Contact health management system based on digital twin

Abstract

The present invention relates to the field of digital twin technology, specifically to a contact health management system based on digital twins, which includes a working condition injection module, a wear mapping module, an offset extraction module, a trend characterization module, and an anomaly capture module. In the present invention, by eliminating data segments that lack synchronization and continuity, a stable trajectory sequence is constructed, the data basis for virtual-real mapping is enhanced, a multi-parameter structure is introduced into the trajectory surface, and the coordinated expression of state changes is achieved. The pressure gradient is extracted in the contact pressure and current intersection section and the trajectory is corrected in combination with the temperature factor to improve the dynamic adaptability of the trajectory to wear evolution. The difference analysis accurately characterizes the mutation behavior, the main axis offset mapping strengthens the behavior path backtracking association, and the action duration and current fluctuation characteristics are combined to calibrate the abnormal points. A health assessment system with multi-layer recognition capabilities is constructed to improve the accuracy of contact state recognition.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and in particular to a contact health management system based on digital twin. Background Art

[0002] The field of digital twin technology involves building virtual digital models of physical objects or systems, and managing and controlling physical objects through monitoring, simulation, prediction, and optimization of these virtual models. The core content of this technology includes dynamic mapping between physical entities and digital models, data collection and integration, operational status simulation, predictive analysis, and optimized decision-making. Its overall technical field involves comprehensive perception of the operational status of physical entities, virtual-real synchronization and interaction, and dynamic simulation and health management driven by real-time data. By integrating the Internet of Things, edge computing, and artificial intelligence, it supports application scenarios such as intelligent manufacturing, smart energy, and intelligent transportation, and achieves full lifecycle management of complex systems.

[0003] Among them, the contact health management system based on digital twins refers to the use of digital twin models to monitor and evaluate the operating status and health status of contact components in electrical equipment. Specifically, it covers the establishment of dynamic physical properties and behavior models of contact components based on real-time collected operating data such as temperature, current, wear, mechanical movement, etc., and the application of mechanism modeling and simulation calculation technology to predict and analyze contact wear, deterioration and life changes. The state parameters of the physical object are continuously adjusted through the digital twin model to form a virtual-real mapping and dynamically updated contact health management mechanism, and the contact anomaly identification and health assessment are carried out in combination with methods based on threshold judgment and empirical rules.

[0004] Existing technologies for contact health management fail to adequately capture behavioral evolution, making it difficult to form a quantifiable dynamic trajectory structure. The lack of a systematic trajectory filtering mechanism makes input data susceptible to occasional perturbations, resulting in fragmented and erratic trajectory representations, hindering the effectiveness of digital modeling. Parameter relationships are often handled in parallel, lacking a structured interactive logic. This limits the trajectory's ability to respond to behavioral path changes. At the anomaly detection level, deviation trends are not considered as a behavioral determinant, relying instead on current state values, failing to capture potential risks caused by early deviation trends. Trajectory retrospective analysis lacks mapping path support for deviation behaviors, making it difficult to accurately map action segments to contact states. During health assessment, outlier identification often relies on empirical thresholds, failing to integrate behavioral feature density and growth rate analysis, resulting in delayed recognition. In complex operating environments, such as those with frequent load fluctuations or severe wear evolution, identification bias and intervention delays are common, leading to reduced equipment operational stability and increased maintenance costs. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a contact health management system based on digital twins.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solutions: The contact health management system based on digital twins includes:

[0007] The working condition injection module obtains the current, contact pressure, and temperature data during the contact opening and closing operation, determines the synchronization and continuity, eliminates abnormal segments, forms a trajectory sequence, inputs the digital twin mapping body, constructs three sets of trajectory surface, and obtains the associated trajectory combination;

[0008] The wear mapping module extracts the intersection segment with the current change based on the contact pressure trajectory in the associated trajectory combination, calculates the matching degree between the pressure gradient and the current fluctuation, and adjusts the trajectory in combination with the temperature change to form an overlapping extension structure;

[0009] The offset extraction module extracts the non-overlapping period of the contact pressure and current difference in the overlapping extension structure, analyzes the difference change and amplitude according to time, identifies the mutation section, and obtains the boundary mutation section;

[0010] The trend characterization module calls the closed cycle segment data in the boundary mutation section to analyze the contact surface state, pressure change and wear trajectory, perform intersection offset analysis, calculate the spindle offset, map it to the original motion segment, and obtain the spindle track offset result;

[0011] The abnormality capture module analyzes the action duration and current fluctuation based on the closed segment where the spindle track deviation result is located, identifies the intersection density and amplification changes, marks the abnormal points, and outputs the contact health management results.

[0012] As a further solution of the present invention, the associated trajectory combination includes synchronous trajectory segments, continuous trajectory segments, and three groups of mapping surfaces; the overlapping extension structure includes pressure change gradient value, current fluctuation matching degree, and temperature adjustment factor; the boundary mutation segment includes contact pressure difference sequence, current difference sequence, and mutation change amplitude; the spindle track offset result includes spindle offset value, intersection point offset characteristics, and contact surface pressure change; the contact health management result includes action duration characteristics, current fluctuation index, intersection density, and amplification abnormality point.

[0013] As a further solution of the present invention, the working condition injection module includes:

[0014] The data screening submodule obtains the working current, contact pressure and ambient temperature data during the contact opening and closing period, determines the synchronization and continuity of the three data in the time dimension, and generates the total amount of synchronous continuous data segments;

[0015] The trajectory construction submodule calls the total amount of the synchronous continuous data segments, connects the data segments in time series, and screens the connectable segments based on the working current, contact pressure difference and ambient temperature change trend to generate a trajectory connection set;

[0016] The facet establishment submodule maps the multiple feature combinations of the trajectories in the trajectory connection set, divides the state sequences into three groups according to the time nodes, inputs them into the digital twin mapping body, establishes three groups of trajectory facets, and generates associated trajectory combinations.

