Precision shaft health state diagnosis method and system based on data twinning

Abstract

The invention provides a precision shaft health state diagnosis method and system based on data twinning, and relates to the technical field of precision shaft health state monitoring, and the method comprises the steps: outputting a fusion health feature of a target precision shaft from a health state model of data twinning, and determining a linear offset during the linear change of the health state of the target precision shaft; determining a health response label of the target precision shaft in the operation state under the variable working condition according to the balance health index, and determining a cross degradation index of the health state of the target precision shaft according to the health response label; determining a health deviation point position of the target precision shaft at each moment during operation, and determining a time sequence health gradient of the full life cycle of the target precision shaft according to the health deviation point position and the linear offset; and performing fusion diagnosis on the health state of the precision shaft according to the cross degradation index and the time sequence health gradient. According to the method, the health state of the precision shaft can be accurately captured in variable working conditions and a complex degradation process, so that the sensitivity of a health diagnosis model to dynamic performance degradation of the precision shaft is improved.

Description

Technical Field

[0001] The present application relates to the technical field of precision shaft health status monitoring, and more specifically, to a precision shaft health status diagnosis method and system based on data twins. Background Art

[0002] Precision shaft health status monitoring refers to the process of determining the working status and degree of degradation of precision shafts through real-time collection, analysis and evaluation of various performance data during the operation of precision shafts, and then predicting their remaining service life and potential failure risks. This monitoring process usually involves dynamically monitoring the key parameters of precision shafts by installing sensors (such as vibration sensors, temperature sensors, pressure sensors, etc.). By analyzing data such as vibration signals, temperature changes, noise, and load, it can be revealed whether the precision shafts have problems such as wear, overheating, and failure during operation.

[0003] The existing data twin-based precision shaft health status diagnosis method is to build a digital twin model of the physical precision shaft, monitor the operating status of the precision shaft in real time, and use data analysis technology to predict faults and perform health assessments. First, the key operating data of the precision shaft is collected through sensors (such as vibration, temperature, pressure, etc.), and these data are transmitted to the digital twin model. The digital twin model compares the real-time data of the physical shaft with the virtual model to simulate the working environment and operating status of the precision shaft, thereby providing support for health status assessment. By analyzing these data in real time, the model can identify the performance degradation of the precision shaft under different working conditions and reveal potential faults or risks. However, the existing data twin-based precision shaft health status diagnosis methods mainly rely on direct analysis of sensor data and preset assumptions of the model, and often ignore the dynamic changes of the shaft under complex working conditions and the interactive effects of multi-dimensional performance degradation. This makes it difficult for traditional methods to fully capture the health change trends of precision shafts when facing complex working conditions and changing environments, resulting in insufficient accuracy and real-time performance of fault prediction, and thus being unable to effectively deal with the sudden degradation of precision shafts in actual operation. Therefore, how to accurately capture the health status of precision shafts under changing working conditions and complex degradation processes to enhance the sensitivity of health diagnosis models to the dynamic performance degradation of precision shafts is a problem faced by the industry. Summary of the invention

[0004] The present application provides a precision shaft health status diagnosis method and system based on data twins, which can accurately capture the health status of the precision shaft under variable working conditions and complex degradation processes, so as to enhance the sensitivity of the health diagnosis model to the dynamic performance degradation of the precision shaft.

[0005] In a first aspect, the present application provides a precision shaft health status diagnosis method based on data twins, the health status diagnosis method comprising the following steps: Scan and measure the target precision axis to obtain the running status data of the target precision axis; Adding the operating status data to the health status model of the data twin, outputting the fused health characteristics of the target precision axis, performing dynamic trend analysis on the fused health characteristics, and obtaining the linear offset when the health status of the target precision axis changes linearly; Determine a balanced health index reflecting the dynamic performance degradation of the precision shaft, determine a health response label of the target precision shaft under variable working conditions according to the balanced health index, and then determine a cross-degradation index of the health state of the target precision shaft according to the health response label; Constructing a health characteristic curve of the target precision axis over its entire life cycle, marking the health characteristic curve with differences, obtaining the health deviation point of the target precision axis at each moment during operation, and then determining the temporal health gradient of the target precision axis over its entire life cycle from the health deviation point and the linear offset; A fusion diagnosis is performed on the health status of the precision axis according to the cross-degradation index and the timing health gradient.

[0006] In this embodiment, the operating status data is added to the health status model of the data twin, and the fusion health characteristics of the output target precision axis specifically include: Based on the characteristics of data twins, a model of the health status of the precision shaft is constructed to obtain the health status model of the data twins; The operating status data is input into the analysis layer of the health status model of the data twin to obtain the fused health characteristics of the target precision axis.

