhancement and histogram equalization on the image to highlight the morphological features of the worn area in preparation for subsequent feature extraction. For three-dimensional morphological images, surface reconstruction and morphological processing are also required to obtain smooth and continuous three-dimensional surfaces.

[0110] After image preprocessing, it is necessary to extract key characteristic parameters that can reflect the wear state and evolution law from the multi-scale wear morphology image. These characteristic parameters should have clear physical meanings, be able to quantitatively describe the morphological characteristics of the wear surface, and have good discrimination ability for the type and degree of wear. For three-dimensional morphology images, a series of three-dimensional surface structure parameters specified by the international standard ISO25178 are usually used, such as arithmetic mean height Sa, root mean square height Sq, maximum height Sz, peak-to-valley height Sv, etc. These parameters reflect the microscopic morphological characteristics of the wear surface such as roughness, undulation, peak-to-valley distribution from different aspects, and are important indicators for evaluating the wear state. For nanoscale morphology images, the geometric contour features of the wear area can be extracted, such as wear depth, wear volume, wear area, etc. By performing difference calculation and cumulative summation on nano-morphology images at different time points, the progress of wear can be quantitatively evaluated and the trend of wear rate can be described.

[0111] After extracting the key characteristic parameters, it is necessary to time-series these parameter data to construct a wear morphology time series data set. Specifically, for each key characteristic parameter, arrange them in chronological order to form a time series. The time series of different characteristic parameters can be stored in parallel in the data set to form a multivariate time series. In this way, each timestamp corresponds to a set of wear morphology feature vectors, reflecting the comprehensive wear state of the sealing surface at that moment. The time series data set can be stored and organized using a variety of data management technologies such as relational databases and NoSQL databases to facilitate subsequent data access and analysis.

[0112] Through the above steps, the original wear morphology image data is digitally processed and feature extracted, and converted into a structured wear morphology time series data set. This data set uses time as the index and the extracted feature parameters as the attribute dimension, and comprehensively and systematically records the evolution of the wear state of the slurry valve sealing surface during long-term service. Compared with the original image, the time series data set has the advantages of small data volume, information concentration, and convenience in calculation. It is an ideal data basis for analyzing and predicting the laws of wear evolution. Based on time series data, data-driven modeling methods such as machine learning and deep learning can be used to discover the intrinsic correlation of the wear evolution process, build a wear prediction model, realize online evaluation of wear status and prediction of remaining life, and provide a basis for preventive maintenance of slurry valves.

[0113] Step S2200, obtain the key operating parameters of the slurry valve, set three-level warning thresholds for the key operating parameters, and establish operating risk level classification rules based on the three-level warning thresholds; collect the real-time operating parameters of the slurry valve sealing structure, and obtain the real-time operating risk level based on the operating risk level classification rules.

[0114] Further, step S2200 includes:

[0115] Step S2210, obtaining key operating parameters of the slurry valve, wherein the key operating parameters include pressure, temperature and vibration;

[0116] During the long-term service of the slurry valve, its sealing structure will be affected by a variety of operating parameters, such as slurry pressure, temperature, vibration, etc. The changes in these operating parameters will directly affect the stress state, lubrication state and wear mechanism of the sealing surface, and then affect the wear rate and failure mode of the sealing structure. Therefore, real-time collection and monitoring of key operating parameters are of great significance for understanding the service status of the slurry valve and evaluating its health level.

[0117] First, it is necessary to identify the key operating parameters that have a significant impact on seal wear based on the structural characteristics and failure mechanism of the slurry valve. Through theoretical analysis, simulation calculations and field experience, the following three key operating parameters can be determined:

[0118] (1) Slurry pressure: Slurry pressure is the main driving force causing seal surface wear. The higher the pressure, the greater the contact stress on the seal surface and the more severe the wear. In addition, pressure fluctuations can also cause fatigue wear on the seal surface. Therefore, it is necessary to focus on monitoring the pressure values ​​at the inlet and outlet of the slurry valve.

[0119] (2) Slurry temperature: The slurry temperature affects the mechanical properties and lubrication state of the sealing material. Too high a temperature will cause the strength and hardness of the sealing material to decrease, which will increase wear. Too low a temperature may lead to poor lubrication, increase the friction coefficient, and induce adhesive wear. Therefore, it is necessary to monitor the temperature distribution inside the slurry valve in real time.

[0120] (3) Slurry valve vibration: The slurry valve will generate large vibration and impact load when opening / closing, which will change the contact state of the sealing pair and aggravate wear. Vibration will also cause fatigue damage to the sealing surface and shorten its service life. Therefore, it is necessary to collect the vibration acceleration of the slurry valve in real time.

[0121] After determining the key operating parameters, it is necessary to arrange sensors at appropriate locations of the slurry valve to realize online monitoring of the operating parameters. Taking pressure monitoring as an example, pressure sensors can be installed on the inlet and outlet pipes of the slurry valve to collect the pressure value in the pipe in real time. Considering the corrosion and scouring effect of slurry on pressure sensors, it is necessary to select corrosion-resistant and wear-resistant pressure sensors, such as ceramic piezoresistive pressure sensors. At the same time, it is also necessary to calibrate and maintain the pressure sensor regularly to ensure the accuracy and stability of its measurement. For temperature monitoring, a thermocouple installation hole can be opened on the inner wall of the sealing cavity of the slurry valve, and a thermocouple sensor can be embedded to measure the temperature distribution inside the slurry valve. Thermocouple sensors have the characteristics of fast response and wide measurement range, which can meet the temperature measurement requirements of slurry working conditions. For vibration monitoring, an acceleration sensor can be attached to the surface of the slurry valve body or the valve stem bearing seat to collect the vibration signal of the slurry valve. The acceleration sensor can measure vibration acceleration in multiple directions and evaluate the vibration level of the slurry valve.

[0122] By arranging sensors reasonably and using the data acquisition system to read sensor data in real time, the time series of key operating parameters during the service life of the slurry valve can be obtained. These operating parameters reflect not only the effect of the slurry medium on seal wear, but also the operating status of the slurry valve itself. Based on the time series of operating parameters, two functions can be achieved: first, to establish a correlation model between operating parameters and wear status, and quantitatively evaluate the wear rate and degree of wear of the sealing structure under different operating condition combinations; second, to set early warning thresholds for key operating parameters, and when the parameters exceed the thresholds, timely warnings are issued to avoid excessive wear of the sealing structure. For example, when the slurry pressure continues to exceed 1.2 times the design pressure, it may accelerate the wear of the sealing surface, and it is necessary to adjust the valve opening or replace the sealing element in time to ensure sealing safety.

[0123] Step S2210 systematically identifies and monitors the key operating parameters of the slurry valve, establishes a quantitative relationship between seal wear and service conditions, and provides an important basis for evaluating the health status of the sealing structure and predicting its remaining life. The real-time collection and analysis of operating parameters also lays a data foundation for the intelligent control and fault diagnosis of the slurry valve. This fully demonstrates the application value of digital and intelligent technologies in the field of process equipment operation and maintenance.

[0124] Step S2220, setting three-level warning thresholds for key operating parameters, wherein the three-level warning thresholds include a pressure three-level warning threshold, a temperature three-level warning threshold, and a vibration three-level warning threshold.

[0125] Further, step S2220 includes:

[0126] Step S2221, obtain the mechanical property parameters and the first thermal safety margin parameters of the slurry valve sealing material, and determine the three-level pressure warning threshold according to the mechanical property parameters and the first thermal safety margin parameters; the three-level pressure warning threshold includes the pressure warning threshold and the pressure danger threshold; the mechanical property parameters include the yield strength σ s , stress fluctuation amplitude of the sealing structure under historical load , Critical failure pressure of sealing structure and the maximum pressure under historical loads The first thermal safety margin parameter includes the maximum temperature that the sealing structure can withstand and critical failure temperature of sealing materials ;

[0127] Further, step S2221 includes:

[0128] Step S22211, obtaining the number of cycles of the sealing structure under the historical load and reference value of fatigue life of sealing materials ,according to , , and σ s , calculate the pressure warning threshold coefficient C w ; According to the yield strength σ s and pressure warning threshold coefficient C w , get the pressure warning threshold P w ;

[0129]

[0130] in,

[0131] is a correction function that takes into account the inhomogeneity of stress distribution, is the stress gradient parameter, which can be obtained through finite element analysis. The function form is ,in It is an empirical coefficient and needs to be determined by fitting a large amount of test data.

[0132] It is the stress fluctuation sensitivity coefficient, which reflects the sensitivity of the material to stress fluctuation and can be obtained through fatigue test.

[0133] is the stress fluctuation amplitude of the sealing structure under the historical load, is the yield strength of the material, and the ratio of the two reflects the degree of load fluctuation.

[0134] It is the nonlinear index of stress fluctuation and can be obtained by the slope of the stress-life curve in double logarithmic coordinates.

[0135] is the cycle number correction factor, which reflects the influence of the load cycle number on the warning threshold and can be obtained by fitting the fatigue life data.

[0136] is the number of cycles of the sealing structure under the historical load, It is the reference value of fatigue life of sealing material. The ratio of the two reflects the proportion of fatigue life consumed.

[0137] When the stress distribution inhomogeneity of the sealing structure is greater, the stress fluctuation amplitude is greater, and the number of cycles is greater, C w The larger the value, the higher the warning threshold, and vice versa. This formula takes into account factors such as material properties, load characteristics, and stress distribution, and dynamically adjusts the warning threshold to avoid excessive or insufficient warnings.

[0138] Step S22212, according to , , and , calculate the pressure danger threshold coefficient C d ; According to the yield strength σ s and pressure warning threshold coefficient C d , get the pressure warning threshold P d .

[0139]

[0140] in,

[0141] is the correction function considering material damage, is the degree of material damage, which can be characterized by nondestructive testing methods such as acoustic emission and ultrasound. The function form is ,in is the damage sensitivity coefficient, which needs to be determined by fitting the failure test data.

[0142] It is the pressure sensitivity coefficient, which reflects the sensitivity of the material to pressure and can be obtained through a burst test.

[0143] is the critical failure pressure of the sealing structure, is the maximum pressure under the historical load. The larger the difference between the two, the farther away from the failure point, and the lower the risk factor.

[0144] It is the pressure nonlinear influence coefficient, which reflects the nonlinear influence of pressure change on the danger threshold and can be obtained by fitting the failure data.

[0145] It is the temperature sensitivity coefficient, which reflects the sensitivity of material properties to temperature and can be obtained through high temperature mechanical properties tests.

[0146] The maximum temperature that the sealing structure can withstand. is the critical failure temperature of the material. The closer the ratio of the two is to 1, the closer it is to the thermal instability point and the lower the risk factor.

[0147] It is the temperature nonlinear index, which reflects the nonlinear effect of temperature change on the danger threshold and can be obtained by fitting high-temperature failure data.