[0017] As a further solution of the present invention, the wear mapping module includes:

[0018] The pressure trajectory extraction submodule extracts the pressure change point sequence in each trajectory unit based on the contact pressure trajectory data in the associated trajectory combination, calculates the pressure change difference between adjacent points according to the time step, aggregates the pressure change gradient data of the corresponding trajectory segment, and screens out the gradient segment that meets the contact characteristic change threshold to obtain the contact pressure change gradient segment value;

[0019] The current cross-segment extraction submodule calls the contact pressure change gradient segment value, combines the current change data sequence, locates the current change cross segment within the gradient segment, establishes a time correspondence by comparing the synchronization of the current change rate and the pressure gradient change rate, and selects the cross segment with a time difference less than the current synchronization threshold to obtain the synchronous current change cross segment value;

[0020] The trajectory structure update submodule calculates the temperature mean difference adjustment factor and the pressure change gradient according to the cross-section value of the synchronous current change and the temperature change trend in each section for weighted correction using the formula:

[0021] ;

[0022] Obtain overlapping extension structure values ​​through calculation, construct trajectory combination update factors based on multi-parameter harmonic structure, and generate overlapping extension structure values;

[0023] in, Represents the overlapping extension structure value, Representative The pressure gradient value of the segment, Representative The mean temperature change of the segment, Representative Segment current change gradient, Represents the number of segments involved in the revision.

[0024] As a further solution of the present invention, the offset extraction module includes:

[0025] The non-overlapping area identification submodule detects and identifies non-overlapping time intervals of contact pressure and current in the overlapping extension structure based on corresponding time periods of the contact pressure and current, extracts corresponding time indexes and section identifiers, and obtains a non-overlapping time index set;

[0026] The difference sequence calculation submodule extracts the contact pressure value sequence and the current value sequence within the corresponding time period according to the non-overlapping time index set, performs item-by-item subtraction on the two sequences along the time axis to obtain a difference sequence, calculates the change in adjacent differences, and generates a difference change sequence to obtain a difference change sequence set;

[0027] The mutation segment screening submodule calls the difference change sequence set, calculates the mutation amplitude index based on the absolute change amplitude of adjacent changes, and introduces the difference change density and the number of adjacent segments, using the formula:

[0028] ;

[0029] Obtain the mutation intensity index B through calculation, filter the segment indexes where B is greater than the mutation intensity threshold, and obtain the boundary mutation segments;

[0030] in, represents the difference change between adjacent time points, S represents the density of difference change, η represents the number of adjacent segments in the mutation segment, B represents the mutation intensity index, and ∑ represents the value of all adjacent segments in the segment. The absolute values ​​of are summed and multiplied by the square root of S.

[0031] As a further solution of the present invention, the trend characterization module includes:

[0032] The contact state acquisition submodule acquires the contact surface state of the contact based on the closed cycle segment identified in the boundary mutation section, acquires the contact area, pressure change and wear trajectory of the contact surface, integrates the acquired contact area, pressure change and wear trajectory, classifies the contact state parameters, and generates a contact state characteristic value;

[0033] The intersection point offset analysis submodule calls the contact state characteristic value, performs intersection point offset analysis, compares the difference in intersection point position change based on the contact area change value and the wear trajectory change value, calculates the initial spindle offset value of the intersection area, and generates the initial spindle offset coefficient;

[0034] The spindle track offset calculation submodule calls the spindle offset initial value coefficient and calculates the spindle track offset based on its combined effect with the pressure change, using the formula:

[0035] ;

[0036] The spindle track offset change value is obtained by calculation, and the offset change value is mapped back to the original action segment to obtain the spindle track offset result;

[0037] in, Represents the change in spindle track offset, Represents the contact area change value, Represents the change in wear trajectory, represents the initial value of pressure change, represents the final value of the pressure change, represents the length of the closed cycle segment, Represents the difference in the position change of the intersection point.

[0038] As a further solution of the present invention, the exception capture module includes:

[0039] The motion data extraction submodule obtains the motion duration and current fluctuation data in the corresponding closed segment based on the spindle track deviation result, integrates the motion duration sequence and the current fluctuation amplitude, and generates the motion current characteristic value;

[0040] The intersection feature recognition submodule calls the action current feature quantity, identifies the density of the intersection points based on the combined characteristics of the action duration sequence and the current fluctuation amplitude, determines the change in current amplification, analyzes the density characteristics and amplification differences of the intersection points, and generates the intersection feature change quantity;

[0041] The abnormal point marking submodule calls the intersection feature change amount, and based on the joint condition of the intersection feature change amount and the current fluctuation difference amplitude, screens the abnormal points in the intersection section, marks the abnormal points, establishes the contact health status index, and obtains the contact health management result.

[0042] Compared with the prior art, the advantages and positive effects of the present invention are:

[0043] In the present invention, by eliminating data segments that lack synchronization and continuity, a stable trajectory sequence is constructed, the data basis of virtual-real mapping is enhanced, and a multi-parameter structure is introduced into the trajectory surface to achieve coordinated expression of state changes. The pressure gradient is extracted in the contact pressure and current intersection section and the trajectory is corrected in combination with the temperature factor to improve the dynamic adaptability of the trajectory to wear evolution. Difference analysis accurately characterizes the mutation behavior, and the main axis offset mapping strengthens the behavioral path backtracking association. The action duration and current fluctuation characteristics are combined to calibrate the abnormal points. A health assessment system with multi-layer recognition capabilities is constructed to improve the accuracy of contact state recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a system flow chart of the present invention;

[0045] Figure 2This is a flow chart of the working condition injection module of the present invention;

[0046] Figure 3 This is a flow chart of the wear mapping module of the present invention;

[0047] Figure 4 This is a flow chart of the offset extraction module of the present invention;

[0048] Figure 5 It is a flow chart of the trend characterization module of the present invention;

[0049] Figure 6 This is a flow chart of the exception capture module of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0051] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0052] See also Figure 1 , the contact health management system based on digital twin includes:

[0053] The working condition injection module obtains the operating current, contact pressure, and ambient temperature data during the contact opening and closing operation, determines the data synchronization and continuity, connects the data to form a trajectory sequence, inputs it into the digital twin mapping body, establishes the corresponding three groups of trajectory surface, and obtains the associated trajectory combination;

[0054] The wear mapping module extracts the intersection segment with the current change based on the contact pressure trajectory in the associated trajectory combination, calculates the matching degree between the pressure change gradient and the current fluctuation, extracts the pressure gradient value and adjusts the trajectory based on the temperature change, and updates the trajectory structure based on this information to obtain the overlapping extension structure.

[0055] The offset extraction module extracts the difference between contact pressure and current based on the non-overlapping time periods in the overlapping extended structure, calculates the change in the difference in chronological order, analyzes the offset amplitude, and screens the sections with sudden changes to form an extended boundary mutation map and obtain boundary mutation sections.

