A simulation method and system for dynamic contact characteristics of a valve sealing pair during an opening and closing process

Abstract

This invention relates to the field of numerical simulation technology and proposes a method and system for simulating the dynamic contact characteristics of valve sealing pairs during the opening and closing process. The method includes: coupling shape parameters of the sealing pairs in the valve to be simulated to obtain basic attribute data; extracting the contour data of the sealing surface and encoding the contour features to obtain a shape feature vector; performing shape-guided point placement on the contact interface of the sealing pairs, pairing and mapping contact nodes and constructing a node correlation matrix; obtaining the sequence of motion parameters of the valve during the opening and closing process, loading them onto the contact node pairs according to time steps to obtain a dynamic contact state sequence; performing time-series difference extrapolation on the state changes of adjacent time steps to obtain the contact transient changes; performing full-process cumulative reconstruction of the contact transient changes, and co-interpreting the cumulative contact field and the shape feature vector to obtain a dynamic contact characteristic distribution map. This invention can improve the simulation efficiency of the dynamic contact characteristics of valve sealing pairs during the opening and closing process.

Description

Technical Field

[0001] This invention relates to the field of numerical simulation technology, and in particular to a method and system for simulating the dynamic contact characteristics of valve sealing pairs during the opening and closing process. Background Technology

[0002] Current simulation analyses of valve sealing pair contact characteristics primarily focus on static sealing conditions, generally employing traditional finite element simulation frameworks for numerical calculations. This approach requires global mesh generation of the entire sealing pair structure. To meet the accuracy requirements of the contact interface, high-density mesh refinement is typically performed on the sealing surface region, directly leading to a sharp increase in the computational degrees of freedom and solution scale of the simulation model, significantly increasing computational resource consumption. In dynamic contact analysis scenarios involving valve opening and closing processes, this method requires repeated nonlinear contact iterations within discrete time steps. The increased mesh size further complicates convergence, prolonging the computation time per time step and ultimately resulting in a substantial increase in the simulation calculation cycle for the entire opening and closing process. This makes it impossible to support rapid optimization design and performance verification of valve sealing structures. In traditional simulation modes, geometric and material parameters are simply superimposed through material assignment and boundary condition settings, failing to form a deep coupling mechanism between shape and properties. This makes it difficult to accurately reflect the true contact response under the combined effects of material properties and geometric morphology, further reducing the engineering reference value of the simulation results.

[0003] Existing valve sealing pair contact simulation technologies have significant limitations in dynamic characteristic analysis. Most schemes can only obtain static contact results at a few discrete time points, lacking a continuous extrapolation mechanism for the evolution of the contact state throughout the entire opening and closing process. They cannot accurately capture the transient changes in the contact state between adjacent motion stages, and cannot fully reproduce the dynamic evolution process of contact stress migration and contact area changes during the opening and closing of the sealing pair. Furthermore, existing technologies lack a correlation mechanism between the sealing surface morphology features and the layout of contact simulation nodes, failing to adaptively allocate computational nodes based on the differentiated distribution of morphological features. This results in insufficient accuracy in analyzing contact characteristics in areas of abrupt morphological changes, while wasting computational resources in areas with gentle morphological changes. In addition, existing simulation schemes do not achieve collaborative interpretation of the contact evolution process and morphological features, failing to explain the distribution and evolution of the contact field from the root cause of morphology. This makes it difficult to support targeted optimization design of the sealing pair structure, ultimately leading to simulation results that fail to meet the expected standards for engineering applications in terms of both reliability and overall computational efficiency. Therefore, improving the simulation efficiency and computational accuracy of the dynamic contact characteristics during the opening and closing process of valve sealing pairs has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for simulating the dynamic contact characteristics of valve sealing pairs during the opening and closing process, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for simulating the dynamic contact characteristics of a valve sealing pair during the opening and closing process, comprising: P1. Couple the geometric parameters and material parameters of the sealing pair in the valve to be simulated with shape parameters to obtain the basic property data of the sealing pair; P2. Extract the sealing surface contour data from the basic attribute data, and encode the contour features of the sealing surface contour data to obtain the shape feature vector of the sealing pair. P3. Based on the morphological feature vector, morphological guidance points are placed on the contact interface of the sealing pair, and the contact nodes after placement are paired and mapped to construct the node association matrix of the sealing pair. P4. Obtain the motion parameter sequence of the valve to be simulated during the opening and closing process, and load the motion parameter sequence onto the contact node pairs in the node association matrix according to the time step to obtain the dynamic contact state sequence of the contact node pairs. P5. Perform time-series difference deduction on the state changes between adjacent time steps in the dynamic contact state sequence to obtain the contact transient changes of the sealing pair; P6. Perform full-process cumulative reconstruction of the contact transient change, and co-interpret the reconstructed cumulative contact field with the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair.

[0006] In a preferred embodiment, the step of coupling the geometric and material parameters of the sealing pair in the valve to be simulated with morphological parameters to obtain the basic property data of the sealing pair includes: Extract the geometric and material parameters of the sealing pair in the valve to be simulated; The geometric parameters and the material parameters are decoupled by intrinsic features to obtain the morphology-dominant component of the geometric parameters and the property-dominant component of the material parameters. Heterogeneous feature matching is performed on the morphology dominant component and the attribute dominant component to obtain the morphological correspondence between the morphology dominant component and the attribute dominant component; Based on the shape correspondence, the shape dominant component and the attribute dominant component are reconstructed by basis weighting to obtain the basis coupling parameters of the shape dominant component and the attribute dominant component. The base coupling parameters are uniformly normalized to obtain the basic attribute data of the sealing pair.

[0007] In a preferred embodiment, the step of extracting sealing surface contour data from the basic attribute data and performing contour feature encoding on the sealing surface contour data to obtain the morphological feature vector of the sealing pair includes: The basic attribute data is subjected to domain feature analysis to obtain the geometric structure features and material property features of the basic attribute data; Based on the material properties, the geometric features are guided to undergo contour calibration to obtain the sealing surface contour data of the sealing pair. The contour direction of the sealing surface contour data is encoded to obtain the contour direction code sequence of the sealing surface contour data; The contour direction code sequence is subjected to principal component feature aggregation to obtain the shape feature vector of the sealing pair.

[0008] In a preferred embodiment, the step of performing shape-guided point placement on the contact interface of the sealing pair based on the shape feature vector, and pairing and mapping the contact nodes after point placement to construct the node association matrix of the sealing pair includes: A difference analysis is performed on the morphological feature vectors to obtain the morphological difference distribution of the morphological feature vectors; Based on the morphological difference distribution, the dimensional components of the morphological feature vector are mapped onto the contact interface of the sealing pair to obtain the encrypted contact nodes and sparse contact nodes of the contact interface. By projecting the encrypted contact node and the sparse contact node onto both sides of the contact surface, the active contact surface node and the passive contact surface node of the sealing pair are obtained. The active contact surface node and the passive contact surface node are matched for nearest neighbor, and the node association matrix of the sealing pair is obtained by constructing an association topology from the matched node pairing sequence.

[0009] In a preferred embodiment, the step of mapping the dimensional components of the morphological feature vector to the contact interface of the sealing pair according to the morphological difference distribution to obtain the encrypted contact nodes and sparse contact nodes of the contact interface includes: The morphological difference distribution is hierarchically quantified to obtain a difference grading map of the contact interface; Based on the difference gradation map, the contact interface is segmented by density clustering to obtain the high-density mapping region and the low-density mapping region of the contact interface. Based on the high-density mapping region and the low-density mapping region, the dimensional components of the shape feature vector are sorted into dense mapping components and sparse mapping components. Non-uniform mapping is performed on the dense mapping component and the sparse mapping component to obtain the encrypted contact node and sparse contact node of the contact interface.

[0010] In a preferred embodiment, obtaining the motion parameter sequence of the valve to be simulated during the opening and closing process, and loading the motion parameter sequence onto the contact node pairs in the node association matrix according to time steps to obtain the dynamic contact state sequence of the contact node pairs, includes: The driving timing parameters and displacement response parameters of the valve to be simulated during the opening and closing process are obtained, and the driving timing parameters and displacement response parameters are time-synchronized and merged to obtain the motion parameter sequence of the valve to be simulated. The motion parameter sequence is discretized by step size to obtain the time step sequence of the motion parameter sequence; Using the time step sequence as the advancement step size, the instantaneous motion parameter values ​​of the motion parameter sequence are extracted step by step, and the instantaneous motion parameter values ​​are coupled with the contact node pairs in the node association matrix to obtain the instantaneous contact state parameters of the contact node pairs. The instantaneous contact state parameters are systematically aggregated in a step-by-step manner to obtain the dynamic contact state sequence of the contact node pairs.

[0011] In a preferred embodiment, the step of performing time-series differential deduction on the state changes between adjacent time steps in the dynamic contact state sequence to obtain the contact transient changes of the sealing pair includes: Along the time step direction, the dynamic contact state sequence is stripped by step differential analysis to obtain the state change of the contact node pair between adjacent time steps; By performing hierarchical trend fitting on the state change, the evolution trend of the change in the contact node pair is obtained; The transient characteristics of the evolution trend of the change are extracted to obtain the contact transient change of the sealing pair.

[0012] In a preferred embodiment, the step of performing hierarchical trend fitting on the state change to obtain the evolution trend of the change in the contact node pair includes: The state change quantity is divided into windows to obtain a time window sequence of the state change quantity; Within each time window of the time window sequence, the state change is fitted using a window rate to obtain a window trend value for the state change. The formula for calculating the window trend value is as follows: ; in, Indicates the first The contact node pair in the th ... Window trend values ​​within a time window, Indicates the first The contact node pair in the th ... Within the first time window The state change at each time step Indicates the first The contact node pair in the th ... The state change at the first time step within a time window Indicates the first The time interval between the first time step and the second time step. Indicates the first The number of time steps contained within a time window; The window trend value is recursively extrapolated window by window to obtain the evolution trend of the change in the contact node pair.

[0013] In a preferred embodiment, the step of performing full-process cumulative reconstruction of the contact transient change and co-interpreting the reconstructed cumulative contact field with the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair includes: Using the time step sequence as the entire time series axis, the contact transient changes are sequentially accumulated and merged, and the merged results are aggregated and reconstructed to obtain the cumulative contact field of the sealing pair. The field topology of the cumulative contact field is deconstructed to obtain the field structure characteristics of the cumulative contact field; Cross-domain feature coupling is performed between the field structure features and the morphological feature vector to obtain the overlapping and related features between the field structure features and the morphological feature vector; A global map of the overlapping and correlated features is plotted to obtain a dynamic contact characteristic distribution map of the sealing pair.