[0148] When the cumulative damage of the sealing structure is greater, the maximum pressure it withstands is closer to the critical failure pressure, and the maximum temperature it withstands is closer to the critical failure temperature, The smaller the value, the lower the danger threshold, and vice versa. This formula takes into account factors such as material damage evolution and extreme working conditions, dynamically adjusts the danger threshold, and improves the accuracy of extreme warning.

[0149] Specifically, when determining the three-level warning threshold of pressure, it is necessary to comprehensively consider the mechanical properties and thermal safety margin of the slurry valve sealing material. Among them, the mechanical performance parameters reflect the deformation and failure characteristics of the sealing material under load, which is an important basis for determining the pressure warning threshold and danger threshold; while the thermal safety margin parameters reflect the performance degradation and failure risk of the sealing material in a high temperature environment, which is particularly important for high-temperature slurry working conditions.

[0150] First, the yield strength σ of the sealing material needs to be obtained through material testing or by consulting the material data manual. s , fatigue life reference value N0 and other key mechanical performance parameters. Among them, the yield strength reflects the critical stress value of the material undergoing plastic deformation, which is an important reference for determining the pressure warning threshold; the fatigue life reference value reflects the failure life of the material under cyclic load, which is different from the actual number of cycles N of the sealing structure. c Together they determine the degree of fatigue damage. In addition, it is also necessary to obtain the stress fluctuation amplitude Δσ of the sealing structure under the historical load through finite element simulation or strain gauge testing. s , and the critical failure pressure P of the sealing structure cr and the historical maximum pressure P max These parameters comprehensively reflect the stress state and safety margin of the sealing structure under actual working conditions.

[0151] Secondly, it is necessary to obtain parameters that reflect the thermal safety margin of the sealing material, including the maximum withstand temperature T of the sealing structure. max and the critical failure temperature T of the sealing material cr The maximum withstand temperature can be obtained by arranging thermocouples for real-time monitoring, while the critical failure temperature needs to be determined through high-temperature tensile, creep and other material tests. Under high temperature conditions, the strength and stiffness of the sealing material will decrease significantly, and microstructural evolutions such as oxidation and grain growth will occur inside the material, eventually leading to material failure. The critical failure temperature is an important indicator for evaluating the safety margin of materials at high temperatures.

[0152] After obtaining the above mechanical performance parameters and thermal safety margin parameters, the three-level pressure warning threshold can be determined by the following method:

[0153] For the pressure warning threshold P w The method for determining is as follows: First, according to , , and σ s , calculate the pressure warning threshold coefficient C w ; After getting C w After that, the pressure warning threshold P can be calculated w , P w =C w ×σ s ;P w It reflects the maximum pressure value that the sealing structure can withstand under the current fatigue damage level. w When the sealing structure is at risk of fatigue failure, an early warning signal needs to be issued.

[0154] For the pressure danger threshold P d The method for determining is as follows: First, according to , , and , calculate the pressure danger threshold coefficient C d ; After getting C d Then, the pressure danger threshold P can be calculated. d , P d =C d ×σ s ;P d It reflects the ultimate pressure value that the sealing structure can withstand under the current thermal safety margin. d When the sealing structure is at high risk of failure and leakage, emergency measures need to be taken in a timely manner.

[0155] Through the above steps, the three-level pressure warning thresholds (including the warning threshold Pw and the danger threshold P d ). In practical applications, the pressure value of the slurry valve can be monitored in real time and compared with the warning threshold to promptly discover safety hazards of the sealing structure. At the same time, in the process of determining the pressure threshold, the fatigue characteristics and temperature characteristics of the material are fully considered, which improves the scientificity and rationality of the threshold setting.

[0156] A significant advantage of this step is that it establishes a quantitative correlation between the pressure warning threshold and the performance parameters of the sealing material, so that the threshold setting has a clear physical meaning and theoretical basis, avoiding empiricism and arbitrariness. At the same time, by introducing fatigue damage degree and temperature degradation factors, the threshold setting process is dynamically adjusted and intelligent, which can adapt to the actual health status and working conditions of the sealing structure. This data-driven threshold optimization method can greatly improve the safety and reliability of slurry valve sealing.

[0157] Step S2222, obtaining the second thermal safety margin parameter of the sealing material, and determining the temperature three-level warning threshold according to the first thermal safety margin parameter and the second thermal safety margin parameter; the temperature three-level warning threshold includes the temperature warning threshold and the temperature danger threshold; the second thermal safety margin parameter includes the safe working temperature T0 of the sealing material, the allowable amplitude ΔT of the sealing material temperature fluctuation w and the safety margin temperature difference ΔT of the sealing material d ;

[0158] Further, step S2222 includes:

[0159] Step S22221, based on the safe operating temperature T0 and the allowable amplitude of temperature fluctuation ΔT s , calculate the temperature warning threshold T w ;

[0160] Step S22222: according to the safety margin temperature difference ΔT d and critical failure temperature , calculate the temperature danger threshold T d .

[0161] Specifically, when determining the temperature level 3 warning threshold, it is necessary to comprehensively consider the performance characteristics and degradation laws of the sealing material under different temperature conditions. First, it is necessary to obtain the first thermal safety margin parameter of the sealing material, that is, the maximum temperature T that the sealing structure can withstand. max and the critical failure temperature T of the sealing material cr These two parameters reflect the temperature load level of the sealing structure under actual working conditions and the high temperature failure characteristics of the material itself. max and T cr , the temperature safety state of the sealing structure can be determined.

[0162] Secondly, it is also necessary to obtain the second thermal safety margin parameters of the sealing material, including the safe working temperature T0 of the sealing material and the allowable amplitude ΔT of the temperature fluctuation of the sealing material. w and the safety margin temperature difference ΔT of the sealing material d The safe operating temperature T0 is the upper limit of the temperature at which the material exhibits stable performance in long-term operation, which can usually be obtained through long-term high temperature tests or by consulting the material manual. The allowable amplitude of temperature fluctuation ΔT w It reflects the sensitivity of material properties to temperature fluctuations. When the temperature fluctuation exceeds ΔT w When the material is subjected to the ΔT, the material properties may change significantly, and in extreme cases, the material may fail. w The size of is closely related to the thermal stability of the material. Safety margin temperature difference ΔT d It reflects the high temperature safety margin of the material, which is the temperature difference between the critical failure temperature of the material and the maximum temperature it can actually withstand. d The larger it is, the higher the high temperature safety margin of the material.

[0163] Based on the above two types of thermal safety margin parameters, the temperature level 3 warning threshold can be determined by the following method:

[0164] For temperature warning threshold T w The method for determining the temperature is as follows. Taking the safe working temperature T0 of the material as the benchmark, considering the allowable amplitude of temperature fluctuation ΔT w , we can get the temperature warning threshold T w :T w =T0+ΔT w ;

[0165] For example, the safe working temperature T0 of a certain polymer material is 120°C, and the allowable amplitude of temperature fluctuation ΔT w is 20℃, then its temperature warning threshold T w This indicates that when the actual temperature of the sealing structure exceeds 140°C, the performance of the material may change significantly, there is a certain risk of failure, and an early warning signal needs to be issued.

[0166] For temperature danger threshold T d The determination method is as follows. At the critical failure temperature T of the material cr Based on this, a certain safety margin temperature difference ΔT is considered d , we can get the temperature danger threshold T d :T d =T cr -ΔT d ;

[0167] Safety margin temperature difference ΔT dThe size of can be determined based on engineering experience and safety factor. d The introduction of T d The determination of provides a certain safety margin to avoid the sealing structure being cr Danger occurs when.

[0168] For example, the critical failure temperature T of a certain polymer material cr The temperature difference ΔT is 200°C, taking into account the safety margin of 20°C. d , then its temperature danger threshold T d This shows that when the actual temperature of the sealing structure exceeds 180°C, the material is at a high risk of failure and emergency measures such as shutdown and cooling need to be taken in time.

[0169] Through the above steps, the three-level temperature warning thresholds can be quantitatively calculated based on the thermal safety margin parameters of the sealing material, including the temperature warning threshold T w and temperature danger threshold T d In practical applications, temperature sensors such as thermocouples can be arranged to monitor the temperature changes of the slurry valve sealing structure in real time and compare them with the early warning threshold to detect temperature anomalies in time.

[0170] An important feature of this step is that it fully considers the dependence and sensitivity of the sealing material performance on temperature. By introducing the second thermal safety margin parameter, the temperature warning threshold is refined and dynamically adjusted. Compared with a single temperature upper limit, this hierarchical warning mechanism can more comprehensively reflect the performance status of the sealing material under different temperature conditions, which helps to achieve the advance and sensitivity of the warning signal. At the same time, the safe operating temperature T0 and the allowable amplitude of temperature fluctuation ΔT w The introduction of makes the setting of temperature warning threshold closer to the actual working state of the material, avoiding the problem of one-sided pursuit of high temperature threshold and ignoring the fluctuation of material performance. d The introduction of reflects the concept of safety redundancy in engineering design and provides additional safety protection for the setting of temperature danger thresholds. In short, through step S2222, the determination of the three-level temperature warning threshold is quantitative, refined and dynamic, which can provide more accurate and reliable criteria for the safety monitoring of the slurry valve sealing structure, and is of great significance for timely detection and prevention of temperature-induced sealing failures. At the same time, the threshold design idea based on material properties embodied in this method also provides a good reference for the health monitoring of other high-temperature and high-pressure process equipment.

[0171] Step S2223, obtaining the first vibration data of the slurry valve sealing structure, and determining the three-level vibration warning threshold according to the first vibration data; the three-level vibration warning threshold includes a vibration warning threshold and a vibration danger threshold; the first vibration data includes an average vibration level V of the slurry valve sealing structure under rated working conditions. a and critical resonance vibration velocity V cr .

[0172] Furthermore, if Figure 5 As shown, step S2223 includes:

[0173] Step S22231, setting the warning margin coefficient R w and the risk margin factor R d , according to the warning margin coefficient R w and the average vibration level V under rated conditions a , calculate the vibration warning threshold V w ;

[0174] Step S22232, according to the risk margin coefficient R d and critical resonance vibration velocity V cr , calculate the vibration danger threshold V d .

[0175] Specifically, when determining the third-level vibration warning threshold, it is necessary to obtain the vibration monitoring data of the slurry valve sealing structure under actual working conditions, that is, the first vibration data. This data reflects the vibration characteristics and evolution law of the sealing structure during long-term service. Among them, the two key parameters are the average vibration level V of the slurry valve sealing structure under rated working conditions and the average vibration level V of the slurry valve sealing structure under rated working conditions. a and critical resonance vibration velocity V cr .

[0176] Average vibration level V a Characterizes the vibration amplitude of the sealing structure under normal working conditions, which can be obtained by statistical analysis of the long-term vibration monitoring data of the slurry valve. Vibration velocity is usually used as an evaluation index because it is directly related to structural stress and fatigue damage. a It is necessary to arrange acceleration sensors at key parts of the sealing structure to collect vibration signals in real time and obtain them through data preprocessing and statistical analysis. a The size of V depends on the structural form, sealing type, medium characteristics and other factors of the slurry valve. The V of slurry valves of different models and working conditions is different. a The values ​​may vary greatly. Therefore, it is necessary to carry out special vibration monitoring tests for specific slurry valves and accumulate the first vibration data.