[0056] The trend characterization module uses the closed cycle segments identified in the boundary mutation section to collect the contact surface status, including contact area, pressure change, and wear trajectory. It then performs intersection offset analysis, calculates the spindle offset in the intersection area, and maps the offset value back to the original motion segment to obtain the spindle track offset result.

[0057] The abnormality capture module calls the action duration and current fluctuation data in the closed segment corresponding to the spindle track offset result, identifies the density and amplification changes of the intersection points, marks the abnormal points in the intersection section, and obtains the contact health management results.

[0058] The associated trajectory combination includes synchronous trajectory segments, continuous trajectory segments, and three sets of mapping surfaces. The overlapping extension structure includes pressure change gradient value, current fluctuation matching degree, and temperature adjustment factor. The boundary mutation segment includes contact pressure difference sequence, current difference sequence, and mutation change amplitude. The main axis track offset results include main axis offset value, intersection point offset characteristics, and contact surface pressure change. The contact health management results include action duration characteristics, current fluctuation index, intersection density, and amplification abnormality points.

[0059] See also Figure 2 , the working condition injection module includes:

[0060] The data screening submodule obtains the working current, contact pressure and ambient temperature data during the contact opening and closing period, determines the synchronization and continuity of the three data in the time dimension, and generates the total amount of synchronous continuous data segments;

[0061] First, the control time of opening and closing is recorded through the contactor control signal, and the start and end times of the sampling time period are set. For example, if the contact closure control signal is issued at t0=2.000s, the starting sampling time is set to t0-0.5s, that is, 1.500s, and the end time is set to t0+1.5s according to the control instruction or action delay time, that is, 3.500s. Three sets of parameters are collected synchronously during this time period. The sampling frequency is fixed at 1000Hz, that is, 1000 sets of data are collected per second to ensure that the three parameters form a data group at each timestamp point. Then, the synchronization of the data in the time dimension is judged by traversing the timestamps of the three sets of data at each time point. If the difference between the timestamp of any set of data and the other two sets of timestamps exceeds 2ms, the data group at that point is eliminated. For example, the current timestamp at a certain time point is 1.502s, the pressure is 1.502s, and the temperature is 1.5045s. The difference between the temperature and the current is 2.5ms, which exceeds the threshold, so the group is eliminated and the next step is continued. The 2ms threshold is set based on the hardware acquisition system synchronization deviation upper limit test. In 99 out of 100 tests, the difference did not exceed 1.8ms, so 2ms was set as the stability boundary. After completing the synchronization test, all remaining data are checked for continuity. Specifically, each data segment is checked for data loss or anomalies with a continuous time interval exceeding 10ms. If a segment between 3.210s and 3.225s has data missing for more than 10ms, it is considered a breakpoint and is discarded or processed in segments. Each valid data segment must have a continuous duration of at least 1.000s and a data integrity rate of at least 95%. For example, a segment starting at 2.100s and ending at 4.400s, with two 0.005s interruptions in between, has a total duration of 2.3s and a data integrity rate of 99.56%. This condition is met and the segment is retained as a valid segment. All remaining valid segments are numbered according to their start and end times, and the total amount of current, pressure, and temperature data is recorded to form the total number of synchronized continuous data segments.

[0062] The trajectory construction submodule calls the total amount of synchronous continuous data segments, connects the data segments according to the time series, combines the working current, contact pressure difference and ambient temperature change trend, screens the connected segments, and generates a trajectory connection set;

[0063] The trajectory construction submodule calls the total amount of synchronous continuous data segments, arranges each segment data in ascending order according to the starting time, and then starts the trajectory splicing process. The connection logic is based on the difference judgment of current, contact pressure and temperature between adjacent segments. Each connection judgment is compared with the parameters of the end data point of the current segment and the starting data point of the next segment, and the difference between the end value and the starting value of the current is extracted. For example, the end value of the current segment is 19.2A, and the starting value of the next segment is 20.1A, with a difference of 0.9A. The contact pressure changes from 84.5N to 85.0N, with a difference of 0.5N, and the temperature changes from 34.0℃ to 34.6℃, with a change of 0.6℃. The three differences are compared with the set thresholds respectively. The connection thresholds are set to current ±1.5A, pressure ±5N, and temperature ±1.0℃. If the differences do not exceed the thresholds, the segment is judged to be a connectable segment. The above thresholds are divided into The current threshold is set based on the historical data of the device. The current threshold is derived from the current fluctuation measurement during the low-speed operation of the device. In 50 opening and closing experiments, the maximum fluctuation range did not exceed 1.4A, and 1.5A was set as the upper limit redundancy value; the pressure change is the allowable error of the contact mechanical structure. In actual tests, the maximum is 4.7N, and it is set to 5N; the temperature changes slowly due to environmental influences. The maximum temperature difference between two data segments is measured to be 0.9℃, and 1.0℃ is set as the threshold; in the judgment process, once any item exceeds the set threshold, the current segment splicing is terminated and a new trajectory segmentation is started; for example, in a certain judgment, the current difference is 2.0A, which exceeds 1.5A, then the connection is skipped and a new segment is recorded. The whole process is repeated for all data segment combinations, and the number range of each group of successfully spliced ​​segments is recorded, marked as a trajectory path, and finally multiple trajectory fragments are generated and summarized as a trajectory connection set.

[0064] The facet establishment submodule maps multiple feature combinations of the trajectories in the trajectory connection set, divides the state sequences into three groups according to the time nodes, inputs them into the digital twin mapping body, establishes three groups of trajectory facets, and generates associated trajectory combinations;

[0065] The facet establishment submodule performs mapping processing based on the combination of multiple features of the trajectory in the trajectory connection set. In each trajectory segment, multiple state nodes are divided according to the time step of 0.5s. The working current change rate, contact pressure change amplitude and temperature change direction of each node are extracted to form a state feature vector. The current rate is obtained by dividing the current difference in the current time period by the time length. For example, at the node t1, the current changes from 16.0A to 17.5A in 0.5s, and the rate is 3.0A / s; the contact pressure changes from 82.0N to 85.0N, which is 3.0N, and the corresponding rate is 6.0N / s; the temperature rises from 33.5℃ to 34.0℃, and the change direction is rising; the state vector [3.0A / s, 6.0N / s, ↑] is formed; after constructing all node vectors, each vector is grouped, first according to The first level of classification is based on whether the current rate is greater than 1.5 A / s. This value is set based on the critical rate analysis of rapid changes in equipment load. Under typical load conditions, the current rise rate fluctuates most frequently between 1.4 and 1.6 A / s, with the median value of 1.5 A / s being taken as the dividing point. The second level of classification is based on the pressure change rate, with a critical value of 3.5 N / s, which is derived from the monitoring results of the mechanical spring buffer's action rate. Fast action rates exceed this value, while slow action rates fall below this value. The third level is classified according to the direction of temperature change into three categories: heating, cooling, and stability. Finally, all state nodes are grouped according to the above three characteristics and marked as trajectory state groups A, B, and C, representing high-change, medium-change, and low-change states, respectively. They are arranged in order as state trajectories and imported into the digital twin mapping body to generate three groups of trajectory facets, which are then output as the final associated trajectory combination.