[0014] To address the aforementioned problems, the present invention also provides a simulation system for the dynamic contact characteristics of valve sealing pairs during the opening and closing process, the system comprising: The shape coupling module is used to couple the geometric parameters and material parameters of the sealing pair in the valve to be simulated with shape parameters to obtain the basic property data of the sealing pair; The contour encoding module, connected to the shape coupling module, is used to extract the sealing surface contour data from the basic attribute data and perform contour feature encoding on the sealing surface contour data to obtain the shape feature vector of the sealing pair. The point matching module, connected to the contour encoding module, is used to perform shape-guided point placement on the contact interface of the sealing pair based on the shape feature vector, and to perform pairing mapping on the contact nodes after point placement to construct the node association matrix of the sealing pair. The timing loading module, connected to the point pairing module, is used to obtain the motion parameter sequence of the valve to be simulated during the opening and closing process, and load the motion parameter sequence onto the contact node pairs in the node association matrix according to the time step to obtain the dynamic contact state sequence of the contact node pairs. The differential deduction module, connected to the timing loading module, is used to perform timing differential deduction on the state changes between adjacent time steps in the dynamic contact state sequence to obtain the contact transient changes of the sealing pair. The collaborative interpretation module, connected to the differential inference module, is used to perform full-process cumulative reconstruction of the contact transient change, and to perform collaborative interpretation of the reconstructed cumulative contact field with the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention integrates the geometric and material parameters of the sealing pair through a shape-parameter coupling mechanism, constructing unified and standardized basic attribute data. This achieves deep matching between geometric morphology features and material property features, accurately characterizing the inherent properties of the sealing pair and providing a reliable data foundation for subsequent contact characteristic simulation. Based on the morphology feature vector generated by contour feature encoding, morphology-guided point placement at the contact interface is achieved. The node distribution density can be adaptively adjusted according to the morphology differences, ensuring the solution accuracy in complex morphological areas while simplifying the overall computational node scale, effectively reducing the computational load and improving the computational efficiency and resource utilization of contact simulation. By constructing a node association matrix through the pairing mapping of contact nodes, a stable bilateral node correspondence can be established, providing a clear computational carrier for the temporal loading of dynamic contact states and ensuring the orderly and stable progress of subsequent dynamic deduction processes.

[0016] 2. This invention generates a continuous and complete dynamic contact state sequence by loading the motion parameters of the valve opening and closing process into the contact node pairs in the node correlation matrix step by step, thus fully reconstructing the contact state evolution process of the entire opening and closing process of the sealing pair. By employing a time-series differential deduction method to extract transient changes in contact, it can accurately capture the contact state change patterns between adjacent time steps and clearly present the transient evolution trend of contact characteristics. Through full-process cumulative reconstruction to generate a cumulative contact field, and combined with morphological feature vectors for collaborative interpretation, a dynamic contact characteristic distribution map is finally output. This map intuitively presents the spatial distribution and temporal evolution patterns of the sealing pair's contact characteristics, providing comprehensive and accurate data support and reference for the structural design, performance optimization, and failure analysis of valve sealing pairs. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for simulating the dynamic contact characteristics of a valve sealing pair during the opening and closing process, according to an embodiment of the present invention. Figure 2 This is a functional block diagram of a simulation system for the dynamic contact characteristics of a valve sealing pair during the opening and closing process, provided in an embodiment of the present invention.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] This application provides a method for simulating the dynamic contact characteristics of a valve sealing pair during the opening and closing process. The execution subject of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for simulating the dynamic contact characteristics of a valve sealing pair during the opening and closing process can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0021] Reference Figure 1 The diagram shown is a flowchart illustrating a method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing, according to an embodiment of the present invention. In this embodiment, the method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing includes: P1. Couple the geometric parameters and material parameters of the sealing pair in the valve to be simulated with shape parameters to obtain the basic property data of the sealing pair; In this embodiment of the invention, the step of coupling the geometric parameters and material parameters of the sealing pair in the valve to be simulated with morphological parameters to obtain the basic property data of the sealing pair includes: Extract the geometric and material parameters of the sealing pair in the valve to be simulated; The geometric parameters and the material parameters are decoupled by intrinsic features to obtain the morphology-dominant component of the geometric parameters and the property-dominant component of the material parameters. Heterogeneous feature matching is performed on the morphology dominant component and the attribute dominant component to obtain the morphological correspondence between the morphology dominant component and the attribute dominant component; Based on the shape correspondence, the shape dominant component and the attribute dominant component are reconstructed by basis weighting to obtain the basis coupling parameters of the shape dominant component and the attribute dominant component. The base coupling parameters are uniformly normalized to obtain the basic attribute data of the sealing pair.

[0022] First, locate the active and driven sealing components of the valve sealing pair to be simulated. Within the complete structural form of the sealing pair, precisely delineate the effective sealing area where the two components are in contact. Eliminate non-sealing functional parts of the sealing pair, such as support structures, connection structures, and transmission structures, that do not participate in the contact action. Perform parameter acquisition operations separately for the active and driven sealing components. Along the three dimensions of the sealing surface—circumferential extension path, radial distribution range, and axial undulation shape—collect all structural morphological information directly related to the contact behavior. This includes the overall radial dimension of the sealing contact surface, the axial contact width, the slope angle of the sealing cone surface, the micro-undulation shape of the sealing surface surface, the texture direction and distribution density formed by surface processing, the transition fillet shape of the inner and outer edges of the sealing surface, the flatness deviation of the sealing surface, and the coaxiality deviation of the sealing surface, among other geometric features. The materials used in the two contact components of the sealing pair are each analyzed, and all performance information of the corresponding materials under contact stress scenarios is retrieved. This includes the elastic deformation response characteristics of the material after being subjected to external force, the critical threshold of plastic deformation, surface contact friction characteristics, hardness index against surface crushing, performance decay law under cyclic stress, performance change trend under temperature fluctuation, and performance evolution law during contact wear. All collected geometric and material information is matched one-to-one with the effective contact area of ​​the sealing surface to ensure that each piece of geometric information corresponds to a specific contact position and each piece of material information is accurately matched to the corresponding contact component. This fully covers all structural and material factors that affect the distribution and change of the contact state during the opening and closing contact of the sealing pair, thus completing the extraction of the geometric and material parameters of the sealing pair in the valve to be simulated.

[0023] The extracted geometric and material parameters were decomposed and screened at the intrinsic feature level. The core criterion for intrinsic features was the direct contribution of the parameters to the contact behavior of the sealing interface. Only parameters that directly changed the contact stress, contact area, interface deformation, and friction state of the sealing surface were classified as core intrinsic features. Parameters that only affected the overall structural strength, external dimensions, and stiffness of the non-contact area of ​​the valve body were uniformly classified as irrelevant secondary features. A hierarchical screening and decoupling process was adopted. The first step was to analyze the action transmission path of each single type of parameter individually. The second step was to distinguish whether the parameter's action object directly affected the sealing contact interface. The third step was to complete the primary and secondary feature segmentation according to the degree of contribution to the contact response. The decoupling operations of the geometric and material parameter data were completed independently throughout the process without interference. Feature levels were divided according to the weight of each feature's influence on the contact stress distribution, contact area change, contact friction response, and contact wear rate of the sealing pair. First, all the parameters were analyzed one by one. The force transmission path and range corresponding to each parameter are analyzed. The morphological features that directly affect the contact interface are distinguished from the structural features that only affect the overall structural bearing capacity. Secondary features in the geometric parameters that only determine the overall dimensions of the sealing pair and do not directly participate in the stress transmission at the contact interface are removed. All core features that can directly change the contact morphology of the sealing surface and directly affect the uniformity of stress distribution at the contact interface are retained, thus obtaining the morphological dominant components of the geometric parameters. Then, the response paths and ranges corresponding to all material parameters are analyzed one by one. The attribute features that directly determine the mechanical response of the contact interface are distinguished from the ontological features that only affect the overall structural strength of the component. Secondary features in the material parameters that only affect the overall resistance to damage of the component and do not directly determine the deformation and friction performance of the contact interface are removed. All core attribute features that can directly determine the deformation response amplitude of the contact interface and directly affect the magnitude of the contact friction resistance are retained, thus obtaining the attribute dominant components of the material parameters.

[0024] A feature correlation analysis framework for morphology-dominant and property-dominant components is constructed. Using the complete contact process during the opening and closing of the sealing pair as the analysis vehicle, and force transmission and deformation response at the contact interface as the correlation link, each morphological feature in the morphology-dominant component is matched one by one. The logic and transmission path of the morphological feature inducing material deformation, changing friction state, and adjusting stress distribution during contact stress are analyzed. The correspondence between the magnitude of morphological feature change and the degree of material property response is clarified. The correlation matching of all morphological features with their corresponding property features is completed sequentially. The matching process follows a unified and standardized coupling matching logic, maintaining a one-to-one binding mapping relationship between morphological components and material property components, avoiding one-to-many or many-to-one cross-matching throughout the process. The matching effect is comprehensively judged based on three indicators: contact stress, effective contact area interface, and friction deformation. After matching, a structured correlation data table in the form of key-value pairs is generated as intermediate output data of the process. Clear and stable correspondence rules between the two types of heterogeneous features are established, fully covering the interaction between all core morphological features and their corresponding material properties, with no missing core feature pairs, thus obtaining the morphological correspondence between the morphology-dominant and property-dominant components.

[0025] A unified feature representation substrate is constructed as the fusion carrier for the two dominant components. The substrate completes the dimensional construction based on the three-dimensional spatial scale of the sealing pair contact interface. The number of substrate dimensions corresponds one-to-one with the three extension directions of the sealing surface: circumferential, radial, and axial. The division density of each dimension matches the sampling accuracy of the previously extracted sealing surface contour. Each spatial unit inside the substrate reserves an independent storage channel to simultaneously store the two types of feature information: morphology and material. After the overall construction of the substrate is completed, the coordinate scale of each dimension is uniformly calibrated to eliminate the difference in dimensions. This feature substrate sets a unified feature representation dimension and data expression rule, and adapts to the fusion requirements of two types of heterogeneous features: geometric morphology features and material property features. It can fully carry the core information of the two types of features without information loss. Based on the degree of correlation and contact action weight of each feature pair in the shape-property correspondence, corresponding fusion weights are assigned to each feature item of the morphology dominant component and the property dominant component. All weight values ​​are obtained through multiple sets of sealing pair benchmark simulation experiments. The experiments cover sealing pair samples with different contour morphologies and different material combinations. The contribution ratio of each feature to the interface contact stress and wear amount is statistically used as the benchmark for weight values. The closer the correlation between a feature and contact behavior, the higher its contribution and the greater its weight. The total weight of the two types of features maintains a fixed baseline ratio. The entire weighted allocation process is uniformly applied to all feature items without any differentiated judgment criteria. The weight parameters are all derived from batch simulation statistical results. The weight allocation process strictly follows the role logic in the form-property correspondence. Features that are closely related and have a direct impact on the contact state are assigned higher fusion weights. The complete reconstruction process is divided into three steps executed sequentially. First, based on the form-property correspondence, the morphological dominant component and the property dominant component are matched to the same spatial unit of the substrate. Then, the fusion processing of the two types of feature components is completed using the calibrated weight values. Finally, the comprehensive parameters after fusion of each spatial unit are output. Once all units have been calculated, a complete and continuous set of substrate parameters is formed. According to the determined weight ratio, all features of the two types of dominant components are mapped item by item to a unified feature substrate for superposition and fusion. The fusion process completely preserves the core role information of the two types of features, while eliminating the differences in the representation system and expression scale between heterogeneous features, forming a complete fused feature set that simultaneously carries geometric morphological information and material property information, and obtaining the substrate coupling parameters of the morphological dominant component and the property dominant component.