[0177] Critical resonance vibration speed V crIt represents the vibration velocity threshold when the sealing structure fails due to resonance. When the frequency of the external excitation is close to or coincides with the natural frequency of the structure, resonance will occur, causing the vibration amplitude to be amplified sharply and triggering structural failure. cr The determination of V needs to be obtained through theoretical analysis, finite element simulation and experimental testing. First, it is necessary to establish its vibration dynamics model based on the structural parameters and material properties of the slurry valve, and calculate the natural frequency and vibration mode; then, use numerical simulation technology to simulate the vibration response under different excitation conditions, and find out the critical excitation frequency and vibration mode that may cause resonance; finally, through vibration table tests or field tests, verify the theoretical and simulation results, make necessary corrections, and finally determine V cr The value of V cr The size of V is closely related to the structural stiffness, damping characteristics, sealing surface clearance and other factors of the slurry valve. Therefore, when conducting theoretical analysis and numerical calculations, it is necessary to model and set boundary conditions as accurately as possible to improve V. cr Reliability of predictions.

[0178] After obtaining the first vibration data, the vibration level 3 warning threshold can be determined by the following method:

[0179] For vibration warning threshold V w The determination method is as follows. Introducing the warning margin coefficient R w , considering a certain safety margin, V w Set to the average vibration level V under rated conditions a V w times, that is: V w =R w ×V a ;

[0180] Among them, R w The value can be determined based on engineering experience and reliability requirements, usually between 1.2 and 1.5. w =1.2, then V w =1.2V a This means that when the actual vibration speed exceeds 1.2 times the average level under rated conditions, the vibration state of the sealing structure enters the warning range, which requires attention and analysis of the cause. w The size of R actually reflects the sensitivity of the vibration warning. w The larger the value, the higher the warning threshold and the lower the warning sensitivity; conversely, the higher the warning sensitivity, but the false alarm rate may increase. w The value of R needs to balance safety and economy, avoiding missed reports and false reports while reducing unnecessary downtime and maintenance. Through the analysis of a large amount of vibration monitoring data and failure case statistics, R can be further optimized. wto make it more scientific and reasonable.

[0181] For vibration danger threshold V d The method for determining is as follows. Introducing the risk margin coefficient R d , considering a certain safety margin, V d Set to the critical resonance vibration speed V cr R d times, that is: V d =R d ×V cr ;

[0182] Among them, R d The value is also determined based on engineering experience and reliability requirements, usually between 0.7 and 0.9. d =0.8, then V d =0.8V cr This means that when the actual vibration speed exceeds 0.8 times the critical resonance vibration speed, the sealing structure is in an extremely dangerous state, with a high risk of resonance failure, and emergency measures must be taken immediately, such as shutting down for maintenance, replacing seals, etc. d The size of R actually reflects the severity of the vibration hazard. d The larger the value, the closer the danger threshold is to the critical value, and the more conservative the risk assessment is; conversely, the more aggressive the risk assessment is, but the best maintenance time may be delayed. d The value of also needs to balance safety and economy, preventing catastrophic consequences while avoiding excessive maintenance and waste of resources. This requires comprehensive consideration of factors such as the health status of the equipment, remaining life, and maintenance cycle, and is determined through quantitative evaluation and expert experience.

[0183] Through the above steps, the three-level vibration warning thresholds can be quantitatively calculated based on the first vibration data, including the vibration warning threshold V w and vibration danger threshold V d In practical applications, the vibration speed of the slurry valve sealing structure can be monitored in real time and compared with the warning threshold to dynamically evaluate its vibration risk level. w When the vibration speed exceeds V d A red alarm signal is issued, indicating that the equipment is on the verge of failure and that it is necessary to immediately activate the emergency plan and take emergency measures to avoid major accidents.

[0184] A notable feature of this step is the introduction of the warning margin coefficient R when determining the vibration warning threshold. w and the risk margin factor R dCompared with directly adopting empirical values ​​or absolute thresholds, this method is more flexible and adaptable. w and R d The value of can be customized according to different equipment characteristics, management strategies and risk preferences. This threshold determination method based on relative values ​​can adaptively balance the warning sensitivity and false alarm rate to meet the actual needs in different scenarios. At the same time, the vibration warning threshold is related to the average vibration level V a Hook, vibration danger threshold and critical resonance vibration speed V cr Linking makes the early warning criteria more scientific and rigorous. Compared with a single empirical value or absolute threshold, this threshold, which is closely related to the actual operating status of the equipment, can more accurately and timely reflect the health level and failure risk of the equipment.

[0185] Step S2230, establishing a working condition risk level classification rule according to the three-level warning threshold;

[0186] Further, step S2230 includes:

[0187] Step S2231, for different key operating condition parameters, corresponding operating condition parameter fuzzy linguistic variables, operating condition parameter fuzzy subsets and operating condition parameter domains are set; the operating condition parameter fuzzy linguistic variables include pressure fuzzy linguistic variables, temperature fuzzy linguistic variables and vibration fuzzy linguistic variables; the operating condition parameter fuzzy subsets include pressure fuzzy subsets, temperature fuzzy subsets and vibration fuzzy subsets; the operating condition parameter domains include temperature domains, pressure domains and vibration domains;

[0188] Specifically, when establishing the risk level classification rules for working conditions, it is first necessary to set the corresponding working condition parameter fuzzy linguistic variables, working condition parameter fuzzy subsets and working condition parameter domains for different key working condition parameters. Fuzzy linguistic variables are fuzzy concepts expressed in natural language, such as "high", "medium", "low", etc.; fuzzy subsets are the value ranges of fuzzy linguistic variables, and membership functions are used to represent different degrees of membership; the domain defines the valid value interval of the fuzzy subset, which is a set of real numbers. By introducing fuzzy theory, the working condition state can be intuitively described with linguistic variables, and uncertainty can be quantified through membership, which is suitable for expressing empirical knowledge and subjective judgment.

[0189] For the pressure parameter, the pressure fuzzy linguistic variable can be defined as {low pressure, medium pressure, high pressure}, which represents the three states of pressure level. Correspondingly, the pressure fuzzy subsets are defined as {low pressure subset, medium pressure subset, high pressure subset}, which represent the degree of membership of the pressure in the three states of low, medium and high. The pressure domain can be set to [0,P d ], that is, from zero to the pressure danger threshold P dIn this way, any pressure value can be mapped to the pressure domain, and its degree of membership to the three pressure fuzzy subsets can be determined through the membership function.

[0190] Similarly, for the temperature parameter, the temperature fuzzy linguistic variable can be defined as {low temperature, normal temperature, high temperature}, indicating the three states of temperature level. Correspondingly, the temperature fuzzy subsets are defined as {low temperature subset, normal temperature subset, high temperature subset}, indicating the degree of membership of the temperature in the three states of low, medium and high, respectively. The temperature domain can be set to [0,T d ], that is, from zero to the temperature danger threshold T d The set of real numbers.

[0191] For vibration parameters, the vibration fuzzy linguistic variables can be defined as {low vibration, medium vibration, high vibration}, which represent the three states of vibration level. Correspondingly, the vibration fuzzy subsets are defined as {low vibration subset, medium vibration subset, high vibration subset}, which represent the membership degree of vibration in the three states of low, medium and high, respectively. The vibration domain can be set to [0, V d ], that is, from zero to the vibration danger threshold V d The set of real numbers.

[0192] Through the above definition, a fuzzy logic system for characterizing the working condition is established. For any set of measured key working condition parameter values, they can be mapped to the corresponding domain, and their membership to different fuzzy subsets can be determined according to the membership function. This membership relationship can quantitatively express the fuzziness and uncertainty of the working condition parameters, providing a basis for subsequent working condition evaluation and decision-making.

[0193] An important feature of this step is the use of fuzzy theory to describe the working condition. Compared with traditional deterministic methods, fuzzy methods can better deal with the imprecision, subjectivity and empiricism in working condition evaluation. By using language variables to characterize the working condition, the cognition and judgment of equipment managers can be expressed intuitively and naturally, reducing the obstacles to knowledge representation. At the same time, the introduction of fuzzy subsets and membership functions allows the organic combination of qualitative description and quantitative calculation, which not only retains the richness of empirical knowledge, but also gives it a strict mathematical form, which is easy to implement on computers.

[0194] Step S2232, according to the operating condition parameter fuzzy linguistic variables, operating condition parameter fuzzy subsets and operating condition parameter domains corresponding to each key operating condition parameter, the operating condition parameter fuzzy subset membership function of the corresponding key operating condition parameter is set; the operating condition parameter fuzzy subset membership function includes a pressure fuzzy subset membership function, a temperature fuzzy subset membership function and a vibration fuzzy subset membership function;

[0195] Specifically, after setting fuzzy linguistic variables, fuzzy subsets and domains for different key operating parameters, it is necessary to further define the membership function of the fuzzy subset. The membership function is a function that maps elements in the domain to the interval [0,1], indicating the degree of membership of the element to the fuzzy subset. The larger the membership, the more the element conforms to the language concept represented by the fuzzy subset. Through the membership function, the correspondence between the fuzzy concept and the actual state can be quantitatively described.

[0196] Pressure fuzzy subset membership function:

[0197] Low voltage subset membership function: is 1, It decreases linearly to 0. It means that when the pressure is below the pressure warning threshold, it is completely in the low pressure state. When it is between the pressure warning threshold and the pressure danger threshold, the degree of membership decreases as the pressure increases.

[0198] Medium voltage subset membership function: is 1, and It decreases linearly to 0. It means that the pressure is completely in the medium pressure state when it is between the pressure warning threshold and the pressure danger threshold.

[0199] High-pressure subset membership function: is 1, It decreases linearly to 0. It means that the pressure is completely in a high pressure state when it is above the pressure danger threshold.

[0200] Temperature fuzzy subset membership function:

[0201] Low temperature subset membership function: is 1, Linearly decreases to 0.

[0202] Medium temperature subset membership function: is 1, and Linearly decreases to 0.

[0203] High temperature subset membership function: is 1, Linearly decreases to 0.

[0204] Vibration fuzzy subset membership function:

[0205] Low oscillator set membership function: is 1, Linearly decreases to 0.

[0206] The membership function of the oscillator set is: is 1, and Linearly decreases to 0.

[0207] High oscillator set membership function: is 1, Linearly decreases to 0.

[0208] Through the above steps, the precise working condition parameters are mapped to the membership function, and the membership relationship between the parameter value and the fuzzy subset is quantitatively described. This mapping and description method helps to introduce expert experience and qualitative knowledge into the working condition risk level classification, and improves the rationality and interpretability of the classification results.

[0209] Step S2233, constructing a fuzzy rule base based on the fuzzy subset membership function of the operating condition parameters.