[0066] See also Figure 3 , the wear mapping module includes:

[0067] The pressure trajectory extraction submodule extracts the pressure change point sequence in each trajectory unit based on the contact pressure trajectory data in the associated trajectory combination, calculates the pressure change difference between adjacent points according to the time step, aggregates the pressure change gradient data of the corresponding trajectory segment, and screens out the gradient segment that meets the contact characteristic change threshold to obtain the contact pressure change gradient segment value;

[0068] The pressure trajectory extraction submodule first extracts contact pressure trajectory data from the trajectory combination. The system segments the raw trajectory data sequence according to a fixed time step. Assuming the time step is set to 0.1 seconds, the entire trajectory contains 1000 time points, which can be divided into 100 trajectory units, each containing 10 pressure sampling points. Next, within each trajectory unit, the program iterates through the pressure values ​​of every two adjacent points and calculates the difference. For example, if the pressure value sequence in a trajectory segment is 2.1, 2.3, 2.4, 2.7, 2.9, 3.2, 3.3, 3.5, 3.8, and 4.0, the pressure change difference in this segment is calculated to be 0.2, 0.1, 0.3, 0.2, 0.3, 0.1, 0.2, 0.3, and 0.2, respectively. These differences constitute the pressure change gradient sequence for that segment. Afterwards, the system screens and judges each gradient segment based on the set contact characteristic change judgment threshold. The threshold is a stable reference value obtained through experimental statistics, which is set to 0.25 MPa here. The screening rule is: if the number of pressure changes greater than or equal to 0.25 MPa in a certain segment is not less than 3, then the segment is marked as an area with obvious pressure change characteristics. According to the above sequence, there are three differences of 0.3 MPa in this segment, which meets the screening conditions and is therefore recorded as a valid pressure change gradient segment. The system records and summarizes the number of all trajectory segments that meet the conditions to form a set of valid pressure change gradient segments for use in the next step of cross-analysis.

[0069] The current cross-segment extraction submodule uses the contact pressure change gradient segment value and, combined with the current change data sequence, locates the current change cross-segment within the gradient segment. By comparing the synchronization of the current change rate and the pressure gradient change rate, a time correspondence is established. The cross-segment with a time difference less than the current synchronization threshold is selected to obtain the synchronous current change cross-segment value.

[0070] After receiving the pressure change gradient segment values ​​filtered out in the previous step, the current cross-segment extraction submodule retrieves the current change data within the time periods corresponding to these segments and performs segment alignment processing. Each segment of current data is obtained by subtracting adjacent points to obtain the current change amount, and then the change rate is calculated. For example, the current data in a certain segment is 1.2, 1.4, 1.5, 1.7, 2.0, 2.1, 2.3, 2.6, 2.9, 3.0, the time step is 0.1 seconds, and the adjacent point difference values ​​are 0.2, 0.1, 0.2, 0.3, 0.1, 0.2, 0.3, 0.3, 0.1, respectively. The corresponding current change rate can be expressed as the current increment per step. These current change rate data are compared one by one with the pressure change rate data recorded in the previous paragraph. The comparison method is to determine whether the difference between the two change rates is less than the synchronization judgment threshold item by item, and the change direction must remain consistent. For example, if the pressure change rates are 0.2, 0.1, 0.3, 0.2, 0.3, 0.1, 0.2, 0.3, and 0.2, and compared to the current change rates, the change directions for steps 1, 2, 4, 6, 7, and 8 are consistent, and the difference in change is less than 0.1 amperes per step, meeting the set criteria. To further determine temporal synchronization, the time differences between points where pressure and current change synchronously are compared. A time difference threshold is set at 0.15 seconds. If the time differences between these pairs of change points in the segment are all around 0.1 seconds, the condition is met. Ultimately, this trajectory segment is identified as a current change intersection segment and recorded as a valid synchronization segment for subsequent trajectory structure correction and extension calculations. This process extracts the time alignment properties of multiple physical quantities, improving the targetedness and accuracy of trajectory segment selection based on temporal consistency.

[0071] The trajectory structure update submodule calculates the temperature mean difference adjustment factor and the pressure change gradient for weighted correction based on the cross-segment value of the synchronous current change and the temperature change trend within each segment. The formula is:

[0072] ;

[0073] Obtain overlapping extension structure values ​​through calculation, construct trajectory combination update factors based on multi-parameter harmonic structure, and generate overlapping extension structure values;

[0074] in, Represents the overlapping extension structure value, Representative The pressure gradient value of the segment, Representative The mean temperature change of the segment, Representative Segment current change gradient, Representatives participate in the revision of the segment number;

[0075] This parameter is used to quantify the comprehensive effect of multi-physical field coupling during the dynamic closing process of mechanical contacts;

[0076] The trajectory structure update submodule is based on the synchronous current change intersection segment, combined with the temperature change trend of each segment, calculates the temperature difference adjustment factor and combines the pressure change gradient and current change gradient for weighted correction, and finally obtains the overlapping extension structure value. The formula is:

[0077] ;

[0078] in:

[0079] : overlap extension structure value, indicating the mean value of the structure after correction of the trajectory segment;

[0080] : No. The pressure gradient of the segment (unit: kPa / step) is calculated based on the pressure values ​​of the actual trajectory sampling points;

[0081] : No. The average temperature of the segment (unit: °C) is obtained by continuous sampling by the temperature sensor and averaging the segments;

[0082] : No. The current change gradient of the segment (unit: A / step) is obtained by calculating the change rate of the current sampling sequence;

[0083] : The total number of segments involved in the correction.