[0026] The representation format, numerical scale, and data benchmark of the base coupling parameters are standardized and adjusted uniformly. The numerical dimensions, calculation precision, and file storage format of all coupling parameters are simultaneously unified. Global normalization scaling is used to eliminate numerical magnitude differences between different feature dimensions. Logical consistency checks are performed on parameters corresponding to all spatial points within the three-dimensional array. Abnormal parameters exceeding the geometric boundaries of the sealing pair or the material mechanical property thresholds are automatically identified and removed. This eliminates differences in representation scale and data benchmark deviations between different feature items, ensuring all coupling feature items are under a unified expression system. This guarantees comparability and calculability between different feature items. The logical self-consistency of each feature item within the coupling parameters is then checked one by one. Potential information deviations and logical conflicts during feature fusion are investigated. For feature items with deviations, adjustments are made based on the action logic in the form-property correspondence, ensuring that the association logic of all coupling feature items is completely consistent with the actual contact interaction law. Finally, a complete attribute dataset with a unified format, logical self-consistency, complete dimensions, and direct usability for subsequent simulation calculations is formed, yielding the basic attribute data of the sealing pair.

[0027] The beneficial effects are that the deep integration of the geometric parameters and material parameters of the sealing pair is achieved through the whole-process processing of shape parameter coupling, breaking the traditional processing mode of setting and loading the two types of parameters independently. This solution is equipped with a complete and unified coupling matching logic, hierarchical weight allocation rules, and standardized three-dimensional array data storage format. The weight values ​​are obtained based on the simulation calibration of multiple sealing pair benchmarks, and the actual contribution ratio of the feature to the contact behavior is used as the allocation basis. The mapping constraints, data format, and standardized processing flow within the entire coupling process are clearly defined. This can make up for the defects of existing technologies such as vague coupling rules, lack of unified processing standards, messy data structure, and difficulty in reproducing simulation results. It enables the generated basic attribute data to accurately reflect the true inherent characteristics under the synergistic effect of the sealing pair's geometric shape and material properties, and avoids the contact response characterization deviation caused by parameter separation processing. The intrinsic feature decoupling link sets clear feature judgment criteria and hierarchical screening decoupling process. It distinguishes the intrinsic core features from secondary irrelevant features based on the degree of direct effect of the parameters on the sealing contact behavior, forming a standardized parameter splitting logic, avoiding the problem of unstable simulation results caused by arbitrary selection of features. The core dominant components identified through intrinsic feature decoupling and screening eliminate interference from irrelevant features, focusing on the core factors at the contact interface. This improves the relevance and accuracy of the coupled data, reduces unnecessary computations in subsequent simulations, and clarifies the transmission paths of the two types of parameters through the shape-property correspondence established by heterogeneous feature matching. The fixed and repeatable one-to-one mapping coupling logic ensures the logical rationality and fusion accuracy of the base weighted reconstruction stage. Furthermore, the simulation-calibrated weighting method provides objective data support for weight allocation, effectively avoiding information misalignment and data distortion issues that easily occur during heterogeneous feature fusion. The resulting consistent and regularized data possesses a unified expression standard, providing a precise and unified data input foundation for subsequent sealing surface contour feature encoding, contact interface node layout, and dynamic contact state timing deduction. This effectively ensures the accuracy of subsequent contact characteristic simulation results. Simultaneously, the unified and standardized data format simplifies subsequent data retrieval and processing, reduces the adaptation costs for different types of parameters, improves the overall simulation efficiency, and supports the stable and efficient simulation of the dynamic contact characteristics of valve sealing pairs.

[0028] P2. Extract the sealing surface contour data from the basic attribute data, and encode the contour features of the sealing surface contour data to obtain the shape feature vector of the sealing pair. In this embodiment of the invention, the step of extracting sealing surface contour data from the basic attribute data and performing contour feature encoding on the sealing surface contour data to obtain the shape feature vector of the sealing pair includes: The basic attribute data is subjected to domain feature analysis to obtain the geometric structure features and material property features of the basic attribute data; Based on the material properties, the geometric features are guided to undergo contour calibration to obtain the sealing surface contour data of the sealing pair. The contour direction of the sealing surface contour data is encoded to obtain the contour direction code sequence of the sealing surface contour data; The contour direction code sequence is subjected to principal component feature aggregation to obtain the shape feature vector of the sealing pair.

[0029] Using the complete dataset of basic attribute data as the processing object, a domain-based judgment standard is pre-established based on the physical representation of the feature items. Features representing the spatial morphology, structural dimensions, surface morphology, and contour trajectory of the entity are divided into the same category, while features representing the mechanical response, contact performance, deformation law, and friction performance of the material are divided into another category. Then, each feature in the dataset is traversed and compared with the judgment standard one by one. All features describing the external dimensions, spatial morphology, surface undulation, contour extension path, and structural boundary position of the sealing pair entity are collected into the geometric structure feature set, and all features describing the deformation characteristics, contact response capability, friction performance, and surface performance parameters of the sealing pair material are collected into the material property feature set. During the collection process, each feature is only included in one feature set, and there is no cross-classification or omission. Finally, two clearly defined and independent feature sets are formed, resulting in the geometric structure features and material property features of the basic attribute data.

[0030] Effective boundary information of contact action is extracted from material property characteristics to clarify the surface depth range and radial action range of the sealing pair material during contact stress. This serves as the benchmark for contour calibration. Segment-by-segment calibration is carried out among all structural contours included in the geometric features. The spatial position of each contour segment is compared along the circumferential extension direction, radial distribution direction, and axial undulation direction of the sealing pair. Only contour segments falling within the effective contact action range of the material are retained, while supporting structure contours, non-sealing surface contours, and structural transition contours that exceed the effective contact range are eliminated. During the calibration process, the effective boundary of the contour is adjusted according to the influence law of material stress deformation to ensure that the calibrated contour completely matches the actual sealing contact action area. Subsequently, all retained contour segments are spliced ​​together in sequence according to spatial connection relationship to form a continuous and complete closed contour information set, thus obtaining the sealing surface contour data of the sealing pair.

[0031] Fixed feature points are selected as the encoding starting positions on the closed trajectory of the sealing surface contour data. A uniform sampling step size, adapted to the overall contour morphology refinement, is set along the uniform extension direction of the contour. The sampling step size is determined based on the actual size and morphology refinement requirements of the sealing surface. Using the maximum radial dimension of the sealing surface as a benchmark, the sampling step size value is calculated according to a preset ratio, with the unit consistent with the size unit of the sealing surface contour data. Simultaneously, the sampling step size must not exceed half the minimum size of the micro-morphological features of the sealing surface to ensure complete capture of all key morphological features such as micro-undulations and slope transitions of the sealing surface, avoiding loss of morphological information due to excessively large sampling step sizes. Alternatively, the sampling step size can be calculated using the pixel resolution of the sealing surface contour data to ensure that the distribution density of sampling points matches the accuracy of the contour data, ensuring complete sampling. The system captures all undulations and changes in orientation, selecting continuous sampling points on the contour trajectory according to the sampling step size. This ensures that all sampling points are evenly distributed along the contour and completely cover the entire contour trajectory. Subsequently, the spatial orientation of the contour line segments between every two adjacent sampling points is determined. A unified contour orientation classification standard is pre-established, using the sealed sub-center reference axis as a reference. Different orientation categories are divided based on the angle range formed by the contour line segments and the reference axis. Each orientation category corresponds to a unique fixed number as a direction code. The number is only used as a distinguishing identifier and does not represent a superior or inferior level. The direction code generation follows a unified execution rule, determining the orientation category of each line segment along a preset fixed detour order and simultaneously assigning a matching exclusive code. All line segment codes are arranged continuously according to the sampling order, with no skipped codes, disordered order, or unclassified line segments throughout the process. Each contour line segment is assigned a direction code corresponding to its direction. The direction code classification covers all possible direction intervals of the contour. Each contour line segment corresponds to a unique direction code. The direction codes of all contour line segments are arranged in the order of sampling to form a continuous, ordered, unbroken and non-repeating code set, thus obtaining the contour direction code sequence of the sealing surface contour data.

[0032] A comprehensive statistical analysis is conducted on all the encoded content contained in the contour direction code sequence. The overall frequency of occurrence of each type of code, the length of consecutive occurrences, the distribution position in different regions of the contour, and the combination patterns of different codes are counted one by one. Based on the contribution of features to the overall morphology of the sealing surface, the core feature components that can determine the overall morphology of the sealing surface are selected. Secondary feature components and random fluctuation features that appear randomly and have no significant effect on the overall morphology are removed. All the selected core feature components are arranged in a fixed dimensional order to form an ordered feature set that is dimensionally unified, structurally stable, and can completely reflect the core morphological characteristics of the sealing surface, thus obtaining the morphological feature vector of the sealing pair.

[0033] The beneficial effects are that domain-specific feature analysis can clearly clarify the boundaries and scope of action of the two types of features in the basic attribute data, establish a clear feature classification system, avoid analysis interference caused by the mixing of different attribute features, and provide clear data support for subsequent contour calibration work. The contour calibration based on material attribute features can overcome the limitations of pure geometric contour extraction, fully combine the actual effective range of material contact action to accurately locate the effective contact contour of the sealing surface, and ensure that the contour data completely corresponds to the actual sealing contact action area, effectively improving the pertinence and accuracy of subsequent morphology analysis and contact simulation. The contour direction coding is equipped with standardized direction division basis, fixed coding system and ordered generation rules, clearly define the direction judgment logic, the meaning of the direction code and the coding mapping rules, and transform continuous contour morphology information into a standardized ordered coding sequence, realize the structured and ordered expression of contour morphology information, eliminate the processing obstacles caused by non-standardized expression of contour morphology, and provide a standardized data carrier for subsequent feature extraction and operation processing. Principal feature aggregation can simplify feature dimensions while fully preserving core morphological information, eliminating the computational burden caused by redundant information and invalid fluctuations, and significantly improving the concentration and representation efficiency of feature information. The resulting morphological feature vector can centrally and accurately represent the core morphological characteristics of the sealing surface, providing accurate morphological basis for subsequent morphological guidance of contact interface placement, and effectively improving the rationality of subsequent node placement and the overall computational efficiency of contact simulation.