[0210] Further, step S2233 includes:

[0211] Step S22331, define the operating condition risk fuzzy linguistic variables {normal, abnormal, dangerous}, the operating condition risk fuzzy subset {normal level, abnormal level, dangerous level} and the operating condition risk domain [0,1];

[0212] Step S22332, calculating the operating condition risk value based on the operating condition parameter fuzzy subset membership function;

[0213] Step S22333, defining a membership function of a corresponding operating condition risk fuzzy subset in the operating condition risk domain; the membership function of the operating condition risk fuzzy subset includes a normal level membership function, an abnormal level membership function and a dangerous level membership function;

[0214] Step S22334, obtaining the operating condition risk level according to the operating condition risk value and the membership function of the operating condition risk fuzzy subset;

[0215] Step S22335, establish fuzzy rules in the form of IF-THEN and generate a fuzzy rule base; each fuzzy rule contains an antecedent part and a consequent part, and the form of the fuzzy rule is: IF antecedent part THEN consequent part, and the operating condition risk level is written into the consequent part.

[0216] Specifically, after completing the fuzzification of key operating condition parameters, it is necessary to further build a fuzzy rule base to realize fuzzy reasoning from operating condition parameters to operating condition risk levels. The fuzzy rule base is a series of IF-THEN language rules summarized based on expert experience and engineering knowledge, which are used to describe the causal relationship between operating condition parameters and operating condition risks. Through the fuzzy rule base, qualitative operating condition assessment knowledge can be formalized and systematized to provide support for intelligent decision-making.

[0217] Step S22331 is the premise and basis for building a fuzzy rule base. First, it is necessary to define the language variables involved in the operating risk assessment. The operating risk fuzzy language variable is defined as {normal, abnormal, dangerous}, which represents the qualitative judgment of the operating risk of the slurry valve; accordingly, the operating risk fuzzy subset is defined as {normal level, abnormal level, dangerous level}, which means that the operating risk is divided into three levels. The setting of these language variables and fuzzy subsets should be consistent with engineering practice and expert experience, and can intuitively and accurately describe the risk status of the slurry valve.

[0218] While defining the linguistic variables and fuzzy subsets, it is also necessary to determine the operating condition risk domain. Considering that the risk value is a continuous variable, the operating condition risk domain can be defined as the interval [0,1], where 0 represents the lowest risk and 1 represents the highest risk. The setting of this domain helps to quantitatively evaluate and compare the risk levels under different operating conditions, and facilitates subsequent fuzzy reasoning and defuzzification processing.

[0219] Step S22332 is the definition and design of the membership function. The membership function is a function that maps the risk value to the fuzzy subset membership, which quantitatively describes the membership relationship between different risk values ​​and risk levels. For each operating condition risk fuzzy subset, it is necessary to design a corresponding membership function in the [0,1] domain. For example:

[0220] Normal level membership function: It can be defined as 1 from 0 to 0.3, indicating that the working condition is completely normal when the working condition risk value is below 0.3; it decreases linearly to 0 with the working condition risk value from 0.3 to 0.5, indicating that when the risk value is between 0.3 and 0.5, the working condition gradually deviates from the normal state; and 0 from 0.5 to 1.

[0221] Abnormal level membership function: It can be defined as 0 from 0 to 0.3; it increases linearly to 1 with the risk value from 0.3 to 0.4, indicating that when the risk value is between 0.3 and 0.4, the working condition begins to enter an abnormal state; it remains at 1 from 0.4 to 0.6, indicating that when the risk value is between 0.4 and 0.6, the working condition is in an abnormal state; it decreases linearly to 0 with the risk value from 0.6 to 0.7, indicating that when the risk value is between 0.6 and 0.7, the working condition gradually deviates from the abnormal state; it is 0 from 0.7 to 1.

[0222] Danger level membership function: It can be defined as 0 from 0 to 0.5; it increases linearly with the risk value from 0.5 to 0.7 to 1, indicating that when the risk value is between 0.5 and 0.7, the working condition begins to enter a dangerous state; it remains at 1 from 0.7 to 1, indicating that when the risk value is above 0.7, the working condition is in a dangerous state.

[0223] The setting of the above membership function comprehensively considers the size and change trend of the risk value and reasonably divides the range of different risk levels. In practical applications, the form and parameters of the membership function can also be modified and optimized based on expert experience and statistical data to improve the accuracy and reliability of risk assessment.

[0224] Steps S22333 and S22334 are the generation and application of the fuzzy rule base. Based on the defined linguistic variables, fuzzy subsets and membership functions, a series of IF-THEN fuzzy rules can be summarized to describe the correspondence between operating parameters and operating risks. The antecedent of the fuzzy rule consists of the linguistic values ​​of the key operating parameters, using AND as the conjunction; the consequent consists of the linguistic values ​​of the operating risk. For example:

[0225] Rule 1: IF pressure is high AND temperature is high AND vibration is high THEN operating risk is dangerous;

[0226] Rule 2: IF pressure is medium AND temperature is medium AND vibration is medium THEN operating risk is abnormal;

[0227] Rule 3: IF pressure is low AND temperature is low AND vibration is low THEN operating risk is normal;

[0228] The above rules focus on the comprehensive impact of the three key working condition parameters of pressure, temperature and vibration on working condition risks, reflecting the interaction between parameters. When building a fuzzy rule base, it is necessary to fully collect expert experience and historical data, cover various typical working condition combinations, and reasonably set the triggering conditions and conclusions of the rules.

[0229] Step S2240, collecting real-time operating parameters of the slurry valve sealing structure, and obtaining the real-time operating risk level according to the operating risk level classification rules.

[0230] Furthermore, if Figure 7 As shown, step S2240 includes:

[0231] Step S2241, collecting real-time operating parameters of the slurry valve sealing structure, wherein the real-time operating parameters include real-time pressure value, real-time temperature value and real-time vibration value;

[0232] Step S2242, calculating the membership of the real-time operating condition parameters to their respective fuzzy subsets, and obtaining the membership calculation result;

[0233] Step S2243, substitute the membership calculation result into the constructed fuzzy rule base to obtain the real-time working condition risk level.

[0234] Specifically, step S2240 monitors the operating parameters of the slurry valve sealing structure in real time and uses the previously established operating risk level classification rules to evaluate the operating risk level of the slurry valve sealing structure in real time, providing a basis for subsequent wear evolution prediction and sealing structure optimization.

[0235] In step S2241, the real-time operating parameters of the slurry valve sealing structure are collected, and the real-time operating parameters include real-time pressure value, real-time temperature value and real-time vibration value. In this step, it is necessary to use the pressure, flow, temperature and vibration sensors installed at the inlet and outlet of the slurry valve and the sealing cavity to collect the operating parameters of the slurry valve sealing structure in real time. Among them, the pressure sensor is used to measure the pressure difference of the inlet and outlet of the slurry valve, the flow sensor is used to measure the slurry flow, the temperature sensor is used to measure the temperature in the sealing cavity, and the vibration sensor is used to measure the vibration level of the sealing structure. The data collected by these sensors will be used as real-time operating parameters for subsequent operating risk level assessment.

[0236] In step S2242, the membership of the real-time operating condition parameters to their respective fuzzy subsets is calculated to obtain the membership calculation results. In this step, the collected real-time operating condition parameters need to be substituted into the previously established pressure, temperature and vibration fuzzy subset membership functions to calculate the membership of the real-time operating condition parameters to their respective fuzzy subsets. For example, the real-time pressure value is substituted into the pressure fuzzy subset membership function to calculate the membership of the real-time pressure value to the three fuzzy subsets of "low-level pressure", "intermediate pressure" and "high-level pressure"; similarly, the membership of the real-time temperature value and the real-time vibration value to their respective fuzzy subsets is calculated. Through this step, the membership calculation results of the real-time operating condition parameters to their respective fuzzy subsets can be obtained, providing input for the subsequent operating condition risk level assessment.

[0237] In step S2243, the membership calculation result is substituted into the constructed fuzzy rule base to obtain the real-time working condition risk level. In this step, it is necessary to substitute the membership calculation result obtained in step S2242 into the previously constructed fuzzy rule base, and obtain the real-time working condition risk level through fuzzy reasoning. Specifically, the corresponding rules in the fuzzy rule base are triggered according to the membership calculation result to obtain the working condition risk value, and then the fuzzy subset to which the real-time working condition risk belongs, that is, the real-time working condition risk level, is determined according to the working condition risk value and the working condition risk fuzzy subset membership function. For example, if the membership calculation result triggers the rule "IF the pressure is high AND the temperature is high AND the vibration is high THEN the working condition risk is dangerous", the working condition risk value is 1, and according to the dangerous level membership function, the real-time working condition risk level is determined to be "dangerous level".

[0238] Step S2240 can grasp the health status of the slurry valve sealing structure in real time and discover potential risks and hidden faults in time by collecting operating parameters in real time and evaluating the operating risk level. By comprehensively considering multiple operating parameters such as pressure, temperature and vibration, the operating risk of the slurry valve sealing structure can be comprehensively evaluated to avoid the one-sidedness of single parameter evaluation. Through fuzzy logic reasoning, the qualitative operating risk description is quantified into a specific risk level, which is convenient for subsequent risk management and decision support. By constructing a fuzzy rule base, the operating risk level can be adaptively evaluated according to the combination of different operating parameters, thereby improving the flexibility and adaptability of risk assessment.

[0239] For example, suppose that the real-time working condition parameters of the slurry valve sealing structure collected at a certain moment are: pressure is 8MPa, temperature is 200℃, and vibration speed is 30mm / s. According to the previously established fuzzy subset membership function, the membership of the real-time working condition parameters to their respective fuzzy subsets is calculated as follows: pressure to "high-level pressure" membership is 0.8, temperature to "high-level temperature" membership is 0.6, and vibration to "high-level vibration" membership is 0.7. Substitute the membership calculation results into the fuzzy rule base, trigger the rule "IF pressure is high AND temperature is high AND vibration is high THEN working condition risk is dangerous", and obtain the working condition risk value of 0.9. According to the dangerous level membership function, the real-time working condition risk level of the slurry valve sealing structure at this moment is determined to be "dangerous level". Based on this, it can be judged that the current working condition risk of the slurry valve sealing structure is very high, and corresponding measures need to be taken to reduce the risk, such as reducing working condition parameters, increasing monitoring frequency, and preparing spare parts.

[0240] To sum up, through step S2240, a quantitative assessment of the real-time operating risk of the slurry valve sealing structure can be achieved, which provides an important basis for subsequent wear evolution prediction and sealing structure optimization, and is of great significance to improving the reliability and safety of the slurry valve sealing structure.