[0084] Operation logic description:

[0085] The pressure gradient of each segment is calculated based on the difference sequence of data within the segment. For example, the pressure data in segment 1 is , the step size is 0.2 seconds, then the gradient is Or converted to , converted to kPa / step , the temperature data is , the mean is , the current data is , then the current gradient is correspond , and the calculation is as follows:

[0086] ;

[0087] ;

[0088] By analogy, there are 3 segments of data, and the other two segments are:

[0089] Paragraph 2: ;

[0090] Paragraph 3: ;

[0091] Substitute respectively:

[0092] ;

[0093] ;

[0094] After summing, substitute into the formula:

[0095] ;

[0096] The results show that the average pressure gradient, after synchronous temperature and current corrections, is 1397.66 kPa / step in the three current trajectories. This is used to construct the overlapping extension structure factor of the trajectory combination, which serves as a key parameter for subsequent structural map updates or wear progression modeling. The formula is beneficial because it introduces a nonlinear coefficient formed by the square root of the mean temperature and the inverse of the current gradient, forming a dynamic weighted correction mechanism, thereby enhancing the ability to identify the physical change characteristics of each segment.

[0097] See also Figure 4 , the offset extraction module includes:

[0098] The non-overlapping area identification submodule detects and identifies the non-overlapping time intervals of the contact pressure and current in the overlapping extension structure based on the corresponding time periods of the contact pressure and current, extracts the corresponding time index and segment identifier, and obtains the non-overlapping time index set;

[0099] When the non-overlapping area identification submodule performs the identification operation based on the corresponding time period of the contact pressure and current in the overlapping extension structure, it is necessary to first clarify the data acquisition cycle and signal synchronization benchmark, and synchronously compare the contact pressure and current signals at the same sampling frequency (such as once every 0.1 seconds). Each monitoring period is set to 10 seconds, which corresponds to 100 data points. By marking the time point corresponding to each value in the current signal sequence, it is confirmed whether there is a valid numerical response to the corresponding time point in the pressure signal. When there is a current value but the corresponding pressure value is zero or missing, or the pressure is significantly lower than the system setting For the period of the benchmark (for example, less than 1.0kPa), the time point is added to the preliminary judgment list of the non-overlapping area. On this basis, the time window sliding method is introduced to form time segments for all consecutive non-overlapping periods. For example, if the current increases significantly but the pressure remains low at the 20th to 25th sampling points, the segment [20, 25] is generated. The index of the first point in each time period is used as the segment identifier, and then a non-overlapping time index set is formed in multiple segments. For example, the final index set is {[12, 17], [20, 25], [48, 51]}, which can be used for subsequent difference processing operations.

[0100] The difference sequence calculation submodule extracts the contact pressure value sequence and the current value sequence within the corresponding time period according to the non-overlapping time index set, performs item-by-item subtraction on the two sequences along the time axis, obtains the difference sequence, calculates the change in adjacent differences, and generates a difference change sequence to obtain a difference change sequence set;

[0101] After receiving the non-overlapping time index set, the difference sequence calculation submodule needs to extract the pressure value sequence and current value sequence of the corresponding segment from the original data sequence one by one. For example, in segment [12, 17], the pressure value is [1.0, 0.8, 0.7, 1.1, 0.6, 0.5], and the current value is [2.2, 1.9, 1.7, 2.0, 1.8, 1.6]. The difference between the two is calculated item by item, that is, the pressure minus the current is calculated item by item to obtain the difference sequence [-1.2, -1.1, -1.0, -0.9, -1.2, -1. 1], and then further calculate the difference change Δv by the difference between adjacent items, that is, [-0.1, -0.1, -0.1, 0.3, 0.1], completing the generation of the entire difference change sequence. The "calculation" process in this operation is specifically as follows: read each difference, subtract the previous item from the current item to form a new list element, perform 5 operations to obtain 5 changes, and continue to repeat the above operations in other segments of the index set to finally form a difference change sequence set. In this process, each group of numerical sequences is clearly marked with the time index and data value to ensure traceability and feasibility.

[0102] The mutation segment screening submodule calls the difference change sequence set and calculates the mutation amplitude index based on the absolute change amplitude of adjacent changes. It also introduces the difference change density and the number of adjacent segments and uses the formula:

[0103] ;

[0104] Obtain the mutation intensity index B through calculation, filter the segment indexes where B is greater than the mutation intensity threshold, and obtain the boundary mutation segments;

[0105] in, represents the difference change between adjacent time points, S represents the density of difference change, η represents the number of adjacent segments in the mutation segment, B represents the mutation intensity index, and ∑ represents the value of all adjacent segments in the segment. The absolute value of sum is multiplied by the square root of S;

[0106] This indicator is used to identify the severity of sudden changes in the contact surface state. Its innovation lies in a three-dimensional composite calculation: first, the pressure-current difference sequence of the non-overlapping period is extracted, and the difference change between adjacent time points is calculated; second, the number of changes exceeding the threshold per unit time is counted as the density; finally, the number of adjacent sections is introduced to perform range compensation;

[0107] The mutation strength index B is calculated for each candidate segment, where Represents the change in adjacent difference items within any segment, is the sum of the absolute values ​​of all changes in the segment, is the density of difference changes within the segment, that is, the number of change points divided by the length of time, The number of adjacent segments with the same trend direction within the segment is obtained by reading the absolute value of each Δv and accumulating them to obtain the first sum value, and then performing square root product calculation. At the same time, η plus 1 is used as the denominator to complete the entire formula calculation process. For example, if the Δv of a segment is [0.6, -0.8, 0.7, -0.9, 1.0], its ∑|v| is obtained as 0.6+0.8+0.7+0.9+1.0=4.0, and S is set to 5. is 3, and we can get:

[0108] ;

[0109] In the second example, Δv is [1.2, -1.5, 1.3, -1.1, 1.4, -1.6], ∑|v|=8.1, S=8, η=4, and we get:

[0110] ;

[0111] For the third example, Δv is [0.3, -0.2, 0.4, -0.5, 0.1], ∑|v|=1.5, S=3, η=2, and we get:

[0112] ;

[0113] By and threshold Comparison revealed that the second data segment was a mutation segment, while the first and third segments were non-mutation segments. This result indicates that the B value only exceeds the threshold when the fluctuations are strong and dense, further improving the accuracy of mutation judgment and eliminating false positives. The formula is beneficial because it combines the absolute magnitude of the difference change with its density, integrating both the intensity of the change and the magnitude of the amplitude, and, combined with the segment structure density, provides a complete characterization of the mutation intensity.