[0034] P3. Based on the morphological feature vector, morphological guidance points are placed on the contact interface of the sealing pair, and the contact nodes after placement are paired and mapped to construct the node association matrix of the sealing pair. In this embodiment of the invention, the step of guiding the placement of points on the contact interface of the sealing pair based on the topographic feature vector, and then pairing and mapping the contact nodes after placement to construct the node association matrix of the sealing pair includes: A difference analysis is performed on the morphological feature vectors to obtain the morphological difference distribution of the morphological feature vectors; Based on the morphological difference distribution, the dimensional components of the morphological feature vector are mapped onto the contact interface of the sealing pair to obtain the encrypted contact nodes and sparse contact nodes of the contact interface. By projecting the encrypted contact node and the sparse contact node onto both sides of the contact surface, the active contact surface node and the passive contact surface node of the sealing pair are obtained. The active contact surface node and the passive contact surface node are matched for nearest neighbor, and the node association matrix of the sealing pair is obtained by constructing an association topology from the matched node pairing sequence.

[0035] The step of mapping the dimensional components of the morphological feature vector to the contact interface of the sealing pair according to the morphological difference distribution to obtain the encrypted contact nodes and sparse contact nodes of the contact interface includes: The morphological difference distribution is hierarchically quantified to obtain a difference grading map of the contact interface; Based on the difference gradation map, the contact interface is segmented by density clustering to obtain the high-density mapping region and the low-density mapping region of the contact interface. Based on the high-density mapping region and the low-density mapping region, the dimensional components of the shape feature vector are sorted into dense mapping components and sparse mapping components. Non-uniform mapping is performed on the dense mapping component and the sparse mapping component to obtain the encrypted contact node and sparse contact node of the contact interface.

[0036] All dimensional components of the morphological feature vector are bound point-to-point to each spatial point on the sealing interface according to a preset spatial correspondence rule. The correspondence process is carried out row-by-row and column-by-column along the circumferential and radial distribution directions of the contact interface, ensuring that each dimensional component accurately matches a unique spatial point on the contact interface, and that each spatial point corresponds to a unique dimensional component, without misalignment, omission, or duplicate correspondence. After binding, the morphological feature vector components corresponding to each sampling point of the sealing surface contour are used as the object of difference calculation. A preset benchmark morphological feature vector is selected as the comparison reference, and normalized Euclidean distance is used as the quantitative evaluation index for difference measurement. During calculation, the morphological feature vector to be analyzed is first compared with the benchmark morphological feature vector. Normalization is used to eliminate dimensional differences between different components. Then, the morphological difference values ​​corresponding to each point are obtained by solving the normalized Euclidean distance. The required input data includes the morphological feature vector output by the aggregation of principal component features and the pre-set baseline morphological feature vector. Then, taking each spatial point on the contact interface as the object, the morphological feature differences of the corresponding dimension component of the point are compared with the corresponding dimension components of all surrounding adjacent points. The morphological feature change amplitude at the point is counted. After the morphological change amplitude of all points is counted, they are arranged in an orderly manner according to the spatial coordinate position of each point to form a spatial distribution set of morphological change amplitude that completely covers the entire contact interface of the sealing pair, and the morphological difference distribution of the morphological feature vector is obtained.

[0037] A comprehensive statistical analysis was conducted on the morphological variation range of all points included in the morphological variation distribution to clarify the overall value range of the morphological variation range and determine the upper and lower limits of the variation range. Subsequently, a fixed number of morphological variation level intervals were defined within the value range. The number of level divisions was uniformly set in conjunction with the simulation accuracy requirements of the sealing pair. Under normal working conditions, five levels of morphological variation intervals were uniformly divided. The level boundaries were determined based on the sealing surface machining accuracy standards and simulation accuracy thresholds. First, the industry-standard sealing surface roughness and allowable deviation of contour undulation were extracted as benchmark boundary values. Then, the upper and lower limits of the variation range obtained from the comprehensive statistical analysis were linearly and evenly adjusted to ensure that each level interval contained the required morphological variation range. The morphological undulations exhibit similar contact mechanical response patterns. The boundaries of each level interval are continuously connected and do not overlap, fully covering the entire range of variation amplitudes. Each level interval corresponds to a clear level of morphological severity. After the level intervals are defined, the morphological variation amplitude of each point on the contact interface is compared with each level interval one by one. A difference level label corresponding to the variation amplitude is matched for each point. All level labels are strictly arranged according to their spatial positions, forming a morphological difference grading map with clear level division labels that fully covers the entire contact interface, thus obtaining the difference grading map of the contact interface.

[0038] Using the grade markers in the difference grading map as the core basis for region segmentation, starting from any edge point of the contact interface as the traversal starting position, all points are traversed sequentially along the circumferential and radial directions of the contact interface. During the traversal, consecutive points that are spatially adjacent and have completely consistent grade markers are aggregated and merged to form multiple independent consecutive region blocks. After traversing all points, all consecutive region blocks corresponding to high difference grades are uniformly grouped and designated as high-density mapping regions, and all consecutive region blocks corresponding to low difference grades are uniformly grouped and designated as low-density mapping regions. The boundaries of the two types of regions are strictly defined by the line connecting the points where the grade markers change. There are no overlapping or intersecting areas, and the regions completely cover the entire contact interface. There are no undivided blank points, thus obtaining the high-density mapping region and low-density mapping region of the contact interface.

[0039] Based on the spatial distribution range of high-density and low-density mapping regions on the contact interface, the dimensional components in the morphological feature vector that match the spatial positions of the two types of regions are searched in reverse. The region category to which the spatial point corresponding to each dimensional component belongs is checked one by one. All dimensional components whose spatial points fall within the range of the high-density mapping region are grouped into the same type of component set, and all dimensional components whose spatial points fall within the range of the low-density mapping region are grouped into another type of component set. The grouping process strictly follows the correspondence of spatial positions. All dimensional components are classified, and there are no omissions or overlaps. Each dimensional component belongs to only one type of component set, resulting in dense mapping components and sparse mapping components.

[0040] For high-density mapping regions corresponding to dense mapping components, all dense mapping components are synchronously mapped to each contact node within the region. Each component is matched with a separate spatial node, and contact nodes are set with a smaller node spacing. The node placement process is carried out uniformly along the circumference and radial direction of the region. The entire placement work is completed using a partitioned adaptive mesh generation algorithm. The criteria for determining node density are entirely based on the morphological difference grading map obtained earlier. The higher the morphological difference level, the more severe the contour undulation changes within the region, and a smaller placement spacing is used to increase the number of nodes. The lower the morphological difference level, the gentler the surface trend of the region, and a larger placement spacing is used to reduce the number of nodes. This scheme divides morphological differences into five levels. The gradient, from level one to level five, progressively increases in the intensity of morphological undulation, accompanied by progressively decreasing reference point spacing and progressively tightening adjacent node distance constraints. A linear conversion is performed based on the total radial dimension of the sealing surface. Under standard operating conditions, when the radial dimension of the sealing surface is 10mm, the reference point spacing in the level one region is 0.5mm, with a minimum adjacent node spacing of 0.2mm and a maximum spacing of 0.8mm; the reference point spacing in the level two region is 0.4mm, with a minimum adjacent node spacing of 0.15mm and a maximum spacing of 0.7mm; the reference point spacing in the level three region is 0.3mm, with a minimum adjacent node spacing of 0.1mm and a maximum spacing of 0.6mm; and the reference point spacing in the level four region is 0.2mm, with a minimum adjacent node spacing of 0. 0.8mm, maximum spacing 0.5mm; the baseline spacing of the fifth-level area is 0.1mm, the minimum spacing between adjacent nodes is 0.05mm, and the maximum spacing is 0.4mm; if the radial dimension of the sealing surface deviates from the standard baseline value of 10mm, all spacing and distance thresholds are simultaneously enlarged or reduced proportionally according to the scaling ratio of the radial dimension. The algorithm will first read the difference level value corresponding to each area and then match the preset spacing parameters, synchronously constraining the minimum and maximum distance thresholds between adjacent nodes to avoid excessive clustering of local nodes or large-scale node loss, ensuring that the nodes can completely cover all morphological change positions in the area, accurately capturing the influence of subtle morphological undulations in the area on the contact state, and targeting the low-density mapping corresponding to sparse mapping components. In the radiation region, all sparse mapping components are uniformly mapped to the contact nodes in the low-density region. The components and nodes maintain spatial matching. A larger node spacing is used to set the contact nodes. The node deployment process is also carried out uniformly along the circumference and radial direction of the region to ensure that the nodes can cover the overall morphological features of the region. There is no need to add extra encryption for fine morphology. After all contact nodes are deployed, they are grouped according to the category of the region where the nodes are located. All nodes in the high-density mapping region are grouped into encrypted contact nodes, and all nodes in the low-density mapping region are grouped into sparse contact nodes. The two types of nodes together completely cover the entire contact interface, resulting in the encrypted contact nodes and sparse contact nodes of the contact interface.

[0041] Using the intermediate reference plane when the two contact surfaces of the sealing pair are fully in contact as the projection symmetry reference, the set of single-sided contact nodes, composed of dense contact nodes and sparse contact nodes, is symmetrically projected onto the other contact surface of the sealing pair along a direction perpendicular to the intermediate reference plane. During the projection process, the relative spatial position, distribution density, and arrangement order of all nodes are strictly maintained without any change, ensuring that the nodes on both contact surfaces are completely aligned in space after the projection is completed, without any misalignment or offset. After the projection is completed, all nodes on the contact surface of the active sealing component are uniformly grouped into active contact surface nodes, and all nodes on the contact surface of the driven sealing component are uniformly grouped into driven contact surface nodes. The number of the two types of nodes is completely equal and their positions are matched, thus obtaining the active contact surface nodes and driven contact surface nodes of the sealing pair.

[0042] Using all nodes on the active contact surface as the matching benchmark, for each active contact surface node, the node with the closest spatial straight-line distance is found among all nodes on the passive contact surface. A unique node pairing relationship is established for each active node. The matching process ensures that each active node corresponds to only one passive node, and each passive node also corresponds to only one active node. There are no many-to-one or one-to-many matching results. After all nodes are paired, they are arranged in order of spatial position to form a complete node pairing sequence. Then, the spatial adjacency relationship and topological connection relationship between all node pairs are sorted out to clarify the positional association between each node pair and its surrounding adjacent node pairs. Using node pairing as the basic constituent element and the association relationship between node pairs as the connection basis, a structured association topology structure is constructed. Finally, the corresponding information and topological connection relationship of all node pairings are carried in matrix form. The elements in the matrix correspond one-to-one with the node pairings and can directly reflect the association status and topological attributes between nodes, thus obtaining the node association matrix of the sealing pair.