[0241] Step S3000: Based on the wear morphology time series data set and the real-time operating condition risk level, a wear evolution prediction model is constructed, and combined with the digital twin model of the slurry valve sealing structure, the key geometric parameters of the sealing structure are optimized to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

[0242] Furthermore, step S3000 includes:

[0243] Step S3100, dividing the wear morphology time series data set into a historical wear morphology time series data set and a real-time wear morphology time series data set; constructing a wear evolution prediction model based on the historical wear morphology time series data set and the real-time working condition risk level;

[0244] Further, step S3100 includes:

[0245] Step S3110, constructing a training sample set for a wear evolution prediction model;

[0246] Furthermore, if Figure 8 As shown, step S3110 includes:

[0247] Step S3111, selecting a time window, and extracting a continuous statistical feature parameter subsequence from the historical wear morphology time series data set according to the selected time window as an input feature vector;

[0248] Step S3112, using the statistical characteristic parameter at the next moment corresponding to the input feature vector as the output target value, constructing an input-output pair to form a training sample;

[0249] Step S3113, traverse the entire historical wear morphology time series data set through a sliding time window to generate a complete training sample set.

[0250] Specifically, the wear morphology time series data set is divided into a historical wear morphology time series data set and a real-time wear morphology time series data set. The purpose of this division is to distinguish the wear data that has been collected from the wear data monitored in real time, so that they can be used for the training and application of the wear evolution prediction model. The historical wear morphology time series data set contains the wear state change data of the slurry valve within a period of time after it is put into use, reflecting the development law and trend of wear under different working conditions. This part of the data is mainly used to construct the training sample set of the wear evolution prediction model. By selecting appropriate time windows and feature indicators, input and output pairs are extracted from the historical data set to form training samples, which are used to train and optimize the parameters of the prediction model. By training on the historical data set, the prediction model can learn the general laws and trends of wear evolution. The real-time wear morphology time series data set corresponds to the wear monitoring data at the current moment and in the recent period, reflecting the latest wear state of the slurry valve sealing structure. This part of the data is mainly used for the application and verification of the prediction model. By inputting real-time data into the trained prediction model, the wear morphology prediction results for a period of time in the future can be obtained. By integrating the prediction results with the current digital twin model, the geometric shape and wear distribution of the slurry valve sealing structure can be dynamically updated, and its wear status can be predicted and monitored online. At the same time, by comparing the prediction results with the subsequent actual wear monitoring data, the accuracy and reliability of the prediction model can be evaluated, and a basis can be provided for further optimization of the model.

[0251] The purpose of step S3110 is to extract representative feature parameter sequences from the historical wear morphology time series data set, and construct them into the form of input-output pairs to form a complete training sample set to prepare for subsequent prediction model training. Select a time window to extract a continuous statistical feature parameter subsequence from the historical wear morphology time series data set as the input feature vector. In this step, it is necessary to preprocess the three-dimensional morphology feature parameter sequence and geometric profile change sequence in the wear morphology time series data set, such as denoising, smoothing, normalization, etc., to improve data quality and reliability. Then, select a time window of fixed length, such as 10 time points, and extract a continuous statistical feature parameter subsequence from the preprocessed wear morphology time series data set as the input feature vector of the model. The statistical feature parameters here can be indicators that can quantitatively characterize the wear state, such as average roughness, maximum profile depth, wear volume, etc. By sliding the time window, multiple input feature vectors can be extracted over the entire time series.

[0252] The statistical characteristic parameters of the next moment corresponding to the input feature vector are used as the output target value to construct an input-output pair to form a training sample. In this step, for each extracted input feature vector, the corresponding statistical characteristic parameters of the next moment are used as the output target value to construct an input-output pair to form a training sample. Each training sample here actually represents the mapping relationship from a wear state to the wear state at the next moment. In this way, the time series data can be converted into a supervised learning problem, laying the foundation for the training of the prediction model. By sliding the time window, the entire historical wear morphology time series data set is traversed to generate a complete training sample set. In this step, steps S3111 and S3112 are repeated on the entire historical wear morphology time series data set by sliding the time window to generate a complete training sample set. In order to improve the generalization performance of the model, the generated training sample set can be randomly divided into a training subset and a validation subset, which are used for model training and performance evaluation respectively.

[0253] Step S3110 can effectively capture the time series characteristics and trend information in the wear evolution process by selecting a time window to extract a continuous statistical feature parameter subsequence, and provide a richer feature representation for the prediction model. By preprocessing the original time series data such as denoising, smoothing, and normalization, the noise and outliers in the data can be effectively eliminated, the quality and reliability of the data can be improved, and the stability of model training can be enhanced. By constructing input-output pairs, the time series prediction problem is converted into a supervised learning problem, and mature machine learning algorithms and frameworks can be used for model training and optimization to improve training efficiency and prediction accuracy. By sliding the time window, multiple training samples can be generated from limited historical data, expanding the scale of the training data set, and improving the generalization ability and robustness of the model.

[0254] Exemplarily, for example, assume that the historical wear morphology time series data set contains 100 time points, and each time point contains three statistical characteristic parameters: average roughness, maximum profile depth, and wear volume. If the time window length is selected as 10, the input feature vectors can be extracted from 91 time windows, 1-10, 2-11, ..., 91-100, and each feature vector contains 30 elements (10 time points x 3 feature parameters). Then, the statistical characteristic parameters of the 11th time point corresponding to each feature vector are used as the output target value to construct 91 input-output pairs to form a training sample set. Next, the training sample set is randomly divided into a training set and a validation set for model training and performance evaluation.

[0255] Step S3120, taking the real-time operating condition risk level as the attribute label of the training sample and integrating it into the training sample set;

[0256] Specifically, the purpose of step S3120 is to introduce the operating condition risk information into the wear evolution prediction model, so that the model can learn and characterize the difference laws of wear evolution according to different operating condition risk levels, and improve the pertinence and accuracy of the prediction. In the process of generating the training sample set, for each training sample, according to the real-time operating condition parameters at the corresponding time point, the previously constructed operating condition risk level classification rules are used to obtain the real-time operating condition risk level at that time point, and use it as the attribute label of the training sample. The real-time operating condition risk level here can be a discrete value such as "normal", "abnormal", "dangerous", or a continuous value between 0-1, indicating the severity of the operating condition risk. By integrating the real-time operating condition risk level into the training sample set, the operating condition information can be associated with the wear status information, so that the prediction model can learn and characterize the influence of the operating condition risk on the wear evolution.

[0257] By introducing the real-time operating risk level, the prediction model can learn and characterize the differential laws of wear evolution under different operating conditions, and improve the pertinence and accuracy of the prediction. For example, under high-risk conditions, the wear rate may be significantly accelerated, while under low-risk conditions, the wear rate is relatively mild. By learning these differential laws, the prediction model can give a more accurate prediction of wear evolution according to the real-time operating risk level. Using the real-time operating risk level as the attribute label of the training sample can enable the prediction model to have the ability to adapt to the working conditions, and can dynamically adjust the results of wear evolution prediction according to the changes in the real-time working conditions. This adaptive ability can improve the practicality and robustness of the prediction model, enabling it to cope with complex and changeable actual working conditions. By incorporating working condition information, the interpretability and explainability of the wear evolution prediction model can be enhanced. The prediction results not only reflect the changing trend of the wear state, but also reveal the degree of influence of the working condition risk on wear, providing richer information support for equipment maintenance and operation and maintenance decisions.

[0258] For example, suppose that at a certain point in time, the real-time operating parameters of the slurry valve are: pressure 8MPa, temperature 200℃, vibration speed 30mm / s. According to the previously constructed operating risk level classification rules, the real-time operating risk level at this point in time is determined to be "dangerous". When generating the training sample corresponding to this time point, "dangerous" is used as the attribute label of the sample, and together with the input feature vector, a complete training sample is formed. In this way, the wear evolution law under different operating risk levels can be introduced into the training of the prediction model, so that the model can learn and characterize the impact of operating risk on wear.

[0259] Step S3130, select the long short-term memory network LSTM model, and use the training sample set incorporating attribute labels to train multiple wear evolution prediction sub-models according to different real-time operating condition risk levels.

[0260] Specifically, the purpose of step S3130 is to utilize the time series modeling capability and gating mechanism of the LSTM model to train multiple specialized wear evolution prediction sub-models for different operating risk levels, thereby improving the accuracy and applicability of the prediction. The LSTM model is a special recurrent neural network that can effectively process time series data and mine long-term dependencies in the data. This is highly consistent with the time series characteristics of the wear evolution of the slurry valve. The LSTM model contains three gating mechanisms: input gate, forget gate, and output gate, which can selectively memorize and forget information, thereby better characterizing the nonlinear and non-stationary characteristics of the wear evolution process. By introducing different real-time operating risk levels, multiple targeted prediction sub-models can be trained. Each sub-model learns and characterizes the wear evolution law under a specific operating risk level.

[0261] The specific model training process is as follows:

[0262] 1. Group the training sample set according to the real-time working condition risk level to obtain sub-training sets corresponding to multiple risk levels.

[0263] 2. For each sub-training set corresponding to each risk level, use cross-validation to optimize the hyperparameters of the LSTM model, such as the number of hidden layers, number of hidden units, learning rate, batch size, etc., to obtain the model parameter configuration with the best performance.

[0264] 3. Using the optimal parameter configuration, train the LSTM model on the sub-training set corresponding to each risk level to obtain the wear evolution prediction sub-model for that risk level.

[0265] 4. Combine the prediction sub-models corresponding to all risk levels into a complete wear evolution prediction model. According to the real-time working condition risk level, select the corresponding sub-model to give the wear evolution prediction result.

[0266] By training a special prediction sub-model for different working condition risk levels, the wear evolution law under each working condition can be better learned and characterized, and the prediction accuracy can be improved. Compared with using a single prediction model, this method can better handle the differences in working conditions and give more accurate and reliable prediction results. By training multiple prediction sub-models, the scope of application of the wear evolution prediction model can be expanded, so that it can meet the wear prediction needs under different working conditions. Regardless of how the real-time working condition risk level changes, the corresponding sub-model can be selected to give the wear prediction results suitable for the current working condition, enhancing the practicality and robustness of the model. Decomposing the entire wear evolution prediction problem into multiple sub-problems and training a special prediction sub-model for each sub-problem can reduce the complexity of each sub-model and improve the computational efficiency of training and prediction. Compared with using a single large and complex model, this divide-and-conquer approach can more efficiently utilize computing resources and achieve real-time wear evolution prediction. By training multiple prediction sub-models, the characteristics and laws of wear evolution under different working conditions can be better understood and explained. The training results of each sub-model can reveal the influencing factors and evolution trends of wear under specific working conditions, providing richer and more targeted information support for equipment maintenance and operation and maintenance decisions.

[0267] For example, assuming that the real-time operating risk level is divided into three levels: "normal", "abnormal", and "dangerous", the training sample set can be grouped according to these three levels to obtain three sub-training sets. Then, for each sub-training set, the hyperparameters of the LSTM model, such as the number of hidden layers, the number of hidden units, the learning rate, the batch size, etc., are optimized using cross-validation to obtain the model parameter configuration with the best performance. Next, using the optimal parameter configuration, the LSTM model is trained on each sub-training set to obtain three prediction sub-models, one for each of the three operating risk levels: "normal", "abnormal", and "dangerous". In actual prediction, the corresponding sub-model is selected according to the current real-time operating risk level to give the wear evolution prediction result.