[0114] See also Figure 5 , the trend characterization module includes:

[0115] The contact state acquisition submodule collects the contact surface state of the contact based on the closed cycle segment identified in the boundary mutation section, collects the contact area, pressure change and wear trajectory of the contact surface, integrates the collected contact area, pressure change and wear trajectory, classifies the contact state parameters, and generates the contact state characteristic value;

[0116] The contact state acquisition submodule starts working based on the closed cycle segment identified in the boundary mutation segment called by the trend characterization module. First, the start and end index values ​​are extracted according to the boundary mutation segment map to confirm whether it belongs to a complete closed cycle. For example, if the contact completes a closing and opening operation between the 10th and 14th seconds, it is considered a complete closed cycle. During this period, the contact area and wear trajectory data are collected by the flexible thin film resistor and wear vision sensor installed on the contact surface of the contactor. The contact area is determined by the two-dimensional pixel projection area. For example, the image recognition area at a certain moment is 84 pixels, and the actual mapped area is 84mm². The pressure is converted according to the change in resistance value, and the wear trajectory is converted by the graph. The contour changes before and after the image comparison form trajectory lines and are converted into coordinate offsets. After the above acquisition is completed, the system arranges the area values, pressure values, and wear trajectory change curves in each cycle in chronological order, and performs normalization and normalization processing. Three groups of vectors, namely area feature vector, pressure change rate sequence, and trajectory offset sequence, are constructed respectively. The maximum change of each vector is used as the characteristic value of the item. For example, if the area decreases from 85mm² to 60mm², the pressure fluctuates from 15kPa to 10kPa, and the maximum amplitude of the wear trajectory offset is 1.4mm, it is classified as an area feature value of 25mm², a pressure change value of 5kPa, and a trajectory offset of 1.4mm, and is finally output as the contact state feature value.

[0117] The intersection offset analysis submodule calls the contact state characteristic value to perform intersection offset analysis. Based on the difference between the change in contact area and the change in wear trajectory, it compares the difference in intersection position change, calculates the initial spindle offset of the intersection area, and generates the initial spindle offset coefficient.

[0118] The intersection point offset analysis submodule calls the above-mentioned contact state characteristic value to start offset analysis. First, the area change value and the wear trajectory offset are extracted from each cycle as input for comparison. The specific operation is to subtract the final value from the initial area value to obtain the area change value α1. At the same time, the difference between the end position and the starting position of the trajectory offset is taken to obtain the wear trajectory change α2. The two are numerically differenced to generate the preliminary intersection point change. Then, the influence of the area change rate on the wear change rate is introduced for numerical proportional analysis to determine the significance of the difference within the threshold range. For example, if α1 is 28mm² and α2 is 21mm, the difference between the two is 7mm, which belongs to the medium difference range (the medium difference range is set to [5, 10] mm). The system records the difference and further extracts the starting and final values ​​of the contact pressure in the cycle, setting β1=12.5kPa and β2=14.2kPa. These two pressure data are brought into the next step of the spindle initial value offset coefficient calculation process. The spindle offset initial value coefficient is stored as a single scalar for easy call in the next step.

[0119] The spindle track offset calculation submodule calls the spindle offset initial value coefficient and performs the spindle track offset calculation based on its combined effect with the pressure change. The formula is:

[0120] ;

[0121] The spindle track offset change value is obtained by calculation, and the offset change value is mapped back to the original action segment to obtain the spindle track offset result;

[0122] in, Represents the change in spindle track offset, Represents the contact area change value, Represents the change in wear trajectory, represents the initial value of pressure change, represents the final value of pressure change, represents the length of the closed cycle segment, represents the magnitude of the difference in the position change of the junction;

[0123] This parameter characterizes the dynamic offset characteristics of the mechanical spindle under the action of thermal-mechanical coupling, and its calculation model includes a dual action mechanism;

[0124] The spindle track offset calculation submodule performs the final offset calculation after receiving the spindle offset initial value coefficient. The entire calculation process is based on the formula:

[0125] ;

[0126] In this formula, Represents the spindle track offset change value, in millimeters (mm). Indicates the contact area change value, its unit is square millimeter (mm²), It represents the change of wear track, and its unit is millimeter (mm). The absolute value of the difference between the two reflects the deformation trend of the contact surface. and are the initial and final values ​​of pressure, respectively, in kilopascals (kPa), and their square sum reflects the pressure gradient effect. is the duration of the closed cycle in seconds (s), is the amplitude of the intersection change, in millimeters (mm). In the formula structure, the numerator is composed of the contact area, trajectory difference, and pressure distribution, and the denominator is the accumulation of time and displacement. It is used to control the scale uniformity and reflect the offset intensity per unit time. If the example value is used: 、 、 、 、 、 , the calculation process is as follows:

[0127] ;

[0128] ;

[0129] ;

[0130] Finally, we get:

[0131] ;

[0132] The results show that the spindle track offset reaches 8.80mm during this period. Combined with the set offset critical value range, such as the standard upper limit of 10mm, the current result is in the critical band of the upper limit of the normal offset range, which can be used to reflect the critical state of the structural contact path. The benefit of this formula is that through the fusion operation of the three factors of contact area, wear trajectory and pressure change, it can construct a quantitative expression of the offset amplitude under dynamic and multi-dimensional coupling conditions, thereby improving the characterization ability of the offset behavior.

[0133] See also Figure 6 , the exception capture module includes:

[0134] The motion data extraction submodule obtains the motion duration and current fluctuation data in the corresponding closed segment based on the spindle track offset results, integrates the motion duration sequence and current fluctuation amplitude, and generates the motion current characteristic value;