[0007] In this embodiment, the dynamic trend analysis of the fused health feature is performed to obtain the linear offset when the health state of the target precision axis changes linearly, which specifically includes: Determine the local discrete characteristics of the target precision axis health state when it changes linearly according to the fused health characteristics; Determine the steady-state trend information of the target precision axis operation status; A linear offset when a health state of a target precision axis changes linearly is determined according to the local discrete characteristics and the steady-state trend information.

[0008] In this embodiment, determining the balance health index reflecting the dynamic performance degradation of the precision shaft specifically includes: Obtain dynamic performance characteristics of precision shafts; determining a balance adjustment amount reflecting dynamic performance degradation of the precision shaft according to the dynamic performance characteristics; A balance health index reflecting dynamic performance degradation of the precision shaft is determined by the dynamic performance characteristics and the balance adjustment amount.

[0009] In this embodiment, determining the health response label of the target precision axis's operating state under variable working conditions according to the balance health index specifically includes: Determine the state attenuation information of the target precision axis's operating state under variable working conditions; Determining the distribution adaptation characteristics of the precision shaft when the running state changes according to the state attenuation information; Determine the coupling identification amount of the health state of the target precision shaft through the distributed adaptation feature; A health response label of the operating state of the target precision shaft under variable working conditions is determined according to the balance health index and the coupling identification amount.

[0010] In this embodiment, the cross-degradation index of the target precision axis health state determined by the health response tag specifically includes: Determine a health status category of a target precision axis health status according to the health response tag; Determine the performance constraint level when the target precision axis performance changes through the health status category; A cross-degradation indicator of a target precision axis health state is determined according to the performance constraint hierarchy.

[0011] In this embodiment, determining the temporal health gradient of the target precision axis throughout its life cycle based on the health deviation point and the linear offset specifically includes: Determining the state affiliation characteristics of the target precision axis during operation according to the health deviation point; Determining the timing operation properties of the target precision axis throughout its life cycle according to the linear offset; The timing health gradient of the target precision axis throughout its life cycle is determined through the state affiliation characteristics and the timing operation attributes.

[0012] In this embodiment, the health characteristic curve refers to a curve reflecting the health change trend of the target precision axis obtained by analyzing the health status data of the target precision axis throughout its entire life cycle.

[0013] In this embodiment, the operating status data refers to multi-dimensional dynamic feature information collected by high-precision sensors and scanning measurement equipment during the actual operation of the precision shaft.

[0014] In a second aspect, the present application provides a precision shaft health status diagnosis system based on data twins, which is used to perform a precision shaft health status diagnosis method based on data twins, and the health status diagnosis system includes: A scanning measurement module is used to scan and measure the target precision axis to obtain the running status data of the target precision axis; A dynamic trend analysis module is used to add the operating status data to the health status model of the data twin, output the fused health characteristics of the target precision axis, perform dynamic trend analysis on the fused health characteristics, and obtain the linear offset when the health status of the target precision axis changes linearly; A health degradation monitoring module, used to determine a balanced health index reflecting the dynamic performance degradation of the precision shaft, determine a health response label of the target precision shaft under variable working conditions according to the balanced health index, and further determine a cross-degradation index of the health state of the target precision shaft from the health response label; A health difference marking module is used to construct a health characteristic curve of the target precision axis throughout its life cycle, to mark the health characteristic curve with differences, to obtain the health deviation point of the target precision axis at each moment during operation, and then to determine the temporal health gradient of the target precision axis throughout its life cycle from the health deviation point and the linear offset; A fusion diagnosis module is used to perform fusion diagnosis on the health status of the precision axis according to the cross-degradation index and the timing health gradient.

[0015] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects: Scan and measure the target precision axis to obtain the operating status data of the target precision axis; add the operating status data to the health status model of the data twin, output the fused health characteristics of the target precision axis, perform dynamic trend analysis on the fused health characteristics, and obtain the linear offset when the health status of the target precision axis changes linearly; determine a balanced health index reflecting the dynamic performance degradation of the precision axis, determine the health response label of the operating status of the target precision axis under variable working conditions based on the balanced health index, and then determine the cross-degradation index of the health status of the target precision axis from the health response label; construct a health characteristic curve of the target precision axis over its entire life cycle, perform difference marking on the health characteristic curve, obtain the health deviation point of the target precision axis at each moment during operation, and then determine the time series health gradient of the target precision axis over its entire life cycle from the health deviation point and the linear offset; perform a fusion diagnosis on the health status of the precision axis based on the cross-degradation index and the time series health gradient.