[0268] In summary, through step S3100, a wear evolution prediction model that is adaptive to the working condition can be constructed, which can give accurate and reliable wear prediction results for different working condition risk levels, and provide strong decision support for the maintenance and operation of the slurry valve. This prediction method makes full use of historical wear data and real-time working condition information, comprehensively considers the time series characteristics of wear evolution and working condition differences, and has good practicality and promotion value.

[0269] Step S3200, obtaining predicted wear morphology data according to the real-time wear morphology time series data set and the constructed wear evolution prediction model; fusing the predicted wear morphology data with the digital twin model of the slurry valve sealing structure, dynamically updating the digital twin model of the slurry valve sealing structure, and obtaining an updated digital twin model;

[0270] The purpose of step S3200 is to use real-time monitoring data and prediction models to predict and update the wear state of the slurry valve sealing structure online, and to build a dynamic digital twin model that can reflect its actual service status. By integrating the prediction results with the current digital twin model, the wear state changes of the slurry valve sealing structure can be presented in real time in the virtual space, providing a basis for subsequent mechanical performance analysis and structural optimization.

[0271] Further, step S3200 includes:

[0272] Step S3210, obtaining measured wear morphology data, fusing the predicted wear morphology data with the measured wear morphology data, mapping them into the three-dimensional topological base model of the slurry valve sealing structure, and obtaining an updated digital twin model;

[0273] Specifically, it is necessary to first perform spatiotemporal registration of the measured wear morphology data and the predicted wear morphology data to ensure the consistency of the two types of data in time and space. The registration method can use algorithms such as feature point matching and iterative nearest point to unify data from different sources into the same coordinate system. Then, the registered measured data and predicted data are weighted fused, and the fusion weight can be determined according to the confidence or uncertainty of the two types of data. For example, algorithms such as Kalman filtering can be used to dynamically adjust the fusion weight of the measured data and the predicted data by establishing a dynamic model and an observation model of the wear state, so that the fusion result can take into account the advantages of the two types of data, which can not only reflect the actual wear state, but also predict the future wear trend. Finally, the fused wear morphology data is converted into a three-dimensional digital representation such as a triangular face mesh or point cloud, and mapped to the three-dimensional topological base model of the slurry valve sealing structure through mesh deformation or reconstruction techniques to obtain dynamically updated geometric shapes and wear distribution. The updated three-dimensional topological base model is recombined with the mechanical properties simulation model to form a digital twin model, which is used to reflect the wear state changes of the slurry valve sealing structure under actual working conditions.

[0274] Step S3220, calculating the mechanical performance index of the sealing structure based on the updated digital twin model;

[0275] Specifically, the updated digital twin model is used to recalculate and evaluate the stress state and sealing performance of the slurry valve sealing structure. First, based on the updated three-dimensional topological base model, the finite element mesh model is regenerated, and the updated material property parameters, boundary conditions and load conditions are input into the mesh model. Then, multi-physics field simulation analysis such as statics, dynamics or fluid-solid coupling is carried out to simulate the stress response and sealing performance of the slurry valve sealing structure under the current wear state. Key mechanical performance indicators are extracted from the simulation results, such as the contact stress distribution of the sealing surface and the leakage of the sealing gap, to quantitatively evaluate the service performance of the sealing structure under the wear state. The calculation of mechanical performance indicators can be achieved with the help of commercial finite element software or independently developed numerical calculation programs. By establishing a quantitative correlation between the wear state and the mechanical properties, the performance degradation caused by wear can be discovered in time, providing a basis for maintenance decisions of the slurry valve.

[0276] Step S3230, judging whether the mechanical performance index meets the design requirements, if not, triggering the parameter optimization process of the sealing structure and executing step S3300;

[0277] Specifically, the calculated mechanical performance index is compared with the design requirements to determine whether the sealing structure meets the requirements for safe and reliable operation under the current wear state. The design requirements can be set from the following aspects: (1) Contact stress: The maximum contact stress of the sealing surface should not exceed the allowable contact stress of the material to avoid local plastic deformation or fatigue damage; (2) Leakage: The leakage of the sealing gap should not exceed the design allowable value to ensure the sealing performance and system efficiency of the slurry valve; (3) Service life: Based on the wear evolution prediction model, the remaining service life of the sealing structure is extrapolated and estimated, which should meet the shortest service life required by the design; (4) Safety margin: Considering factors such as material performance discreteness, operating condition fluctuations and model uncertainty, the mechanical performance index of the sealing structure should have sufficient safety margin. If the mechanical performance index does not meet any of the above design requirements, the sealing structure parameter optimization process of step S3300 is automatically triggered to adjust and optimize the key geometric parameters of the sealing structure in order to improve its mechanical properties and extend its service life.

[0278] Step S3200 can realize the online prediction of the wear state of the sealing structure of the slurry valve, the dynamic update of the digital twin model, and the real-time evaluation of the mechanical properties, forming a closed-loop state monitoring and performance optimization process. This method based on digital twins has the following advantages: (1) By integrating the measured data and the predicted data, the digital twin model can more realistically reflect the actual wear state of the sealing structure and improve the accuracy of state monitoring; (2) Through the quantitative correlation between the wear state and the mechanical properties, the performance degradation problems caused by wear can be discovered in time, providing a basis for maintenance decisions; (3) By comparing the mechanical performance indicators with the design requirements, it can be determined whether the sealing structure meets the requirements of safe and reliable operation, and potential risks and hidden faults can be discovered; (4) Through the parameter optimization process, the sealing structure design can be adjusted and improved online to extend its service life and improve the reliability and economy of the slurry valve.

[0279] For example, suppose that after a slurry valve sealing structure has been put into use for a period of time, it is found through real-time monitoring that its sealing surface wear has intensified and the leakage has exceeded the design allowable value. At this time, the measured wear morphology data can be input into the wear evolution prediction model to obtain the wear development trend in the future. The prediction results are integrated with the current digital twin model of the sealing structure to update the geometry and wear distribution. Based on the updated digital twin model, the mechanical performance indicators such as contact stress and leakage of the sealing structure are recalculated. The results show that with the intensification of wear, the uneven distribution of contact stress on the sealing surface increases, the local stress peak exceeds the allowable stress of the material, and the leakage further increases, which does not meet the design requirements. According to these evaluation results, the sealing structure parameter optimization process is automatically triggered. By optimizing key geometric parameters such as the width and inclination angle of the sealing surface, the contact stress distribution is more uniform and the leakage is reduced to within the allowable value. The optimized sealing structure design scheme is verified in the digital twin model to ensure that its mechanical properties meet the requirements, and then applied to the actual slurry valve sealing structure to guide its maintenance and transformation.

[0280] In summary, the digital twin-driven online monitoring and optimization process of the sealing structure formed by step S3200 can realize the real-time prediction and update of the wear state of the slurry valve sealing structure, the quantitative evaluation and feedback of the sealing performance, and the online optimization and improvement of the sealing structure design. This process makes full use of the advantages of digital twin technology, interactively integrating the actual physical system with the virtual digital model to form a closed-loop state monitoring and performance optimization mechanism. By coupling and integrating technologies such as wear evolution prediction, digital twin modeling, mechanical performance analysis, and parameter optimization, the reliability, safety, and economy of the slurry valve sealing structure can be significantly improved, providing strong support for its full life cycle management.

[0281] Step S3300: Based on the updated digital twin model, optimize the key geometric parameters of the sealing structure to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

[0282] The purpose of this step is to find the optimal combination of key geometric parameters of the sealing structure under given constraints through the method of optimization design, so that the mechanical performance indicators of the sealing structure are optimized and meet the design requirements. Among them, the key geometric parameters of the sealing structure mainly include the sealing surface width, the sealing surface inclination angle, etc. The values ​​of these parameters will directly affect the contact stress distribution and sealing performance of the sealing structure. By optimizing parameters in the digital twin model, the performance of different parameter combinations can be quickly evaluated in a virtual environment, avoiding the high cost and long cycle of physical experiments, and improving design efficiency and success rate.

[0283] Furthermore, if Fig. 9 As shown, step S3300 includes:

[0284] Step S3310, setting key geometric parameters of the sealing structure of the slurry valve as optimized design variables, and establishing an optimized design model for the sealing structure;

[0285] Step S3320, setting the mechanical performance index of the sealing structure as the optimization objective function, and setting the constraint conditions according to the design requirements to improve the sealing structure optimization design model;

[0286] Step S3330, solving the sealing structure optimization design model to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

[0287] Specifically, in step S3310, it is necessary to select key geometric parameters that have a greater impact on the sealing performance as optimization design variables based on the parametric geometric model of the sealing structure. These variables can be continuous, such as the sealing surface width, inclination angle, etc.; they can also be discrete, such as the number of sealing teeth, material type, etc. These variables and their value ranges are encoded into the optimization design vector to form the decision space of the optimization problem. At the same time, it is necessary to convert the mechanical performance indicators of the sealing structure, such as contact stress, leakage, etc., and various constraints, such as strength, stiffness, geometric constraints, etc., into mathematical expressions to construct an optimization design model for the sealing structure. The optimization design model of the sealing structure usually includes three parts: objective function, design variables and constraints.

[0288] In step S3320, it is necessary to further refine and quantify the various components of the sealing structure optimization design model. First, the contact stress minimization and leakage minimization are set as two optimization objectives to form a multi-objective optimization problem. Since there may be contradictions and conflicts between the two objectives, they need to be weighed and compromised. A common processing method is to non-dimensionalize multiple optimization objectives, eliminate dimensional differences, and convert them into normalized fitness functions. Then, a weight coefficient is introduced, and a weighted sum is performed on multiple fitness functions to obtain a comprehensive fitness function as the final objective function of the optimization design model. In this way, the multi-objective optimization problem can be converted into a single-objective optimization problem, which is easy to solve. Secondly, according to the design requirements, the value range of the optimization design variables is constrained to ensure the manufacturability and assemblability of the optimization results. For example, the sealing surface width cannot be too small, otherwise it will cause excessive wear of the sealing surface; it cannot be too large, otherwise it will increase the volume and cost of the sealing structure. Thirdly, the difference between the mechanical performance index value of the slurry valve sealing structure and the design requirement value is constructed as an inequality constraint condition and incorporated into the optimization design model. For example, it can be required that the peak value of contact stress does not exceed the allowable stress of the material, the leakage does not exceed the design allowable value, etc. By reasonably setting the objective function, design variables and constraints, a complete, accurate and solvable sealing structure optimization design model can be established.