[0135] After receiving the spindle track offset result, the motion data extraction submodule first segments the data to identify closed segments with mutation trends. Specifically, the spindle track offset data is divided into units of 0.1 seconds according to the time axis, and the offset difference between multiple consecutive units is gradually slid backward. If the difference between the offset of a unit and the offset of the previous unit is greater than 0.05mm, it is used as the starting point of the mutation segment. The tracking continues until the offset difference recovers to less than 0.02mm, marking the end of the mutation segment. In this way, the start and end time of each mutation can be clearly determined. Then, the sampling time points in the mutation segment are traversed, and the difference between the first and last time points is calculated as the action duration of the segment. For example, if it starts at 10.3 seconds and ends at 10.9 seconds, the action duration is 0.6 seconds. Then, all current data in the period are read, and the maximum and minimum values ​​are counted and the difference is calculated to obtain the current fluctuation amplitude. For example, if the current data is 2.2A, 2.8A, 3.0A, 2.5A, and 2.4A, the maximum value is 3.0A, the minimum value is 2.2A, and the fluctuation amplitude is 0.6 seconds. The amplitude of the action is 0.8A. In order to determine whether it is a valid action, the system sets the minimum action duration to 0.4 seconds. Segments below this value will not be recorded. At the same time, the current fluctuation amplitude must be greater than 1.0A to generate a feature value. The threshold is set by extracting a total of 200 groups of offset segment samples from the operation of 20 spindle equipment. Statistics show that the action time distribution of most valid operations is concentrated between 0.5 seconds and 1.2 seconds, and the current fluctuation is concentrated between 1.2A and 4.8A. Therefore, 0.4 seconds and 1.0A are used as screening thresholds, which is an empirical statistical method. In the example, the fluctuation amplitude is less than 1.0A and is not adopted. If the current data of another segment is 3.2A, 3.8A, 4.1A, 3.9A, and 3.4A, and the fluctuation is 0.9A, it will also be eliminated. However, if the data is 2.0A, 4.2A, 4.5A, 3.8A, and 4.0A, and the fluctuation is 2.5A and the action time is 0.7 seconds, it meets the threshold condition and is recorded as a set of valid action feature quantities. The execution process performs the same processing on all closed segments in sequence, and finally forms a complete set of action duration and current fluctuation amplitude.

[0136] The intersection feature recognition submodule uses the action current feature to identify the density of intersection points based on the combined characteristics of the action duration sequence and the current fluctuation amplitude, judge the change in current amplification, analyze the density characteristics and amplification differences of the intersection points, and generate the intersection feature change;

[0137] After the intersection feature recognition submodule receives the action duration and current fluctuation amplitude set output by the action data extraction submodule, it uses a sliding window method with a length of 3 to perform a combined analysis on three consecutive sets of feature quantities. For example, the three consecutive sets of data are (0.5 seconds, 1.8A), (0.7 seconds, 2.2A), and (0.6 seconds, 2.1A). The system first calculates the difference between the maximum and minimum values ​​of the three sets of action duration, recorded as the duration difference, which is 0.7-0.5=0.2 seconds. Then, the maximum and minimum difference of the three sets of current fluctuation amplitudes is calculated, which is 2.2-1.8=0.4A. The two differences are then averaged as the intersection feature value of the window, and the result is (0.2+0.4) / 2=0.3. This value is used to reflect the density of the action duration and current fluctuation changes in this segment of data. If the intersection feature value is lower than 0.5, it is determined to be a dense section. If it is higher than 0.5, it is a non-dense section. The threshold is also derived from the historical samples. Clustering the dense characteristics of intersection points revealed that, among 500 groups of intersection point samples, more than 80% of the dense point eigenvalues ​​were concentrated between 0.2 and 0.45. Therefore, a threshold of 0.5 was empirically set. After obtaining the intersection eigenvalue, the system proceeded to determine whether there was a significant current increase by sequentially comparing the current amplitudes of the three segments. If the current amplitude of the middle segment increased by more than 1.0A compared with the previous segment, and the current amplitude of the last segment increased by more than 0.8A compared with the middle segment, a current increase trend was determined. This determination criterion was derived from an analysis of typical increase cases during equipment startup. For example, if the current amplitudes of the three segments were 1.5A, 2.8A, and 3.7A, the increase in the middle segment was 1.3A compared with the previous segment, and the increase in the last segment was 0.9A compared with the middle segment, both of which met the determination criteria. The window was marked as "dense and increase"; otherwise, it was a normal intersection segment. Finally, the intersection eigenvalues ​​and increase in each window were recorded as the intersection feature variation for subsequent processing.

[0138] The abnormal point marking submodule calls the intersection feature variation. Based on the combined condition of the intersection feature variation and the current fluctuation difference amplitude, it screens and marks the abnormal points in the intersection section, establishes the contact health status index, and obtains the contact health management results.

[0139] After obtaining the intersection feature change, the abnormal point marking submodule first normalizes the feature change values ​​of all intersection segments, and uses the minimum and maximum value normalization processing method to map each feature value to the interval [0, 1]. After processing, the screening threshold of each normalized result is set to 0.75. Those exceeding this value are candidate abnormal segments. The threshold is analyzed from 200 groups of intersection segment samples. On the eve of the equipment abnormality, the intersection change is mostly concentrated between 0.78 and 0.96. It is set to 0.75 based on the statistical distribution. After the initial screening is completed, the current fluctuation difference of this segment is further combined to determine whether it can be confirmed as an abnormal point. If the difference between the maximum current and the minimum current of this segment is greater than 2.2A, it is finally marked as an abnormal point. The 2.2A threshold is obtained from the operation of 30 devices. The abnormal records during the operation were analyzed and found that more than 95% of the segments with current fluctuations exceeding 2.2A were true abnormal segments. Therefore, they were set as the basis for current amplitude judgment. For example, if the normalized intersection eigenvalue of a segment was 0.83 and the current fluctuation was 2.5A, it was confirmed as an abnormal point if both conditions were met. The system then summarized the distribution of all abnormal points on the time axis and counted the number of abnormal points in each segment. If the number of abnormal points exceeded 3 in a continuous time period, the health status of the device contact corresponding to the segment was determined to be "poor". If only 1 point appeared, it was "good", and 2 to 3 points were "general". This standard was set after 20 rounds of equipment startup processes tested on 10 sets of experimental platforms. It can fully cover the health level distribution of real scenarios and ultimately form a segmented contact health assessment result.