[0016] It can be seen that in this application, the lack of attribute interaction in the health status diagnosis of precision shafts under the dynamic evolution of variable working conditions can be effectively compensated; among them, through the scanning measurement of the precision shaft, the real-time operation status data of the shaft can be accurately obtained, overcoming the limitations of the traditional method that relies on the preset model, ensuring the real-time and accuracy of the data, and providing a reliable data basis for subsequent analysis, thereby enhancing the accuracy and real-time response capability of health status diagnosis; by fusing the real-time operation status data with the data twin model, the model parameters can be dynamically adjusted to reflect the real-time operation characteristics of the precision shaft. This fusion method helps to more comprehensively evaluate the health status of the precision shaft and improve the adaptability and accuracy of the diagnostic model to complex working conditions; by obtaining the linear offset, the changing pattern of the health status can be accurately captured, providing a scientific basis for predicting the degradation trend and fault occurrence of the shaft, and further improving the accuracy of fault warning; combined with the health response labels under different working conditions, the health status of the shaft under various working conditions can be more accurately described. This provides a comprehensive evaluation standard for health diagnosis in multiple working conditions, which helps to detect potential problems in advance and take preventive measures; by integrating cross-degradation indicators and time-series health gradients, the health status of precision shafts can be evaluated from multiple angles and throughout the life cycle to identify potential failure modes. This multi-dimensional diagnostic method improves the monitoring accuracy of the health status of precision shafts.

[0017] In summary, the technical solution adopted in this application can accurately capture the health status of the precision shaft under variable working conditions and complex degradation processes, so as to enhance the sensitivity of the health diagnosis model to the dynamic performance degradation of the precision shaft. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0019] Figure 1 is a flow chart of a precision shaft health status diagnosis method based on data twins provided in the present application; Figure 2 is a schematic diagram of a process for determining a health response label according to the present application; Figure 3 It is a schematic diagram of the process of determining the health deviation point provided by this application; Figure 4 This is a module structure diagram of the precision shaft health status diagnosis system based on data twin provided in this application. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0021] The embodiment of the present application provides a precision axis health status diagnosis method and system based on data twin, the core of which is to scan and measure the target precision axis to obtain the operating status data of the target precision axis; add the operating status data to the health status model of the data twin, output the fused health characteristics of the target precision axis, perform dynamic trend analysis on the fused health characteristics, and obtain the linear offset when the health status of the target precision axis changes linearly; determine a balanced health index reflecting the dynamic performance degradation of the precision axis, determine the health response label of the operating status of the target precision axis under variable working conditions according to the balanced health index, and then determine the cross-degradation index of the health status of the target precision axis from the health response label; construct a health characteristic curve of the target precision axis for the entire life cycle, perform difference marking on the health characteristic curve, obtain the health deviation point of the target precision axis at each moment during operation, and then determine the time series health gradient of the target precision axis for the entire life cycle from the health deviation point and the linear offset; perform a fusion diagnosis of the health status of the precision axis according to the cross-degradation index and the time series health gradient. The above scheme can accurately capture the health status of the precision shaft under variable working conditions and complex degradation processes, so as to improve the sensitivity of the health diagnosis model to the dynamic performance degradation of the precision shaft.

[0022] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 As shown, this figure is an exemplary flow chart of the precision shaft health status diagnosis method based on data twins according to this embodiment of the present application. The health status diagnosis method includes the following steps: In step S1, a target precision axis is scanned and measured to obtain operating status data of the target precision axis.

[0023] In the specific implementation, high-precision measuring equipment and sensors should be selected first to fully obtain its operating status data. Among them, in terms of morphological measurement, a three-dimensional laser scanner or a three-coordinate measuring instrument is used to scan the geometric features of the shaft shape (such as diameter, roundness, and cylindricity) point by point through a non-contact laser beam to form high-density point cloud data, and the complete three-dimensional morphological model of the precision shaft is obtained through surface reconstruction technology. In terms of dynamic state measurement, vibration sensors (such as accelerometers), temperature sensors, and stress and strain sensors are arranged at key positions of the shaft to monitor its vibration signals, temperature distribution, and force changes in real time. The analog signal output by the sensor is digitally converted through the data acquisition system, and the timing synchronization technology is used to ensure the time consistency of various types of data. During operation, data is continuously collected according to different working conditions (such as different speeds and loads) to record the state evolution process of the precision shaft during work. Finally, all data are stored as a structured time series data set through data fusion and preprocessing technology (such as denoising and filtering), that is, the operating status data of the target precision shaft is obtained, which will not be repeated here.