[0043] The beneficial effects are that by conducting a global difference analysis of the morphological feature vectors, the degree of morphological change and feature distribution patterns at various points on the sealing surface can be accurately grasped, establishing a precise correspondence between morphological features and the spatial position of the contact interface. This provides a clear quantitative basis for the subsequent layout of differentiated nodes, ensuring a high degree of fit between the layout process and morphological features, avoiding insufficient accuracy or waste of resources caused by blindly uniform layout. Through hierarchical quantitative calibration, continuous difference values ​​can be transformed into clear grade identifiers. The number of grades and interval boundaries are determined based on sealing processing industry standards and simulation accuracy requirements, and the grading standards have a unified and objective judgment. Based on established criteria, a standardized difference-level map is formed, providing a clear and executable division standard for subsequent region segmentation, thus improving the accuracy and standardization of region segmentation. Density clustering segmentation based on the difference-level map can accurately distinguish the boundaries between regions with drastic morphological changes and regions with gentle morphological changes, forming two types of mapped regions with clear boundaries and complete coverage. This provides a clear basis for subsequent non-uniform node mapping. Based on the dimensional component sorting and non-uniform mapping completed by region segmentation, a partitioned adaptive mesh generation algorithm is introduced to standardize the overall node placement process, using the morphological difference level as a unified criterion for densifying sparse nodes. By clearly defining the node spacing rules for different regions, true topography-guided adaptive node deployment is achieved. In regions with complex topographic variations, contact nodes are densely deployed, ensuring the accuracy of contact characteristic calculations. This allows for precise capture of contact stress and state changes caused by subtle topographic undulations. In regions with gentle topography, node density is reasonably reduced to minimize unnecessary computational load. This effectively simplifies the overall node scale while maintaining simulation accuracy, reducing computational resource consumption and improving overall computational efficiency. Symmetrical projection of the contact surface on both sides ensures that the node positions on both sides are aligned, maintaining consistency and correspondence in node distribution. This provides a precise positional basis for subsequent node pairing, avoiding matching errors caused by node misalignment. The node association matrix formed by nearest neighbor matching and associated topology construction clearly carries the correspondence and topological association of all contact node pairs, forming a standardized and structured computational carrier. This provides a stable and standardized data structure for subsequent dynamic contact state timing loading, differential deduction, and other computational processes, effectively improving the computational efficiency and accuracy of subsequent dynamic contact deduction processes. From the perspectives of node deployment and data structure, this jointly supports the efficient and stable operation of the dynamic contact characteristic simulation of valve sealing pairs.

[0044] P4. Obtain the motion parameter sequence of the valve to be simulated during the opening and closing process, and load the motion parameter sequence onto the contact node pairs in the node association matrix according to the time step to obtain the dynamic contact state sequence of the contact node pairs. In this embodiment of the invention, obtaining the motion parameter sequence of the valve to be simulated during the opening and closing process, and loading the motion parameter sequence onto the contact node pairs in the node association matrix according to time steps to obtain the dynamic contact state sequence of the contact node pairs, includes: The driving timing parameters and displacement response parameters of the valve to be simulated during the opening and closing process are obtained, and the driving timing parameters and displacement response parameters are time-synchronized and merged to obtain the motion parameter sequence of the valve to be simulated. The motion parameter sequence is discretized by step size to obtain the time step sequence of the motion parameter sequence; Using the time step sequence as the advancement step size, the instantaneous motion parameter values ​​of the motion parameter sequence are extracted step by step, and the instantaneous motion parameter values ​​are coupled with the contact node pairs in the node association matrix to obtain the instantaneous contact state parameters of the contact node pairs. The instantaneous contact state parameters are systematically aggregated in a step-by-step manner to obtain the dynamic contact state sequence of the contact node pairs.

[0045] The system collects drive timing parameters for the entire opening and closing cycle of the valve to be simulated, from the fully closed state to the fully open state and back to the fully closed state. This covers the time nodes and action logic of all stages of valve opening (acceleration, constant speed, and deceleration) and closing (acceleration, constant speed, and buffer). Specifically, it includes all information related to the drive action time, such as the timing of the actuator's drive force application, drive rate switching nodes, drive stroke time nodes, and the action sequence arrangement of the opening and closing stages. Simultaneously, it collects all displacement response parameters of the active sealing component under the drive action, including the axial displacement, radial offset, relative gap change of the sealing surface, circumferential rotation angle, and other parameters of the active seal. All spatial position change information related to the relative motion of the sealing pair, such as the change in the cover's contact position, is collected. Then, using the start time of the valve's opening and closing action as a unified time zero point reference, all time nodes of the drive timing parameters and displacement response parameters are matched and aligned one by one. For parts of the two types of parameters with inconsistent sampling frequencies, time alignment interpolation is used to complete the parameter values ​​at the corresponding time. Invalid data segments with misaligned time axes and abnormal parameter correspondences are removed to ensure that each time scale contains the corresponding drive state information and displacement response information. Finally, a motion information set with a unified time reference, matching time sequence of the two types of parameters, and complete coverage of the entire valve opening and closing process is formed, resulting in the motion parameter sequence of the valve to be simulated.

[0046] A unified discrete time step is determined based on the total duration of the valve opening and closing process and the intensity of the motion state changes. The time step is uniformly measured in seconds, and the step size selection is based on the duration of the transient characteristics of the sealing pair contact. The overall main time series framework uses a fixed step size as the global main division interval. In the motion change interval, a local sub-step adaptive refinement method is used, inserting smaller sub-steps only between two adjacent reference main time nodes to improve sampling accuracy. The fixed step size interval between global reference main nodes remains consistent. Local refinement follows standardized execution rules. Intervals where sealing surface contact, sealing surface separation, or component displacement changes occur are identified as motion change intervals and refinement processing is initiated. Four sub-steps are equally inserted between any two adjacent reference main nodes. The duration between two reference main nodes is evenly divided into five segments, with equal intervals between each sub-segment. The duration of each sub-segment is one-fifth of the reference main step duration. No sub-steps are added in the smooth motion interval; only the reference main step is used in the calculation. The time step value is determined by combining the total valve opening and closing time with the transient characteristics of the sealing surface contact. Priority is given to extracting the key feature durations of sudden contact state changes during the valve opening and closing phase. The baseline main step size is set as several equal parts of the key feature durations to ensure that the drastic changes such as the contact and separation of the sealing surfaces during opening and closing are fully sampled and captured. For smooth motion intervals, only the baseline main step is used without adding sub-steps. The entire main timing framework uses a fixed-interval baseline main step as the backbone division mode. After division, it is verified that all key action transition nodes fall on the baseline main time node to avoid losing key motion details. The step size setting is verified through opening and closing motion characteristics to ensure that key motion state change points are covered by the baseline main node or local sub-steps. This prevents the loss of motion details due to an excessively large step size and avoids unnecessary computational burden due to an excessively small step size, thus ensuring complete capture of the opening and closing process. The process details all transitions in the motion state, covering all key motion nodes such as the moment of opening contact, the moment of full opening, and the moment of closing contact, without omitting any motion changes that affect the contact state. Then, taking the start time of the valve opening and closing process as the first reference master time node, the entire domain master time nodes are divided sequentially according to a determined uniform reference step size. All reference master nodes are arranged strictly in chronological order, and adjacent reference master nodes maintain a fixed step size interval. All reference master nodes completely cover the entire time interval of valve opening and closing, without any time discontinuities or overlapping intervals. In the region of abrupt motion change, sub-steps are added between adjacent reference master nodes, while in the region of non-abrupt motion change, only the reference master steps are retained. Finally, the reference master nodes and local sub-steps are integrated to form an ordered set of complete discrete time nodes, resulting in the time step sequence of the motion parameter sequence.

[0047] Following the sequential progression of the time step sequence, the instantaneous motion parameter values ​​in the motion parameter sequence at each corresponding time step are extracted in turn. These instantaneous values ​​include the specific relative displacement, spatial position, and motion rate of the active component of the sealing pair at that moment. Subsequently, these instantaneous motion parameter values ​​are loaded onto all contact node pairs in the node association matrix. A global synchronous loading method is adopted, with all contact node pairs synchronously substituted with the same set of instantaneous motion parameters within the same time step. During loading, the relative spatial position of each group of active and passive nodes is synchronously adjusted based on the double-sided node pairing topology stored within the node association matrix. The real-time gap between the two groups of nodes is used as the basic judgment criterion. Combined with the geometric and material coupling parameters in the basic attribute data obtained from the shape coupling mentioned above, multiple state indicators such as the degree of fit, contact stress, and effective contact area ratio are calculated synchronously. The entire state calculation logic is unified throughout and synchronously applies to all node pairs, without any situation where local nodes are calculated separately or parameters are loaded with lag. The parameter coupling loading process strictly follows the topological structure of the node association matrix. The relative position adjustment of each node pair is consistent with the overall motion displacement, and there is no node misalignment or loading deviation. The relative spatial distance and contact position of the active and driven nodes in each pair of contact nodes are adjusted according to the instantaneous displacement. The contact-related state information such as the degree of contact, contact intensity, and contact area ratio of each pair of contact nodes are calculated one by one in combination with the basic attribute data of the sealing pair. During the calculation process, the contact state is determined by combining the geometric morphology and material mechanical properties of the sealing pair to ensure that the calculation results of the instantaneous contact state are consistent with the actual contact law. Each contact node pair generates instantaneous contact state information corresponding to its spatial position. The instantaneous contact state information of all contact node pairs constitutes the complete contact state set at that time step, and the instantaneous contact state parameters of the contact node pair are obtained.

[0048] Following the sequential progression of the time step sequence, the instantaneous contact state parameters calculated for each time step are collected and organized in turn. The collection process strictly adheres to the chronological order to ensure that each set of instantaneous contact state parameters accurately corresponds to its respective time step, without any disordered order or time misalignment. During the collection process, the consistency of the number of node pairs and state dimensions for each time step is simultaneously verified to ensure that the contact state parameters of different time steps have the same structural dimension, facilitating subsequent time series comparison and differential calculation. After all the instantaneous contact state parameters for all time steps are collected, a set of contact state changes is formed, covering the complete opening and closing process of the valve and arranged in chronological order. The contact state of each contact node pair changes continuously with the time step, fully presenting the dynamic evolution of the contact state throughout the opening and closing process. The resulting dynamic contact state sequence can directly reflect the contact state of any node pair at any opening and closing moment, completely restoring the evolution of the contact state with the opening and closing action, thus obtaining the dynamic contact state sequence of the contact node pair.