[0289] Step S3330 solves the sealing structure optimization design model to obtain the optimal geometric parameter combination of the slurry valve sealing structure. In this step, it is necessary to select an appropriate optimization algorithm to solve the established sealing structure optimization design model to obtain the optimal sealing structure geometric parameter combination. Commonly used optimization algorithms can be divided into two categories: one is swarm intelligence optimization algorithms, such as genetic algorithms, particle swarm algorithms, etc., which simulate biological evolution or group behavior, perform global searches in the design space, and find the optimal solution; the other is traditional numerical optimization algorithms, such as sequential quadratic programming (SQP), interior point method (IPM), etc., which perform local searches in the design space through iterative approximation or gradient descent to find the optimal solution. These two types of algorithms have their own advantages and disadvantages, and the appropriate algorithm can be selected according to the characteristics of the optimization problem and the solution requirements. Taking genetic algorithm as an example, the solution steps are as follows: first, the optimization design variables are discretely encoded to generate the initial optimization iterative population, and the objective function value and constraint condition value of each individual are calculated; then, according to the fitness function, the population is selected, crossed and mutated to generate a new generation of population; the iteration is repeated until the termination conditions such as the maximum number of iterations or the convergence of the objective function value are met, and the optimal individual, that is, the optimal geometric parameter combination, is output; finally, the optimal geometric parameter combination is post-processed, such as rounding, adjustment, etc., to ensure that it meets the discrete requirements of actual processing and manufacturing.

[0290] Step S3300 can quickly evaluate a large number of design schemes by optimizing parameters in a virtual environment, avoiding the high cost and long cycle of physical experiments, greatly shortening the design cycle, and improving design efficiency. By combining the optimization design with the digital twin model, the mechanical behavior and failure mechanism of the sealing structure under actual working conditions can be fully considered, the accuracy and reliability of the optimization results can be improved, and the number of design iterations can be reduced. Through multi-objective optimization and multi-disciplinary constraints, multiple performance indicators and multiple constraints of the sealing structure can be taken into account, and the coordinated optimization of multi-disciplinary factors such as structure, material, and process can be achieved to obtain the optimal design scheme for the system. By exploring and searching in a broader design space, the optimization algorithm can break out of the limitations of conventional design ideas, discover novel and superior design schemes, and achieve design innovation and breakthroughs.

[0291] For example, for a certain type of slurry seal structure, the optimized design variable is the sealing surface width (scope ) and the sealing surface inclination angle (scope ), the optimization goal is to minimize the contact stress and minimize leakage , the constraint condition is that the peak contact stress does not exceed the allowable stress of the material The leakage does not exceed the design allowable value The optimization design model is constructed as follows:

[0292] Optimization goal: ;

[0293] Subject to constraints: ;

[0294]

[0295]

[0296]

[0297] in, and They were calculated by finite element analysis and fluid dynamics analysis in the digital twin model. The model was solved by genetic algorithm, with a population size of 50, an evolutionary generation of 100, a crossover probability of 0.8, and a mutation probability of 0.1. The calculation results show that when the sealing surface width is 15mm and the sealing surface inclination angle is 3°, the contact stress peak and leakage are the smallest, which are reduced by 20% and 30% respectively, meeting the design requirements. The parameter combination was verified by test, and the results were in good agreement with the prediction of the digital twin model, proving the effectiveness of the optimization design.

[0298] In summary, the digital twin-driven sealing structure parameter optimization design method formed by step S3300 can realize an integrated design process from demand analysis, conceptual design to detailed design and optimization design throughout the entire life cycle of the sealing structure, and is closely integrated with manufacturing, testing, and service to form a closed-loop iteration of design, simulation, optimization, and verification, which greatly improves the performance level and development quality of the slurry valve sealing structure and has broad application prospects. This method has certain demonstration and guiding significance for the design of other types of mechanical sealing structures.

[0299] Example 2

[0300] This embodiment provides a large-diameter slurry valve sealing structure design device suitable for high-wear working conditions on the basis of embodiment 1, such as Fig. 9 As shown, including:

[0301] Digital twin model construction module: used to obtain the key geometric parameters of the slurry valve sealing structure, and construct a three-dimensional topological base model of the slurry valve sealing structure based on the key geometric parameters; construct a mechanical performance simulation model of the slurry valve sealing structure based on the three-dimensional topological base model; fuse and associate the three-dimensional topological base model of the slurry valve sealing structure with the mechanical performance simulation model to construct a digital twin model of the slurry valve sealing structure;

[0302] Multi-scale data acquisition module: used to obtain the multi-scale wear morphology image time series of the slurry valve sealing structure, and obtain the wear morphology time series data set according to the multi-scale wear morphology image time series;

[0303] Working condition risk classification module: used to obtain key working condition parameters of the slurry valve, set three-level warning thresholds for the key working condition parameters, and establish working condition risk level classification rules based on the three-level warning thresholds; collect the real-time working condition parameters of the slurry valve sealing structure, and obtain the real-time working condition risk level based on the working condition risk level classification rules;

[0304] Geometric parameter optimization module: Based on the wear morphology time series data set and the real-time operating condition risk level, a wear evolution prediction model is constructed. Combined with the digital twin model of the slurry valve sealing structure, the key geometric parameters of the sealing structure are optimized to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

[0305] In the digital twin model construction module, the mechanical performance simulation model of the slurry valve sealing structure based on the three-dimensional topological base model includes:

[0306] Step S1210, meshing the three-dimensional topological base model to generate a preliminary finite element model;

[0307] Step S1220, defining material property parameters of the preliminary finite element model to obtain a material property enabled finite element model;

[0308] Step S1230, setting the boundary conditions and current load conditions of the material property-enabled finite element model to obtain a mechanical property simulation model of the slurry valve sealing structure.

[0309] In the digital twin model construction module, the digital twin model of the slurry valve sealing structure is constructed including:

[0310] Step S1310, performing finite element solution calculation on the mechanical performance simulation model of the slurry valve sealing structure to obtain the stress and strain distribution results of the slurry valve sealing structure under the current working condition;

[0311] Step S1320, mapping the stress-strain distribution results obtained by finite element solution calculation to the three-dimensional topological base model of the slurry valve sealing structure to obtain a digital twin model of the slurry valve sealing structure.

[0312] In the multi-scale data acquisition module, the wear morphology time series data set is obtained including:

[0313] Step S2110, selecting a key area of ​​the sealing surface of the slurry valve;

[0314] Step S2120, performing in-situ characterization tests on key areas of the slurry valve sealing surface at different time points to obtain a multi-scale wear morphology image time series;

[0315] Step S2130, digitally processing the multi-scale wear morphology image time series to obtain a wear morphology time series data set;

[0316] The step S2130 includes:

[0317] Step S2131, calculating the surface roughness parameters of the three-dimensional topography image time series to obtain a three-dimensional topography feature parameter sequence;

[0318] Step S2132, performing difference calculation on the nanoscale topography image time series to obtain a geometric profile variation sequence;

[0319] Step S2133, the three-dimensional morphological feature parameter sequence and the geometric profile variation sequence are merged to obtain a wear morphology time series data set.

[0320] In the operating condition risk classification module, the three-level warning thresholds for key operating condition parameters are set as follows:

[0321] Step S2210, obtaining key operating parameters of the slurry valve, wherein the key operating parameters include pressure, temperature and vibration;

[0322] Step S2220, setting three-level warning thresholds for key operating parameters, wherein the three-level warning thresholds include a pressure three-level warning threshold, a temperature three-level warning threshold, and a vibration three-level warning threshold.

[0323] The step S2220 includes:

[0324] Step S2221, obtain the mechanical property parameters and the first thermal safety margin parameters of the slurry valve sealing material, and determine the three-level pressure warning threshold according to the mechanical property parameters and the first thermal safety margin parameters; the three-level pressure warning threshold includes the pressure warning threshold and the pressure danger threshold; the mechanical property parameters include the yield strength σ s , stress fluctuation amplitude of the sealing structure under historical load , Critical failure pressure of sealing structure and the maximum pressure under historical loads The first thermal safety margin parameter includes the maximum temperature that the sealing structure can withstand and critical failure temperature of sealing materials ;

[0325] Step S2222, obtaining the second thermal safety margin parameter of the sealing material, and determining the temperature three-level warning threshold according to the first thermal safety margin parameter and the second thermal safety margin parameter; the temperature three-level warning threshold includes the temperature warning threshold and the temperature danger threshold; the second thermal safety margin parameter includes the safe working temperature T0 of the sealing material, the allowable amplitude ΔT of the sealing material temperature fluctuation w and the safety margin temperature difference ΔT of the sealing material d ;

[0326] Step S2223, obtaining the first vibration data of the slurry valve sealing structure, and determining the three-level vibration warning threshold according to the first vibration data; the three-level vibration warning threshold includes a vibration warning threshold and a vibration danger threshold; the first vibration data includes an average vibration level V of the slurry valve sealing structure under rated working conditions. a and critical resonance vibration velocity V cr .

[0327] The step S2221 includes:

[0328] Step S22211, obtaining the number of cycles of the sealing structure under the historical load And the fatigue life reference value of sealing materials ,according to , , and σ s , calculate the pressure warning threshold coefficient C w ; According to the yield strength σ s and pressure warning threshold coefficient C w , get the pressure warning threshold P w ;

[0329] Step S22212, according to , , and , calculate the pressure danger threshold coefficient C d ; According to the yield strength σ s and pressure warning threshold coefficient C d , get the pressure warning threshold P d .

[0330] The step S2222 includes:

[0331] Step S22221, based on the safe operating temperature T0 and the allowable amplitude of temperature fluctuation ΔT s , calculate the temperature warning threshold T w ;

[0332] Step S22222: according to the safety margin temperature difference ΔT d and critical failure temperature , calculate the temperature danger threshold T d .

[0333] The step S2223 includes:

[0334] Step S22231, setting the warning margin coefficient R w and the risk margin factor R d , according to the warning margin coefficient R w and the average vibration level V under rated conditions a , calculate the vibration warning threshold V w ;

[0335] Step S22232, according to the risk margin coefficient R d and critical resonance vibration velocity V cr , calculate the vibration danger threshold V d .

[0336] In the working condition risk classification module, the working condition risk level classification rules are established according to the three-level warning thresholds, including:

[0337] Step S2231, for different key operating condition parameters, corresponding operating condition parameter fuzzy linguistic variables, operating condition parameter fuzzy subsets and operating condition parameter domains are set; the operating condition parameter fuzzy linguistic variables include pressure fuzzy linguistic variables, temperature fuzzy linguistic variables and vibration fuzzy linguistic variables; the operating condition parameter fuzzy subsets include pressure fuzzy subsets, temperature fuzzy subsets and vibration fuzzy subsets; the operating condition parameter domains include temperature domains, pressure domains and vibration domains;

[0338] Step S2232, according to the operating condition parameter fuzzy linguistic variables, operating condition parameter fuzzy subsets and operating condition parameter domains corresponding to each key operating condition parameter, the operating condition parameter fuzzy subset membership function of the corresponding key operating condition parameter is set; the operating condition parameter fuzzy subset membership function includes a pressure fuzzy subset membership function, a temperature fuzzy subset membership function and a vibration fuzzy subset membership function;

[0339] Step S2233, constructing a fuzzy rule base based on the fuzzy subset membership function of the operating condition parameters.