[0024] It should be noted that in this application, the operating status data refers to the multi-dimensional dynamic characteristic information collected by high-precision sensors and scanning measurement equipment during the actual operation of the precision shaft. These data include parameters such as vibration amplitude, temperature distribution, stress strain, speed, displacement, and geometric characteristic changes of the shaft surface (such as roundness and diameter).

[0025] In step S2, the operating status data is added to the health status model of the data twin, the fused health characteristics of the target precision axis are output, and a dynamic trend analysis is performed on the fused health characteristics to obtain the linear offset when the health status of the target precision axis changes linearly.

[0026] In this embodiment, the operating status data is added to the health status model of the data twin, and the fusion health characteristics of the target precision axis are outputted, which can be achieved by the following steps: Based on the characteristics of data twins, a model of the health status of the precision shaft is constructed to obtain the health status model of the data twins; The operating status data is input into the analysis layer of the health status model of the data twin to obtain the fused health characteristics of the target precision axis.

[0027] In the specific implementation, first, according to the design parameters (such as material properties, geometric dimensions, working load) and physical properties of the precision shaft, the physical mapping model of the precision shaft is established using finite element analysis (FEA) or multi-body dynamics simulation software (such as ANSYS, ABAQUS) to simulate the operation status of the shaft, and the physical model is linked with the actual measurement data. Through sensor data mapping and model parameter update, the virtual model can reflect the operation status changes of the precision shaft in real time, forming a data twin health status model. Then, the collected operation status data (such as vibration, temperature, stress) is preprocessed, including denoising (such as wavelet transform), signal filtering and time series alignment to ensure data accuracy and timeliness, and the preprocessed operation status data is input into the analysis layer of the data twin health status model, and the multi-source data is integrated and analyzed using data fusion technology (such as weighted average method, principal component analysis PCA), and the fusion health feature vector is extracted. Then, the difference between the actual data and the simulation data is compared through the model to extract the fusion health feature of the precision shaft. Finally, the fused health features are output in the form of health feature vectors, which contain multi-dimensional data (such as temperature, vibration, and stress values), forming a comprehensive description of the health status of the precision shaft.

[0028] It should be noted that in this application, the health status model of the data twin refers to a virtual model built based on physical simulation and operating status data, which can map the actual health status of the equipment in real time and evaluate and predict it; the fused health characteristics are characteristic indicators that reflect the current health status of the precision axis.

[0029] In this embodiment, the dynamic trend analysis of the fused health feature is performed to obtain the linear offset when the health state of the target precision axis changes linearly, which can be achieved by the following steps: Determine the local discrete characteristics of the target precision axis health state when it changes linearly according to the fused health characteristics; Determine the steady-state trend information of the target precision axis operation status; A linear offset when a health state of a target precision axis changes linearly is determined according to the local discrete characteristics and the steady-state trend information.

[0030] In the specific implementation, first, the fused health features (such as vibration, temperature, and stress) are segmented in chronological order, and the local data points in each time window are extracted to form local discrete features. For example, if the time window is 10 seconds, the health feature data points within every 10 seconds form a local subset; then the statistics of the health features in each local time window, such as the mean, standard deviation, and skewness, are calculated to quantify the degree of change of the local health features, and by analyzing the changes of the local features over time, it is possible to identify whether the health status shows a linear change trend. For example, when the local mean gradually decreases and the standard deviation remains stable, the linear characteristics of the state change can be preliminarily determined. Then, a global time series analysis is performed on the fused health features. The sliding average method or the least squares method can be used to fit the health status change trend to obtain steady-state trend information, which will not be repeated here. Finally, by jointly analyzing the local discrete features and the steady-state trends, the offset of the fused health features relative to the ideal health trends is calculated. The linear offset can be determined in the following way: linear offset = actual value of the local discrete features - (steady-state trend prediction value) / total number of data points, where the steady-state trend prediction value can be determined by the least squares method based on the rate of change of the health features over time, the deviation of the initial health state, and the health feature data, which will not be repeated here.

[0031] It should be noted that in this application, local discrete features refer to local statistical indicators of health feature data extracted by statistical methods within a smaller time window; steady-state trend information represents the trend characteristics of long-term changes in the health state of the precision axis, including the rate of change and the initial state deviation; linear bias describes the deviation of the actual health feature relative to the linear prediction value of the steady-state trend, and is used to quantify the degree of deviation between the linear change of the health state and the ideal state.

[0032] In step S3, a balance health index reflecting the dynamic performance degradation of the precision shaft is determined, and a health response label of the target precision shaft operating state under variable working conditions is determined based on the balance health index, and then a cross-degradation index of the health state of the target precision shaft is determined by the health response label.