[0049] The beneficial effects are as follows: By synchronously acquiring and merging the driving timing parameters and displacement response parameters, a unified time reference can be established, eliminating time misalignment between the two types of parameters, ensuring the consistency and accuracy of the motion parameters, and avoiding distortion in the contact state calculation caused by time axis mismatch. The full-stage coverage of driving and displacement parameters can completely restore the real motion process of valve opening and closing, providing a reliable and comprehensive motion input basis for subsequent dynamic contact simulation. By using a fixed-step discrete partitioning, the continuous opening and closing motion process is transformed into an ordered discrete time step sequence. The time step has a clear selection and judgment logic, using the transient characteristic duration of the sealing pair contact as the core reference standard, balancing simulation accuracy and overall computational load, clearly standardizing the complete process of time step partitioning, and decomposing the continuous motion process into discrete moments that can be calculated sequentially, providing a clear time-series advancement reference for the sequential calculation of the contact state. At the same time, the step size setting verified by motion characteristics can reasonably control the computational scale while ensuring the capture of all key motion details, ensuring the orderly and stable advancement of the dynamic simulation process. Instantaneous motion parameters are extracted step by step and compared with the contact nodes. Coupled loading employs a globally synchronized and unified loading mode, coupled with standardized node contact state synchronous calculation logic. At the same time, all paired nodes synchronously complete displacement updates and contact index solutions. Clear and complete rules for parameter transfer and state determination are established, enabling precise calculation of the specific contact state of each pair of contact nodes at every moment. Strict adherence to the node topology loading method ensures that node displacement remains consistent with the overall motion. Combined with the state calculation of the basic properties of shape coupling, the results are guaranteed to closely match actual contact laws, ensuring a precise correspondence between contact state and motion parameters. This accurately reflects the real-time impact of motion displacement on the contact state, improving the precision and accuracy of instantaneous contact state calculations. Through a step-by-step, orderly aggregation of dynamic contact state sequences, the continuous evolution of contact states throughout the valve's opening and closing process can be fully presented, forming a temporally coherent, structurally unified, and comprehensively covered contact state dataset. This provides a continuous, complete, and structurally standardized data foundation for subsequent temporal differential derivation and full-process cumulative reconstruction, effectively ensuring the coherence and accuracy of dynamic contact characteristic simulation results, and comprehensively improving the realism and computational reliability of dynamic contact characteristic simulation during the valve sealing pair's opening and closing process.

[0050] P5. Perform time-series difference deduction on the state changes between adjacent time steps in the dynamic contact state sequence to obtain the contact transient changes of the sealing pair; In this embodiment of the invention, the step of performing time-series differential deduction on the state changes between adjacent time steps in the dynamic contact state sequence to obtain the contact transient changes of the sealing pair includes: Along the time step direction, the dynamic contact state sequence is stripped by step differential analysis to obtain the state change of the contact node pair between adjacent time steps; By performing hierarchical trend fitting on the state change, the evolution trend of the change in the contact node pair is obtained; The transient characteristics of the evolution trend of the change are extracted to obtain the contact transient change of the sealing pair.

[0051] The step of performing hierarchical trend fitting on the state change to obtain the evolution trend of the change in the contact node pair includes: The state change quantity is divided into windows to obtain a time window sequence of the state change quantity; Within each time window of the time window sequence, the state change is fitted using a window rate to obtain a window trend value for the state change. The formula for calculating the window trend value is as follows: ; in, Indicates the first The contact node pair in the th ... Window trend values ​​within a time window, Indicates the first The contact node pair in the th ... Within the first time window The state change at each time step Indicates the first The contact node pair in the th ... The state change at the first time step within a time window Indicates the first The time interval between the first time step and the second time step. Indicates the first The number of time steps contained within a time window; The window trend value is recursively extrapolated window by window to obtain the evolution trend of the change in the contact node pair.

[0052] Following the sequential progression of time steps, starting from the opening / closing start time step corresponding to the dynamic contact state sequence, differential calculations are performed on each pair of adjacent time steps. For all contact node pairs in the node association matrix, all contact state dimensions of the corresponding node pairs under the two time steps are compared one by one, including core state items such as contact fit degree, contact intensity, and effective contact area ratio. The difference between the state value of the later time step and the state value of the previous time step is calculated item by item to clarify the magnitude and direction of the change in contact state of each contact node pair between adjacent time steps. A positive difference represents an enhanced contact effect, and a negative difference represents a weakened contact effect. After the state difference calculations of all node pairs corresponding to all adjacent time steps are completed, they are strictly arranged and collected according to the chronological order and the spatial position of the nodes to form a complete set of state difference values ​​covering the entire valve opening and closing process and the entire contact interface, thus obtaining the state change of the contact node pair between adjacent time steps.

[0053] Based on the overall time span of the state change and the intensity of the contact state change, a time window length and a fixed window sliding step size are determined, containing a fixed number of continuous state change quantities. Under normal simulation conditions, a single time window is uniformly set to contain data from eight consecutive time steps, and the window sliding step size is set to half of the total window length. A 50% data overlap ratio is maintained between adjacent time windows. The window length, sliding step size, and overlap ratio are all uniformly calibrated based on the valve opening and closing motion fluctuation frequency to ensure the window smoothing filtering effect without causing time series data redundancy. All parameters are uniformly applicable to the division and processing of state change quantities throughout the entire process. The window length setting ensures that it can cover a preset number of single-step change data to achieve a smoothing effect. The sliding step size setting ensures that a fixed proportion of overlapping data segments are retained between adjacent windows. Starting from the beginning position of the state change quantity sequence, continuous state change quantity data groups are sequentially extracted according to the set sliding step size. Each group of data corresponds to an independent time window. All time windows are arranged in strict chronological order to ensure that the original baseline time series data is completely covered and there is no omission of baseline time series intervals. A preset overlapping data segment is retained between adjacent windows to achieve smoothing processing, resulting in the time window sequence of the state change quantities.

[0054] For each time window in the time window sequence, taking the moment corresponding to the first state change within the window as the starting point for calculation, the change amplitude and corresponding time interval of all state changes within the window are analyzed one by one. The overall trend and average rate of change of contact state within the window are statistically analyzed. By comprehensively calculating multiple sets of continuous data within the window, the abnormal deviation caused by single-step random fluctuations is weakened, the interference of single-point data mutations on the overall trend judgment is eliminated, and a feature value that can represent the overall trend of contact state changes within the window is generated. Each contact node pair generates a unique window trend feature value within each time window, thus obtaining the window trend value of the state change.

[0055] The window trend value corresponds to each contact node pair in the node association matrix and each time window in the time window sequence. The window trend value uses a dimension-by-dimensional independent calculation mode, applying a formula separately to calculate the window trend component for each dimension of the contact state parameter. The default output format is the dimension-by-dimensional component, used for detailed monitoring of contact state in each dimension. In the overall global state assessment scenario, a comprehensive scalar fusion mode is enabled. Based on overall analysis needs, the results of each dimension can be further fused to obtain a single comprehensive scalar trend value. The fusion method uses a normalized equal-weighted average method, first normalizing the dimensions of each trend component, and then calculating the equal-weighted average. Among these, the state change... For multidimensional contact parameters, different parameters adopt corresponding physical dimensions. The unit for contact stress parameter is megapascal, the unit for contact gap parameter is millimeter, and the unit for time interval is... The unit is seconds; the final dimension-wise result represents the average rate of change of the contact parameter of the corresponding dimension within the current window for that node, and the comprehensive scalar represents the average rate of change of the overall contact state. The contact node pair is derived from the node pairing set formed after the active contact surface node and the passive contact surface node complete the nearest neighbor matching, and is carried in the topological structure of the node association matrix. The time window is derived from the time window sequence obtained by dividing the state change quantity into windows. Each time window corresponds to a continuous opening and closing time interval. The state change quantity is derived from the contact state difference between adjacent time steps obtained by step differential stripping of the dynamic contact state sequence. Each time window contains a fixed number of continuous state change quantities, which correspond to the contact state change results of each time step within the window. All state change quantities are precisely corresponding to specific contact node pairs and time steps. The time interval data is derived from the time step sequence, corresponding to the time difference between each time step within the window and the first time step of the window. It is obtained by sequentially accumulating the fixed step long step sequence obtained by discrete division. The total number of time steps contained in the window is directly determined by the pre-set time window length, and the number of time steps in all time windows remains uniform.

[0056] The core function of this calculation process is to smooth out the random fluctuations caused by single-step discrete calculations by comprehensively calculating multiple sets of continuous state change data within a window. This yields characteristic values ​​that represent the overall rate of contact state change within the time window. The calculation uses the state change corresponding to the first time step within the window as a benchmark reference value. The difference between the state change at each subsequent time step and the benchmark reference value is calculated sequentially. Each difference is then compared with the corresponding time interval to obtain the rate of change of each time step relative to the benchmark time. Finally, the arithmetic mean of all relative rates of change within the window is calculated to obtain the window trend value of the corresponding contact node pair within the window. This directly reflects the overall trend and average rate of change of contact state within the corresponding time window, effectively weakening the interference of single-point data mutations on the overall trend judgment and improving the smoothness and reliability of trend representation.

[0057] The calculated window trend value clearly reflects the overall trend direction and rate of change of the state of the corresponding contact node pair within the time window. When the calculation result is positive, it means that the amount of contact state change within the window is increasing, the magnitude of contact state change is continuously increasing, and the rate of change of contact action is gradually accelerating. When the calculation result is negative, it means that the amount of contact state change within the window is decreasing, the magnitude of contact state change is continuously decreasing, and the rate of change of contact action is gradually slowing down. When the calculation result approaches zero, it means that the amount of contact state change within the window remains stable, the magnitude of contact state change is basically constant, and the rate of change of contact action tends to be stable. After connecting and recursively extrapolating the window trend values ​​corresponding to all time windows in chronological order, the overall evolution trend of the state change of the contact node pair with the entire process of valve opening and closing can be fully presented. This provides a smooth and continuous trend foundation for the construction of subsequent change trend and supports the accurate condensation of subsequent contact transient change.

[0058] According to the sequential arrangement of the time window sequence, the trend values ​​of all windows corresponding to each contact node pair are sequentially connected and recursively extrapolated. The end trend state of the overlapping segment of the previous time window is used as the starting trend benchmark of the next time window. The trend values ​​at the window connection position are smoothed by using the common data of the window overlapping area to eliminate possible trend abrupt changes and discontinuities at the window connection. Finally, a continuous and smooth trend trajectory covering the entire opening and closing process of the valve is formed, which fully presents the overall evolution trend and change law of the state change of each contact node pair as the opening and closing time progresses, and obtains the change trend of the contact node pair.