[0340] In the geometric parameter optimization module, the construction of the wear evolution prediction model includes:

[0341] Step S3110, constructing a training sample set for a wear evolution prediction model;

[0342] Step S3120, taking the real-time operating condition risk level as the attribute label of the training sample and integrating it into the training sample set;

[0343] Step S3130, select the long short-term memory network LSTM model, and use the training sample set incorporating attribute labels to train multiple wear evolution prediction sub-models according to different real-time operating condition risk levels.

[0344] In the working condition risk division module, the key geometric parameters of the optimized sealing structure are obtained to obtain the optimal geometric parameter combination of the slurry valve sealing structure, including:

[0345] According to the real-time wear morphology time series data set and the constructed wear evolution prediction model, the predicted wear morphology data is obtained; the predicted wear morphology data is integrated with the digital twin model of the slurry valve sealing structure, and the digital twin model of the slurry valve sealing structure is dynamically updated to obtain an updated digital twin model;

[0346] Based on the updated digital twin model, the key geometric parameters of the sealing structure are optimized to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

[0347] The key geometric parameters of the sealing structure are optimized to obtain the optimal geometric parameter combination of the slurry valve sealing structure, including:

[0348] Step S3310, setting key geometric parameters of the sealing structure of the slurry valve as optimized design variables, and establishing an optimized design model for the sealing structure;

[0349] Step S3320, setting the mechanical performance index of the sealing structure as the optimization objective function, and setting the constraint conditions according to the design requirements to improve the sealing structure optimization design model;

[0350] Step S3330, solving the sealing structure optimization design model to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

[0351] The method and apparatus of the present application may be implemented in many ways. For example, the method and apparatus of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above, unless otherwise specifically stated.

[0352] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0353] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A design method for a large-diameter slurry valve sealing structure suitable for high-wear working conditions, characterized in that: The method comprises: Obtain key geometric parameters of the slurry valve sealing structure, and construct a three-dimensional topological base model of the slurry valve sealing structure based on the key geometric parameters; construct a mechanical performance simulation model of the slurry valve sealing structure based on the three-dimensional topological base model; fuse and associate the three-dimensional topological base model of the slurry valve sealing structure with the mechanical performance simulation model to construct a digital twin model of the slurry valve sealing structure; Obtain a time series of multi-scale wear morphology images of the sealing structure of the slurry valve, and obtain a wear morphology time series data set based on the time series of multi-scale wear morphology images; obtain key operating parameters of the slurry valve, set three-level warning thresholds for the key operating parameters, and establish operating risk level classification rules based on the three-level warning thresholds; collect real-time operating parameters of the sealing structure of the slurry valve, and obtain the real-time operating risk level based on the operating risk level classification rules; Based on the wear morphology time series data set and the real-time operating risk level, a wear evolution prediction model is constructed. Combined with the digital twin model of the slurry valve sealing structure, the key geometric parameters of the sealing structure are optimized to obtain the optimal geometric parameter combination of the slurry valve sealing structure. The construction of the wear evolution prediction model comprises: Obtain a historical wear morphology time series dataset from the wear morphology time series dataset; A time window is selected, and according to the selected time window, a continuous statistical feature parameter subsequence is extracted from the historical wear morphology time series data set as an input feature vector; The statistical characteristic parameters of the next moment corresponding to the input feature vector are used as the output target value to construct input-output pairs to form training samples; Through the sliding time window, the entire historical wear morphology time series data set is traversed to generate a complete training sample set; the real-time working condition risk level is used as the attribute label of the training sample and integrated into the training sample set; A long short-term memory network (LSTM) model was selected, and multiple wear evolution prediction sub-models were trained using a training sample set incorporating attribute labels according to different real-time operating condition risk levels.

2. The method for designing a sealing structure of a large-diameter slurry valve suitable for high-wear conditions according to claim 1, characterized in that: The digital twin model of the slurry valve sealing structure is constructed as follows: The finite element method is used to solve the mechanical performance simulation model of the slurry valve sealing structure to obtain the stress and strain distribution results of the slurry valve sealing structure under the current working conditions; The stress-strain distribution results obtained by finite element solution and calculation are mapped to the three-dimensional topological base model of the slurry valve sealing structure to obtain a digital twin model of the slurry valve sealing structure.

3. The method for designing a sealing structure of a large-diameter slurry valve suitable for high-wear conditions according to claim 1, characterized in that: The multi-scale wear morphology image time series includes a three-dimensional morphology image time series and a nanoscale morphology image time series; The method of obtaining a time series of multi-scale wear morphology images of a slurry valve sealing structure comprises: Select the critical area of ​​the slurry valve sealing surface; In-situ characterization tests were carried out on key areas of the slurry valve sealing surface at different time points to obtain a time series of multi-scale wear morphology images; The step of obtaining a wear morphology time series data set according to the multi-scale wear morphology image time series includes: Calculate the surface roughness parameters of the three-dimensional topography image time series to obtain a three-dimensional topography feature parameter sequence; Perform difference calculation on the time series of nanoscale topography images to obtain a series of geometric profile changes; The three-dimensional morphological feature parameter sequence and the geometric profile variation sequence are fused to obtain the wear morphology time series data set.

4. The method for designing a sealing structure of a large-diameter slurry valve suitable for high-wear conditions according to claim 1, characterized in that: The three-level warning threshold includes a pressure three-level warning threshold, a temperature three-level warning threshold and a vibration three-level warning threshold; The three-level warning thresholds for key operating parameters are set as follows: Obtain the mechanical performance parameters and the first thermal safety margin parameters of the slurry valve sealing material, and determine the three-level pressure warning threshold according to the mechanical performance parameters and the first thermal safety margin parameters; the three-level pressure warning threshold includes the pressure warning threshold and the pressure danger threshold; the mechanical performance parameters include the yield strength σ s , stress fluctuation amplitude of the sealing structure under historical load , Critical failure pressure of sealing structure and the maximum pressure under historical loads The first thermal safety margin parameter includes the maximum temperature that the sealing structure can withstand and critical failure temperature of sealing materials ; Obtain the second thermal safety margin parameter of the sealing material, and determine the temperature three-level warning threshold according to the first thermal safety margin parameter and the second thermal safety margin parameter; the temperature three-level warning threshold includes the temperature warning threshold and the temperature danger threshold; the second thermal safety margin parameter includes the safe working temperature T0 of the sealing material, the allowable amplitude ΔT of the sealing material temperature fluctuation w and the safety margin temperature difference ΔT of the sealing material d ; The first vibration data of the slurry valve sealing structure is obtained, and a three-level vibration warning threshold is determined according to the first vibration data; the three-level vibration warning threshold includes a vibration warning threshold and a vibration danger threshold; the first vibration data includes an average vibration level V of the slurry valve sealing structure under rated working conditions. a and critical resonance vibration velocity V cr .

5. The method for designing a sealing structure of a large-diameter slurry valve suitable for high-wear conditions according to claim 4, characterized in that: Determining the third-level pressure warning threshold according to the mechanical performance parameter and the first thermal safety margin parameter includes: Get the number of cycles of the sealing structure under historical loads And the fatigue life reference value of sealing materials ,according to , , and σ s , calculate the pressure warning threshold coefficient C w ; According to the yield strength σ s and pressure warning threshold coefficient C w , get the pressure warning threshold P w ; according to , , and , calculate the pressure danger threshold coefficient C d ; According to the yield strength σ s and pressure warning threshold coefficient C d , get the pressure warning threshold P d .

6. The method for designing a sealing structure of a large-diameter slurry valve suitable for high-wear conditions according to claim 4, characterized in that: Determining the temperature level 3 warning threshold according to the first thermal safety margin parameter and the second thermal safety margin parameter includes: According to the safe working temperature T0 and the allowable amplitude of temperature fluctuation ΔT s , calculate the temperature warning threshold T w ; According to the safety margin temperature difference ΔT d and critical failure temperature , calculate the temperature danger threshold T d .

7. The method for designing a sealing structure of a large-diameter slurry valve suitable for high-wear conditions according to claim 4, characterized in that: Determining the vibration level 3 warning threshold according to the first vibration data includes: Set the warning margin coefficient R w and the risk margin factor R d , according to the warning margin coefficient R w and the average vibration level V under rated conditions a , calculate the vibration warning threshold V w ; According to the risk margin factor R d and critical resonance vibration velocity V cr , calculate the vibration danger threshold V d .

8. The method for designing a sealing structure of a large-diameter slurry valve suitable for high-wear conditions according to claim 1, characterized in that: The establishment of the operating condition risk level classification rules according to the three-level warning thresholds includes: According to different key working condition parameters, corresponding working condition parameter fuzzy linguistic variables, working condition parameter fuzzy subsets and working condition parameter domains are set; According to the working condition parameter fuzzy linguistic variables, working condition parameter fuzzy subsets and working condition parameter domains corresponding to each key working condition parameter, the working condition parameter fuzzy subset membership function of the corresponding key working condition parameter is set; Define the operating condition risk fuzzy linguistic variables {normal, abnormal, dangerous}, the operating condition risk fuzzy subset {normal level, abnormal level, dangerous level} and the operating condition risk domain [0,1]; Based on the fuzzy subset membership function of the working condition parameters, the working condition risk value is calculated; the membership function of the corresponding working condition risk fuzzy subset is defined in the working condition risk domain; According to the operating condition risk value and the membership function of the operating condition risk fuzzy subset, the operating condition risk level is obtained; According to the IF-THEN format, fuzzy rules are established and a fuzzy rule base is generated; each fuzzy rule contains an antecedent part and a consequent part, and the format of the fuzzy rule is: IF antecedent part THEN consequent part, and the operating condition risk level is written into the consequent part.

9. A device for designing a sealing structure of a large-diameter slurry valve suitable for high-wear working conditions, which is used to implement a method for designing a sealing structure of a large-diameter slurry valve suitable for high-wear working conditions as described in any one of claims 1 to 8, characterized in that: The device comprises: Digital twin model construction module: used to obtain the key geometric parameters of the slurry valve sealing structure, and construct a three-dimensional topological base model of the slurry valve sealing structure based on the key geometric parameters; construct a mechanical performance simulation model of the slurry valve sealing structure based on the three-dimensional topological base model; fuse and associate the three-dimensional topological base model of the slurry valve sealing structure with the mechanical performance simulation model to construct a digital twin model of the slurry valve sealing structure; Multi-scale data acquisition module: used to obtain the multi-scale wear morphology image time series of the slurry valve sealing structure, and obtain the wear morphology time series data set according to the multi-scale wear morphology image time series; Working condition risk classification module: used to obtain key working condition parameters of the slurry valve, set three-level warning thresholds for the key working condition parameters, and establish working condition risk level classification rules based on the three-level warning thresholds; collect the real-time working condition parameters of the slurry valve sealing structure, and obtain the real-time working condition risk level based on the working condition risk level classification rules; Geometric parameter optimization module: Based on the wear morphology time series data set and the real-time operating condition risk level, a wear evolution prediction model is constructed. Combined with the digital twin model of the slurry valve sealing structure, the key geometric parameters of the sealing structure are optimized to obtain the optimal geometric parameter combination of the slurry valve sealing structure.

Images