[0033] In this embodiment, determining the balance health index reflecting the dynamic performance degradation of the precision shaft can be achieved by using the following steps: Obtain dynamic performance characteristics of precision shafts; determining a balance adjustment amount reflecting dynamic performance degradation of the precision shaft according to the dynamic performance characteristics; A balance health index reflecting dynamic performance degradation of the precision shaft is determined by the dynamic performance characteristics and the balance adjustment amount.

[0034] In specific implementation, first, high-precision sensors (such as acceleration sensors, displacement sensors, and temperature sensors) can be arranged at key parts of precision shafts to collect dynamic parameters closely related to the shaft operation status. These mainly include: vibration amplitude, frequency components, and harmonic distribution during shaft operation; radial runout of the shaft or the offset of the axis trajectory; heat distribution and temperature rise of the shaft under different load conditions. The dynamic performance data is preprocessed (filtering and denoising, signal detrending, etc.) to extract core dynamic features, such as the root mean square value (RMS) of the vibration signal, frequency domain features (FFT transformation results), and deviation values ​​of the displacement curve. Then, after comparing the dynamic performance characteristics with the design benchmark value or initial state, the deviation value of the dynamic performance degradation degree is quantified. The performance benchmark parameters can be determined based on the design standards of the precision shaft (such as vibration threshold, displacement tolerance, and temperature upper limit). The dynamic characteristics collected during the initial operation of the equipment are used as the health reference benchmark. The balance adjustment amount can be obtained by calculating the difference between the actual dynamic performance characteristics and the benchmark value, that is, the balance adjustment amount is equal to the absolute value of the difference between the dynamic performance characteristic value at the current moment and the initial health state value. This will not be repeated here. Finally, it is necessary to determine multiple dynamic performance characteristics, such as vibration, displacement, and temperature, and evaluate the weight of their impact on the health state based on these characteristics. These weights are usually assigned according to the importance of the characteristics. For example, the vibration characteristic may be more critical, with a weight of 0.5; the displacement characteristic is second, with a weight of 0.3; and the temperature characteristic is set to 0.2. The balance adjustment for each dynamic performance characteristic is calculated, that is, the deviation between the actual measured characteristic value and the design reference value (or initial health state). For example, if the reference value of vibration is 0.2 and the actual measured value is 0.3, the deviation is 0.1. By combining these adjustments with the corresponding weights, the health level of each dynamic characteristic can be evaluated. The health level of each dynamic performance characteristic is weighted and summarized to finally obtain a numerical balance health index. In order to ensure that the balance health index is easy to compare, it is normalized to the range of 0 to 1. The closer the value is to 1, the healthier the dynamic performance of the precision shaft is; the lower the value, the more serious the degradation of the shaft.

[0035] It should be noted that in this application, dynamic performance characteristics refer to the core parameters that reflect the dynamic behavior of the precision shaft during operation, including vibration amplitude, displacement deviation and temperature distribution, etc., which are used to describe the dynamic performance state of the shaft; the balance adjustment amount is an indicator that describes the degree of dynamic performance degradation; the balance health index refers to the degree of deviation of the current health state of the precision shaft from the ideal reference state.

[0036] Preferably, in this embodiment, the health response label of the target precision axis operating state under variable working conditions is determined according to the balance health index, referring to Figure 2The figure is a schematic diagram of the process of determining a health response tag in some embodiments of the present application. In this embodiment, determining a health response tag can be implemented by the following steps: In step S31, the state attenuation information of the operating state of the target precision shaft under the variable working condition is determined; In step S32, the distribution adaptation characteristics of the precision shaft when the running state changes are determined according to the state attenuation information; In step S33, a coupling identification value of a target precision shaft health state is determined by using the distribution adaptation feature; In step S34, a health response label of the operating state of the target precision shaft under variable working conditions is determined according to the balance health index and the coupling identification value.