[0059] For the evolution trend of the change in each contact node pair, the instantaneous change features at the corresponding moment are extracted point by point along the time step sequence. Combining the trend direction and instantaneous change rate at that moment, the transient intensity and direction of the contact state change at that time step are condensed. The transient change features of all contact node pairs are integrated in the whole domain according to the spatial location arrangement and time sequence to form a complete transient change data set covering the entire contact interface and the entire opening and closing process. This set can accurately reflect the instantaneous change of the contact state at any moment and any position during the opening and closing process, and obtain the contact transient change amount of the sealing pair.

[0060] The beneficial effects include the ability to accurately extract the contact state difference between adjacent time steps through step-by-step differential stripping, intuitively presenting the amplitude and direction of contact state changes brought about by each step of motion displacement, providing basic change data support for subsequent trend analysis, ensuring the temporal and spatial correspondence between the change data and the original contact state sequence, and using a hierarchical trend fitting method based on time window division and window rate fitting to effectively eliminate random fluctuations and abnormal deviations caused by single-step calculations through comprehensive calculation of multiple sets of continuous data, smoothly restoring the overall evolution of contact state changes, avoiding result oscillations caused by discrete calculations, and improving the continuity and reliability of the change trend. The overlapping number is utilized during the window-by-window trend recursion process. The smoothing process ensures the continuity and consistency of the entire process trend, fully presenting the evolution of state changes during the opening and closing process, avoiding trend discontinuities at window boundaries, and improving the overall smoothness and realism of the trend trajectory. The contact transient changes obtained through transient feature condensation can accurately capture the instantaneous changes in the contact state during the opening and closing process, clearly reflecting the speed and direction of contact state changes. This provides accurate transient data support for subsequent full-process cumulative reconstruction and morphological collaborative interpretation, effectively improving the precision and accuracy of dynamic contact characteristic analysis, fully restoring the transient evolution of the contact state during the opening and closing of the valve sealing pair, and supporting the accurate generation of subsequent dynamic contact characteristic distribution maps.

[0061] P6. Perform full-process cumulative reconstruction of the contact transient change, and co-interpret the reconstructed cumulative contact field with the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair.

[0062] In this embodiment of the invention, the step of performing full-process cumulative reconstruction of the contact transient change, and co-interpreting the reconstructed cumulative contact field with the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair includes: Using the time step sequence as the entire time series axis, the contact transient changes are sequentially accumulated and merged, and the merged results are aggregated and reconstructed to obtain the cumulative contact field of the sealing pair. The field topology of the cumulative contact field is deconstructed to obtain the field structure characteristics of the cumulative contact field; Cross-domain feature coupling is performed between the field structure features and the morphological feature vector to obtain the overlapping and related features between the field structure features and the morphological feature vector; A global map of the overlapping and correlated features is plotted to obtain a dynamic contact characteristic distribution map of the sealing pair.

[0063] Using the complete valve opening and closing time interval corresponding to the time step sequence as the full-process time axis, and the initial contact state of the sealing pair when there is no relative movement at the start of opening and closing as the baseline initial state for cumulative calculation, sequential cumulative calculation is carried out step by step along the time progression direction of the time axis, starting from the first time step. For each contact node pair in the node association matrix, the contact transient change corresponding to the current time step is added to the cumulative contact state of the node pair at the end of the previous time step. The cumulative contact intensity, cumulative action duration ratio, cumulative deformation degree, and other related cumulative state values ​​of each contact node pair are updated sequentially. After traversing all time steps, The sequential accumulation and merging of all contact transient changes during the entire opening and closing process is completed to obtain the cumulative state data of all discrete contact node pairs throughout the entire process. Then, using the topological distribution of the node correlation matrix as the spatial reference, the cumulative state data of all discrete contact node pairs are continuously aggregated and reconstructed according to the spatial topological distribution relationship of the contact interface. Spatial interpolation is used to fill the field gaps between nodes to ensure a smooth transition of field quantities between adjacent nodes without numerical abrupt changes or discontinuities. Finally, a continuous field data set that completely covers the entire contact interface of the sealing pair and can reflect the cumulative effect of contact action throughout the entire opening and closing process is formed, thus obtaining the cumulative contact field of the sealing pair.

[0064] A systematic topological decomposition analysis was conducted on the global field data of the cumulative contact field. The magnitude, distribution density, spatial gradient, and aggregation degree of the field quantities were statistically analyzed region by region and point by point along the circumferential extension and radial distribution directions of the contact interface. Fixed graded thresholds were defined according to the numerical range of the cumulative contact intensity, uniformly divided into five levels. The threshold boundaries were based on the allowable contact stress of the sealing pair and the effective bearing range of the seal as basic references. First, the maximum and minimum values ​​of the global cumulative contact intensity were extracted to achieve full range coverage. Then, segmented calibration was performed in conjunction with the safe stress range allowed by the seal structure design. Each threshold boundary was initially uniformly divided. After the initial uniform division, the boundary fine-tuning was performed using the seal failure judgment benchmark as the core basis. The single fine-tuning amplitude did not exceed five percent of the initial interval width, and this process was repeated. The verification continues until the cumulative contact effect within each range has a similar impact on seal wear and deformation. Different ranges can clearly distinguish high-load contact areas, medium-load transition areas, and low-load light contact areas. Based on the graded threshold, high-value clustered areas, low-value flat areas, and intensity transition areas between high and low values ​​of the cumulative contact effect within the field are identified. The boundary contour morphology, spatial distribution location, area ratio, and interconnection relationship and adjacency order of each area are analyzed one by one. From this, core feature items that can represent the spatial distribution law and structural morphology of the entire cumulative contact field are extracted. All feature items completely cover the intensity distribution characteristics, regional division characteristics, boundary morphology characteristics, and topological connection characteristics of the field, without any omission of key structural features, thus obtaining the field structure characteristics of the cumulative contact field.

[0065] A one-to-one spatial correspondence is pre-established between the spatial locations of each region of the cumulative contact field and the components of each dimension of the morphological feature vector. This ensures that every region of the field structure features can be accurately matched with the corresponding morphological feature component. Each regional feature in the field structure features is matched and associated with the corresponding morphological feature component one by one. The entire collaborative interpretation process is divided into multiple progressive processing steps. The first step is to complete the spatial location alignment and matching operation, based on the spatial mapping relationship established in the previous point layout stage, to complete the one-to-one correspondence between the field regions and the morphological components. The second step is to conduct bidirectional feature association analysis to simultaneously extract the synchronous change law of the morphological undulation trend and the strength distribution of the contact field. The third step is to select the overlapping regions where the two types of features simultaneously show extreme values ​​or abrupt changes as the core association feature carriers. Throughout the process, the field data and the shape vector dimension are kept aligned in a unified manner to avoid spatial misalignment or missing feature dimensions. The intrinsic relationship between the undulations, contours, texture distribution, and the distribution, region division, and gradient changes of the contact accumulation field is analyzed in depth. The correspondence between changes in shape features and changes in contact accumulation effect is verified one by one. The correlation feature items that are highly overlapping in spatial distribution and correspond to each other in action logic are extracted to clarify the influence path and action law of shape features on contact accumulation effect. All correlation feature items are orderly collected according to spatial location to form a feature set that can completely reflect the synergistic relationship between shape features and contact accumulation field, and the overlapping correlation features of the field structure features and the shape feature vector are obtained.

[0066] The planar projection morphology of the sealing interface is used as the spatial basis for the atlas drawing. The base outline is completely consistent with the actual outline of the sealing surface. According to the spatial position corresponding to the overlapping and related features, the information on the cumulative contact intensity, dynamic change trend, and corresponding morphological influence law is transformed into standardized visual atlas elements. A hierarchical coloring method is used to characterize the strength of dynamic contact characteristics, with the color depth strictly corresponding to the strength of the characteristics. Clear outline markers are superimposed to indicate the boundaries, distribution positions, and characteristic types of key feature areas. The drawing range completely covers all areas of the entire contact interface, with no blank or omitted areas. All atlas elements correspond one-to-one with their actual spatial positions. The output of the map follows a unified output rule, simultaneously outputting two types of results: layered data files and visualization images. The data files store the cumulative contact values, morphological feature parameters, and the correlation and matching level of each location in spatial point order. The visualization images simultaneously label key areas such as high stress accumulation areas and morphological change areas, and include corresponding correlation explanations. All intermediate feature data generated during the interpretation process are completely retained for retrospective analysis. Finally, a visualization map is formed that can intuitively present the spatial distribution law of the dynamic contact characteristics of the sealing pair during the entire valve opening and closing process, and at the same time reflect the influence of morphological features on the contact characteristics, thus obtaining the dynamic contact characteristic distribution map of the sealing pair.

[0067] The beneficial effects are as follows: by using a full-epoch procedure based on time-step sequences for cumulative aggregation and reconstruction, discrete transient data can be integrated into a continuous global contact cumulative field. This fully restores the overall cumulative effect of contact action throughout the entire valve opening and closing process, breaking the limitation that transient data can only reflect the state at a single moment. It achieves a complete characterization of the effect of contact action throughout the entire process. The aggregation and reconstruction process uses spatial interpolation to fill in node gaps, ensuring a smooth transition of field values ​​and effectively improving the continuity and data reliability of the cumulative contact field. Through field topology deconstruction, the core structural distribution can be extracted from the global field data. This approach features streamlined field data dimensions, standardized judgment rules for contact strength grading thresholds, and a clearly defined number of grading levels. Threshold boundaries are determined by combining multiple conditions, including allowable sealing stress, extreme values ​​of the overall field quantity, and sealing failure criteria. The grading rules are unified and objective, ensuring consistent standards for the division of cumulative field regions under different working conditions. This improves the reproducibility and feasibility of the field structure feature extraction process, clearly presenting the spatial distribution patterns and regional structural characteristics of the cumulative contact field. It provides a clear and concise feature carrier for subsequent cross-domain feature correlation, reducing the computational complexity of subsequent correlation analysis. This solution adopts a hierarchical and progressive approach. This standardized collaborative interpretation process clarifies the spatial alignment and matching, bidirectional correlation analysis, and overlapping feature screening between the cumulative contact field and morphological feature vectors. It also specifies unified dual-type output rules for data and images, clearly defining the operational logic and final output format for the fusion and correlation of the two types of features. This addresses the shortcomings of previous technologies, such as ambiguous interpretation logic, unclear feature correlation methods, and lack of unified output standards. It establishes the intrinsic correlation between the morphological features of the sealing surface and the cumulative contact characteristics, clarifies the influence mechanism of morphological factors on contact characteristics, and explains the distribution and evolution of the contact field from the root cause of morphology. This overcomes the limitations of traditional simulations, which can only obtain contact results but cannot correlate morphological causes, effectively improving the analytical depth and engineering guidance value of simulation results. Through full-domain mapping, abstract feature data is transformed into intuitive and visual distribution maps, clearly presenting the spatial distribution patterns and morphological influence relationships of dynamic contact characteristics. This facilitates engineers' rapid understanding of the contact performance characteristics and weak areas of the sealing pair, providing intuitive and comprehensive data support and decision-making basis for the structural optimization, performance improvement, and failure analysis of valve sealing pairs. Overall, it enhances the engineering application value and practical guidance significance of the simulation results of the dynamic contact characteristics of valve sealing pairs.