[0140] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. The contact health management system based on digital twin is characterized by: The system comprises: The working condition injection module obtains the current, contact pressure, and temperature data during the contact opening and closing operation, determines the synchronization and continuity, eliminates abnormal segments, forms a trajectory sequence, inputs the digital twin mapping body, constructs three sets of trajectory surface, and obtains the associated trajectory combination; The wear mapping module extracts the intersection segment with the current change based on the contact pressure trajectory in the associated trajectory combination, calculates the matching degree between the pressure gradient and the current fluctuation, and adjusts the trajectory in combination with the temperature change to form an overlapping extension structure; The offset extraction module extracts the non-overlapping period of the contact pressure and current difference in the overlapping extension structure, analyzes the difference change and amplitude according to time, identifies the mutation section, and obtains the boundary mutation section; The trend characterization module calls the closed cycle segment data in the boundary mutation section, analyzes the contact surface state, pressure change and wear trajectory, performs intersection point offset analysis, calculates the spindle offset, maps it to the original action segment, and obtains the spindle track offset result; The abnormality capture module analyzes the action duration and current fluctuation based on the closed segment where the spindle track deviation result is located, identifies the density and amplification changes of the intersection points, marks the abnormal points, and outputs the contact health management results; The associated trajectory combination includes synchronous trajectory segments, continuous trajectory segments, and three sets of mapping facets; the overlapping extension structure includes pressure change gradient value, current fluctuation matching degree, and temperature adjustment factor; the boundary mutation segment includes contact pressure difference sequence, current difference sequence, and mutation change amplitude; the spindle track offset result includes spindle offset value, intersection point offset characteristics, and contact surface pressure change; the contact health management result includes action duration characteristics, current fluctuation index, intersection density, and amplification abnormality point.

2. The contact health management system based on digital twin according to claim 1, characterized in that: The working condition injection module includes: The data screening submodule obtains the working current, contact pressure and ambient temperature data during the contact opening and closing period, determines the synchronization and continuity of the three data in the time dimension, and generates the total amount of synchronous continuous data segments; The trajectory construction submodule calls the total amount of the synchronous continuous data segments, connects the data segments in time series, and screens the connectable segments based on the working current, contact pressure difference and ambient temperature change trend to generate a trajectory connection set; The facet establishment submodule maps the multiple feature combinations of the trajectories in the trajectory connection set, divides the state sequences into three groups according to the time nodes, inputs them into the digital twin mapping body, establishes three groups of trajectory facets, and generates associated trajectory combinations.

3. The contact health management system based on digital twin according to claim 2, characterized in that: The wear mapping module includes: The pressure trajectory extraction submodule extracts the pressure change point sequence in each trajectory unit based on the contact pressure trajectory data in the associated trajectory combination, calculates the pressure change difference between adjacent points according to the time step, aggregates the pressure change gradient data of the corresponding trajectory segment, and screens out the gradient segment that meets the contact characteristic change threshold to obtain the contact pressure change gradient segment value; The current cross-segment extraction submodule calls the contact pressure change gradient segment value, combines the current change data sequence, locates the current change cross segment within the gradient segment, establishes a time correspondence by comparing the synchronization of the current change rate and the pressure gradient change rate, and selects the cross segment with a time difference less than the current synchronization threshold to obtain the synchronous current change cross segment value; The trajectory structure update submodule calculates the temperature mean difference adjustment factor and the pressure change gradient according to the cross-section value of the synchronous current change and the temperature change trend in each section for weighted correction using the formula: ； Obtain overlapping extension structure values ​​through calculation, construct trajectory combination update factors based on multi-parameter harmonic structure, and generate overlapping extension structure values; in, Represents the overlapping extension structure value, Representative The pressure gradient value of the segment, Representative The mean temperature change of the segment, Representative Segment current change gradient, Represents the number of segments involved in the revision.

4. The contact health management system based on digital twin according to claim 3 is characterized in that: The offset extraction module includes: The non-overlapping area identification submodule detects and identifies non-overlapping time intervals of contact pressure and current in the overlapping extension structure based on corresponding time periods of the contact pressure and current, extracts corresponding time indexes and section identifiers, and obtains a non-overlapping time index set; The difference sequence calculation submodule extracts the contact pressure value sequence and the current value sequence within the corresponding time period according to the non-overlapping time index set, performs item-by-item subtraction on the two sequences along the time axis to obtain a difference sequence, calculates the change in adjacent differences, and generates a difference change sequence to obtain a difference change sequence set; The mutation segment screening submodule calls the difference change sequence set, calculates the mutation amplitude index based on the absolute change amplitude of adjacent changes, and introduces the difference change density and the number of adjacent segments, using the formula: ； Obtain the mutation intensity index B through calculation, filter the segment indexes where B is greater than the mutation intensity threshold, and obtain the boundary mutation segments; in, represents the difference change between adjacent time points, S represents the density of difference change, η represents the number of adjacent segments in the mutation segment, B represents the mutation intensity index, and ∑ represents the value of all adjacent segments in the segment. The absolute values ​​of are summed and multiplied by the square root of S.

5. The contact health management system based on digital twin according to claim 4 is characterized in that: The trend characterization module includes: The contact state acquisition submodule acquires the contact surface state of the contact based on the closed cycle segment identified in the boundary mutation section, acquires the contact area, pressure change and wear trajectory of the contact surface, integrates the acquired contact area, pressure change and wear trajectory, classifies the contact state parameters, and generates a contact state characteristic value; The intersection point offset analysis submodule calls the contact state characteristic value, performs intersection point offset analysis, compares the difference in intersection point position change based on the contact area change value and the wear trajectory change value, calculates the initial spindle offset value of the intersection area, and generates the initial spindle offset coefficient; The spindle track offset calculation submodule calls the spindle offset initial value coefficient and calculates the spindle track offset based on its combined effect with the pressure change, using the formula: ； The spindle track offset change value is obtained by calculation, and the offset change value is mapped back to the original action segment to obtain the spindle track offset result; in, Represents the change in spindle track offset, Represents the contact area change value, Represents the change in wear trajectory, represents the initial value of pressure change, represents the final value of the pressure change, represents the length of the closed cycle segment, Represents the difference in the position change of the intersection point.

6. The contact health management system based on digital twin according to claim 5, characterized in that: The exception capture module includes: The motion data extraction submodule obtains the motion duration and current fluctuation data in the corresponding closed segment based on the spindle track deviation result, integrates the motion duration sequence and the current fluctuation amplitude, and generates the motion current characteristic value; The intersection feature recognition submodule calls the action current feature quantity, identifies the density of the intersection points based on the combined characteristics of the action duration sequence and the current fluctuation amplitude, determines the change in current amplification, analyzes the density characteristics and amplification differences of the intersection points, and generates the intersection feature change quantity; The abnormal point marking submodule calls the intersection feature change amount, and based on the joint condition of the intersection feature change amount and the current fluctuation difference amplitude, screens the abnormal points in the intersection section, marks the abnormal points, establishes the contact health status index, and obtains the contact health management result.

Images