[0037] In the specific implementation, first, the dynamic performance characteristics of the precision shaft under different working conditions, such as vibration, temperature, speed, etc., are continuously monitored by sensors, and the attenuation mode and trend are extracted according to the data changes of the precision shaft in different operating stages. Statistical analysis, machine learning models and other methods are used to analyze the change rate and degradation trajectory of health characteristics under various working conditions. For example, under higher loads, the frequency or amplitude of the vibration signal may increase, and the temperature may rise, indicating that the operating state of the precision shaft has degraded. Then, by performing statistical analysis on the operating data under different working conditions, the distribution characteristics of the data (such as mean, standard deviation, skewness, etc.) are extracted, and regression models, cluster analysis or adaptation algorithms can be used to fit the state attenuation information to generate distribution adaptation features that reflect the state changes of the precision shaft. For example, Gaussian distribution is used to fit the changes in vibration amplitude under different working conditions, and the health change characteristics of the shaft are judged based on the fitting results. Next, the state attenuation information and distribution adaptation characteristics under different working conditions are analyzed to quantify the degree of change in the health state of the precision shaft. A coupling model (such as least squares-based regression analysis, support vector machine or neural network) can be used to analyze the relationship between the distribution characteristics and the health state under different working conditions to obtain the coupling identification quantity. For example, under different loads, the change amplitudes of vibration characteristics and displacement characteristics may have specific correlation patterns, which can be used to further identify the change in the health state, that is, to obtain the coupling identification quantity of the health state of the target precision shaft. Finally, the change in the health state of the precision shaft is determined based on the comprehensive evaluation of the balance health index and the coupling identification quantity. If the balance health index is close to 1 and the coupling identification quantity shows no obvious change, the label is "healthy"; if the health index is low and the coupling identification quantity shows a large change, the label is "severely degraded". For example, when the balance health index is 0.9 and the coupling identification quantity shows a slight change, it can be judged that the precision shaft is in a "healthy" state; when the balance health index drops to 0.5 and the coupling identification quantity changes greatly, the health response label is "severely degraded".

[0038] It should be noted that in this application, state attenuation information refers to the changing trajectory of the dynamic performance characteristics of the precision shaft under different working conditions, revealing the trend of equipment performance degradation; distributed adaptation characteristics refer to the statistical characteristics based on operating data under different working conditions, which are used to describe the changing trend of the operating state; coupling identification quantity refers to quantifying the degree of change in the health state by analyzing the coupling relationship between dynamic performance characteristics and health state changes; the health response label is an indicator that describes the degree of change when the health state of the precision shaft changes.

[0039] In this embodiment, the cross-degradation index of the target precision axis health state is determined by the health response tag by the following steps: Determine a health status category of a target precision axis health status according to the health response tag; Determine the performance constraint level when the target precision axis performance changes through the health status category; A cross-degradation indicator of a target precision axis health state is determined according to the performance constraint hierarchy.

[0040] In the specific implementation, first, based on the health response labels of the precision axis, such as "healthy", "slightly degraded", "moderately degraded", "severely degraded", etc., the health status is divided into different categories, where each health response label represents a different health status of the precision axis. Based on these labels, a health status category system is created. For example, the label "healthy" corresponds to category 1, indicating no degradation; the label "slightly degraded" corresponds to category 2, indicating a slight performance degradation; the label "moderately degraded" corresponds to category 3, indicating a significant performance degradation; the label "severely degraded" corresponds to category 4, indicating severe performance degradation. Then, based on the health status categories of the precision axis, the performance constraints under different health states are determined. Each health status category corresponds to a performance constraint level. A higher level health status (such as the "healthy" category) indicates that there is no obvious performance restriction, while a lower level health status (such as the "severely degraded" category) indicates that the performance of the precision axis is highly restricted. Then, a constraint level is assigned to each health status category. For example, category 1 (healthy) may have no performance constraints (level 0), category 2 (slightly degraded) may have small performance limitations (level 1), category 3 (moderately degraded) may show obvious performance constraints (level 2), and category 4 (severely degraded) represents severe performance limitations (level 3). Finally, the cross-degradation index is calculated based on the performance constraint level under different working conditions and the changes in the health status category. If the health status category is low (such as "severely degraded") and the performance constraint level is high, the cross-degradation index is large, indicating that the health status of the precision axis has deteriorated. For example, the weighted average method or other data fusion methods can be used to combine the performance constraint level and the health status category to calculate the cross-degradation index.

[0041] It should be noted that in this application, the health status category represents the category of health status classification, which is used to describe the health level of the precision axis; the performance constraint level represents the performance limitation degree of the precision axis under different health states; the cross-degradation index represents the performance change and degradation degree of the precision axis under different health states and affected by different working conditions.

[0042] In step S4, a health characteristic curve of the target precision axis over its entire life cycle is constructed, and the health characteristic curve is differentially marked to obtain the health deviation point of the target precision axis at each moment during operation, and then the temporal health gradient of the target precision axis over its entire life cycle is determined by the health deviation point and the linear offset.