[0068] like Figure 2 The diagram shown is a functional block diagram of a simulation system for the dynamic contact characteristics of a valve sealing pair during the opening and closing process, provided in an embodiment of the present invention.

[0069] The simulation system 100 for the dynamic contact characteristics of a valve sealing pair during opening and closing, as described in this invention, can be installed in an electronic device. Depending on the functions implemented, the simulation system 100 may include a shape coupling module 101, a contour encoding module 102, a point matching module 103, a timing loading module 104, a differential deduction module 105, and a collaborative interpretation module 106. The module described in this invention can also be called a unit, referring to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0070] In this embodiment, the functions of each module / unit are as follows: The shape coupling module 101 is used to couple the geometric parameters and material parameters of the sealing pair in the valve to be simulated with shape parameters to obtain the basic property data of the sealing pair; The contour encoding module 102 is connected to the shape coupling module and is used to extract the sealing surface contour data from the basic attribute data and perform contour feature encoding on the sealing surface contour data to obtain the shape feature vector of the sealing pair. The point matching module 103 is connected to the contour encoding module and is used to perform shape-guided point placement on the contact interface of the sealing pair based on the shape feature vector, and to perform pairing mapping on the contact nodes after point placement to construct the node association matrix of the sealing pair. The timing loading module 104 is connected to the point pairing module and is used to obtain the motion parameter sequence of the valve to be simulated during the opening and closing process, and load the motion parameter sequence onto the contact node pairs in the node association matrix according to the time step to obtain the dynamic contact state sequence of the contact node pairs. The differential deduction module 105 is connected to the timing loading module and is used to perform timing differential deduction on the state change between adjacent time steps in the dynamic contact state sequence to obtain the contact transient change of the sealing pair. The collaborative interpretation module 106 is connected to the differential inference module and is used to perform full-process cumulative reconstruction of the contact transient change, and to perform collaborative interpretation of the reconstructed cumulative contact field and the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair.

[0071] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0072] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0073] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0074] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for simulating the dynamic contact characteristics of a valve sealing pair during the opening and closing process, characterized in that, The method includes: P1. Couple the geometric parameters and material parameters of the sealing pair in the valve to be simulated with shape parameters to obtain the basic property data of the sealing pair; P2. Extract the sealing surface contour data from the basic attribute data, and encode the contour features of the sealing surface contour data to obtain the shape feature vector of the sealing pair. P3. Based on the morphological feature vector, morphological guidance points are placed on the contact interface of the sealing pair, and the contact nodes after placement are paired and mapped to construct the node association matrix of the sealing pair. P4. Obtain the motion parameter sequence of the valve to be simulated during the opening and closing process, and load the motion parameter sequence onto the contact node pairs in the node association matrix according to the time step to obtain the dynamic contact state sequence of the contact node pairs. P5. Perform time-series differential deduction on the state changes between adjacent time steps in the dynamic contact state sequence to obtain the contact transient changes of the sealing pair; P6. Perform full-process cumulative reconstruction of the contact transient change, and co-interpret the reconstructed cumulative contact field with the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair.

2. The method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 1, characterized in that, The process of coupling the geometric and material parameters of the sealing pair in the valve to be simulated with morphological parameters to obtain the basic property data of the sealing pair includes: Extract the geometric and material parameters of the sealing pair in the valve to be simulated; The geometric parameters and the material parameters are decoupled by intrinsic features to obtain the morphology-dominant component of the geometric parameters and the property-dominant component of the material parameters. Heterogeneous feature matching is performed on the morphology dominant component and the attribute dominant component to obtain the morphological correspondence between the morphology dominant component and the attribute dominant component; Based on the shape correspondence, the shape dominant component and the attribute dominant component are reconstructed by basis weighting to obtain the basis coupling parameters of the shape dominant component and the attribute dominant component. The base coupling parameters are uniformly normalized to obtain the basic attribute data of the sealing pair.

3. The method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 1, characterized in that, The step of extracting sealing surface contour data from the basic attribute data and encoding the contour features of the sealing surface contour data to obtain the morphological feature vector of the sealing pair includes: The basic attribute data is subjected to domain feature analysis to obtain the geometric structure features and material property features of the basic attribute data; Based on the material properties, the geometric features are guided to undergo contour calibration to obtain the sealing surface contour data of the sealing pair. The contour direction of the sealing surface contour data is encoded to obtain the contour direction code sequence of the sealing surface contour data; The contour direction code sequence is subjected to principal component feature aggregation to obtain the shape feature vector of the sealing pair.

4. The method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 1, characterized in that, The step of guiding the placement of points on the contact interface of the sealing pair based on the morphological feature vector, and then pairing and mapping the contact nodes after placement to construct the node association matrix of the sealing pair includes: A difference analysis is performed on the morphological feature vectors to obtain the morphological difference distribution of the morphological feature vectors; Based on the morphological difference distribution, the dimensional components of the morphological feature vector are mapped onto the contact interface of the sealing pair to obtain the encrypted contact nodes and sparse contact nodes of the contact interface. By projecting the encrypted contact node and the sparse contact node onto both sides of the contact surface, the active contact surface node and the passive contact surface node of the sealing pair are obtained. The active contact surface node and the passive contact surface node are matched for nearest neighbor, and the node association matrix of the sealing pair is obtained by constructing an association topology from the matched node pairing sequence.

5. The method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 4, characterized in that, The step of mapping the dimensional components of the morphological feature vector to the contact interface of the sealing pair according to the morphological difference distribution to obtain the encrypted contact nodes and sparse contact nodes of the contact interface includes: The morphological difference distribution is hierarchically quantified to obtain a difference grading map of the contact interface; Based on the difference gradation map, the contact interface is segmented by density clustering to obtain the high-density mapping region and the low-density mapping region of the contact interface. Based on the high-density mapping region and the low-density mapping region, the dimensional components of the shape feature vector are sorted into dense mapping components and sparse mapping components. Non-uniform mapping is performed on the dense mapping component and the sparse mapping component to obtain the encrypted contact node and sparse contact node of the contact interface.

6. The method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 1, characterized in that, The step of obtaining the motion parameter sequence of the valve to be simulated during the opening and closing process, and loading the motion parameter sequence onto the contact node pairs in the node association matrix according to time steps to obtain the dynamic contact state sequence of the contact node pairs, includes: The driving timing parameters and displacement response parameters of the valve to be simulated during the opening and closing process are obtained, and the driving timing parameters and displacement response parameters are time-synchronized and merged to obtain the motion parameter sequence of the valve to be simulated. The motion parameter sequence is discretized by step size to obtain the time step sequence of the motion parameter sequence; Using the time step sequence as the advancement step size, the instantaneous motion parameter values ​​of the motion parameter sequence are extracted step by step, and the instantaneous motion parameter values ​​are coupled with the contact node pairs in the node association matrix to obtain the instantaneous contact state parameters of the contact node pairs. The instantaneous contact state parameters are systematically aggregated in a step-by-step manner to obtain the dynamic contact state sequence of the contact node pairs.

7. The method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 1, characterized in that, The step of performing time-series differential deduction on the state changes between adjacent time steps in the dynamic contact state sequence to obtain the contact transient changes of the sealing pair includes: Along the time step direction, the dynamic contact state sequence is stripped by step differential analysis to obtain the state change of the contact node pair between adjacent time steps; By performing hierarchical trend fitting on the state change, the evolution trend of the change in the contact node pair is obtained; The transient characteristics of the evolution trend of the change are extracted to obtain the contact transient change of the sealing pair.

8. The method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 7, characterized in that, The step of performing hierarchical trend fitting on the state change to obtain the evolution trend of the change in the contact node pair includes: The state change quantity is divided into windows to obtain a time window sequence of the state change quantity; Within each time window of the time window sequence, the state change is fitted using a window rate to obtain a window trend value for the state change. The formula for calculating the window trend value is as follows: ； in, Indicates the first The contact node pair in the th ... Window trend values ​​within a time window, Indicates the first The contact node pair in the th ... Within the first time window The state change at each time step Indicates the first The contact node pair in the th ... The state change at the first time step within a time window Indicates the first The time interval between the first time step and the second time step. Indicates the first The number of time steps contained within a time window; The window trend value is recursively extrapolated window by window to obtain the evolution trend of the change in the contact node pair.

9. The method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 6, characterized in that, The process of performing full-process cumulative reconstruction of the contact transient changes, and then co-interpreting the reconstructed cumulative contact field with the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair includes: Using the time step sequence as the entire time series axis, the contact transient changes are sequentially accumulated and merged, and the merged results are aggregated and reconstructed to obtain the cumulative contact field of the sealing pair. The field topology of the cumulative contact field is deconstructed to obtain the field structure characteristics of the cumulative contact field; Cross-domain feature coupling is performed between the field structure features and the morphological feature vector to obtain the overlapping and related features between the field structure features and the morphological feature vector. A global map of the overlapping and correlated features is plotted to obtain a dynamic contact characteristic distribution map of the sealing pair.

10. A simulation system for the dynamic contact characteristics of a valve sealing pair during the opening and closing process, characterized in that, A method for simulating the dynamic contact characteristics of a valve sealing pair during opening and closing as described in claim 1, the system comprising: The shape coupling module is used to couple the geometric parameters and material parameters of the sealing pair in the valve to be simulated with shape parameters to obtain the basic property data of the sealing pair; The contour encoding module, connected to the shape coupling module, is used to extract the sealing surface contour data from the basic attribute data and perform contour feature encoding on the sealing surface contour data to obtain the shape feature vector of the sealing pair. The point matching module, connected to the contour encoding module, is used to perform shape-guided point placement on the contact interface of the sealing pair based on the shape feature vector, and to perform pairing mapping on the contact nodes after point placement to construct the node association matrix of the sealing pair. The timing loading module, connected to the point pairing module, is used to obtain the motion parameter sequence of the valve to be simulated during the opening and closing process, and load the motion parameter sequence onto the contact node pairs in the node association matrix according to the time step to obtain the dynamic contact state sequence of the contact node pairs. The differential deduction module, connected to the timing loading module, is used to perform timing differential deduction on the state changes between adjacent time steps in the dynamic contact state sequence to obtain the contact transient changes of the sealing pair. The collaborative interpretation module, connected to the differential inference module, is used to perform full-process cumulative reconstruction of the contact transient change, and to perform collaborative interpretation of the reconstructed cumulative contact field with the morphological feature vector to obtain the dynamic contact characteristic distribution map of the sealing pair.

Images