[0043] In specific implementation, the construction of the health characteristic curve of the target precision shaft throughout its life cycle can be achieved in the following ways: first, by regularly collecting operating status data at each life cycle stage of the precision shaft, covering dynamic performance characteristics such as vibration, temperature, and displacement under different working conditions. These data reflect the real-time health status of the precision shaft and provide a basis for subsequent analysis. Next, feature extraction is performed on the collected raw data to extract key parameters closely related to the health status, such as vibration amplitude, temperature change amplitude, load change, etc. These parameters can effectively reflect the degradation trend of the precision shaft. After the feature extraction is completed, the extracted feature data is input into the health status model of the precision shaft using data twin technology. By dynamically modeling historical data and combining the existing health diagnosis model, a health characteristic curve reflecting the health changes of the precision shaft at different life cycle stages is constructed. This curve integrates the health change trend of the precision shaft over time from installation to retirement, which will not be repeated here.

[0044] It should be noted that, in the present application, the health characteristic curve refers to a curve reflecting the health change trend of a target precision axis obtained by analyzing the health status data of the target precision axis throughout its life cycle.

[0045] Preferably, in this embodiment, the health characteristic curve is differentially marked to obtain the health deviation point of the target precision axis at each moment during operation, and the reference Figure 3 The figure is a schematic diagram of the process of determining the health deviation point in some embodiments of the present application. In this embodiment, the health deviation point can be determined by the following steps: In step S41, a health feature set of the target precision shaft over its entire life cycle is extracted from the health feature curve; In step S42, determining the state decision information of the target precision shaft throughout its life cycle according to the health feature set; In step S43, the operation difference node of the target precision axis at each moment during operation is determined; In step S44, the operation difference node is mapped to the state decision information to obtain the healthy deviation point position of the target precision axis at each moment during operation.

[0046] In the specific implementation, first, the key characteristic parameters affecting the health status, such as dynamic performance indicators such as vibration, temperature, and load, are extracted from the health characteristic curve of the target precision axis. These characteristic data reflect the health changes of the precision axis at different times and working conditions. Then, a health feature set is established through historical data and real-time monitoring data. These features will cover the entire life cycle of the precision axis and reflect the health changes at different use stages. Then, based on the extracted health feature set, a data analysis method (such as statistical analysis, trend analysis, machine learning, etc.) is used to build a health status model of the precision axis at different life cycle stages, determine the health status of each stage, and output the state decision information of the target precision axis throughout the life cycle from the health status model. The state decision information includes the health status category, performance change trend, and degradation level of the precision axis at each stage. Then, by comparing the health data collected in real time during the operation of the precision axis with the health feature set of the entire life cycle, the moments with significant deviations from the normal state during the operation are identified. These moments are the operation difference nodes. Among them, algorithms (such as dynamic time warping (DTW), anomaly detection, etc.) can be used to compare the current health status with the historical health status to determine the time point when the deviation occurs. Finally, each operation difference node is mapped with the corresponding health status decision information to obtain the health deviation point of the precision axis at that moment, which will not be repeated here.

[0047] It should be noted that in this application, the health feature set represents a set of key health parameters extracted from the health feature curve of the precision axis throughout its life cycle, which is used to describe the health status of the precision axis at each stage of its life cycle; the state decision information represents the health status category of the precision axis at each stage of its life cycle and its performance change trend information; the operation difference node represents the moment when the health status deviates significantly from the normal health status during the operation of the precision axis; the health deviation point represents the deviation of the health status of the precision axis at a certain moment relative to the normal state, which is used to calibrate the degree of health deviation of the equipment.

[0048] In this embodiment, determining the temporal health gradient of the target precision axis over its entire life cycle by using the health deviation point and the linear offset can be achieved by using the following steps: Determining the state affiliation characteristics of the target precision axis during operation according to the health deviation point; Determining the timing operation properties of the target precision axis throughout its life cycle according to the linear offset; The timing health gradient of the target precision axis throughout its life cycle is determined through the state affiliation characteristics and the timing operation attributes.

[0049] In the specific implementation, first, by analyzing the health deviation points, the health status of the precision axis can be divided into different categories, such as "healthy", "slightly degraded", "moderately degraded" and "severely degraded". Under a specific health deviation point, the membership degree of the health status of the precision axis in each category is used as the state membership feature. For example, in the "slightly degraded" state, the health deviation point of the precision axis may indicate that the health status deviates from the normal value by 10%. At this time, the membership value of the state membership feature can be given to reflect its weight in different health status categories. Then, by calculating the trend of the health status change over time, the timing operation properties of the precision axis in the whole life cycle can be determined, including the accelerated degradation or steady-state change of the health status over time. By analyzing the linear offset, the operation properties of the precision axis can be divided into different stages, such as "steady-state operation", "slow degradation" or "accelerated degradation". These stages describe the health change pattern of the precision axis at different life cycle stages. Finally, by comprehensively considering the state membership characteristics of the health deviation points and the timing operation properties of the linear offset, the health status change gradient of the precision axis, that is, the timing health gradient, is obtained. The time series health gradient can describe the health change rate of the precision axis at different time points and can determine whether the precision axis has entered the accelerated degradation stage or the stable stage.

[0050] It should be noted that in this application, the state affiliation feature refers to the degree of affiliation of the current health state of the precision axis in different health categories (such as healthy, slightly degraded, etc.), which is used to quantify its health state; the timing operation attribute describes the trend of the health state of the precision axis changing over time during its entire life cycle, such as steady-state operation, slow degradation or accelerated degradation; the timing health gradient represents an indicator of the rate of change of the health state of the precision axis, which is used to reflect the trend and rate of change of the health state over time.

[0051] In step S5, a fusion diagnosis is performed on the health status of the precision axis according to the cross-degradation index and the sequential health gradient.

[0052] In specific implementation, firstly, through the cross-degradation index, the system can identify the degradation of the precision axis under different operating conditions, especially how the overall performance of the precision axis is affected when the working conditions change. The cross-degradation index can determine the health changes of the precision axis under different working conditions, especially the performance degradation trend under complex working conditions. Secondly, the temporal health gradient focuses on the time evolution of the health state of the precision axis and describes the rate of change during its degradation process. The temporal health gradient provides a dynamic evaluation standard for the health state of the precision axis by quantifying the rate at which the health state of the precision axis changes over time. By analyzing the temporal health gradient, the changing trend of the health state of the precision axis can be determined, and whether it is in a state of accelerated degradation. For example, if the health gradient continues to increase, it may mean that the degradation rate of the equipment is accelerating and needs special attention. Among them, the fusion diagnosis comprehensively evaluates the health state of the precision axis by combining the cross-degradation index and the temporal health gradient. The cross-degradation index provides multi-dimensional health information, revealing the potential degradation mode of the precision axis under different working conditions, while the temporal health gradient supplements the dynamic change of the degradation rate of the precision axis. By integrating the information of these two, the diagnostic system can accurately determine the health trend of the precision shaft in the current and future operating states, thereby providing more scientific decision support for equipment maintenance and fault prediction. Ultimately, this fusion diagnostic method not only improves the assessment accuracy of the health status of the precision shaft, but also can identify the potential risks of equipment degradation in advance, help managers formulate more accurate maintenance strategies, avoid equipment failures, and improve equipment reliability and operating efficiency.

[0053] It can be seen that in this application, the lack of attribute interaction in the health status diagnosis of precision shafts under the dynamic evolution of variable working conditions can be effectively compensated; among them, through the scanning measurement of the precision shaft, the real-time operation status data of the shaft can be accurately obtained, overcoming the limitations of the traditional method that relies on the preset model, ensuring the real-time and accuracy of the data, and providing a reliable data basis for subsequent analysis, thereby enhancing the accuracy and real-time response capability of health status diagnosis; by fusing the real-time operation status data with the data twin model, the model parameters can be dynamically adjusted to reflect the real-time operation characteristics of the precision shaft. This fusion method helps to more comprehensively evaluate the health status of the precision shaft and improve the adaptability and accuracy of the diagnostic model to complex working conditions; by obtaining the linear offset, the changing pattern of the health status can be accurately captured, providing a scientific basis for predicting the degradation trend and fault occurrence of the shaft, and further improving the accuracy of fault warning; combined with the health response labels under different working conditions, the health status of the shaft under various working conditions can be more accurately described. This provides a comprehensive evaluation standard for health diagnosis in multiple working conditions, which helps to detect potential problems in advance and take preventive measures; by integrating cross-degradation indicators and time-series health gradients, the health status of precision shafts can be evaluated from multiple angles and throughout the life cycle to identify potential failure modes. This multi-dimensional diagnostic method improves the monitoring accuracy of the health status of precision shafts.

[0054] In summary, the technical solution adopted in this application can accurately capture the health status of the precision shaft under variable working conditions and complex degradation processes, so as to enhance the sensitivity of the health diagnosis model to the dynamic performance degradation of the precision shaft.

[0055] This application provides a precision shaft health status diagnosis system based on data twins, referring to Figure 2 As shown, this figure is a schematic diagram of a precision shaft health status diagnosis system based on data twins according to this embodiment of the present application, and the health status diagnosis system includes: The scanning and measuring module 100 is used to scan and measure the target precision shaft to obtain the running status data of the target precision shaft; A dynamic trend analysis module 200 is used to add the operating status data to the health status model of the data twin, output the fused health characteristics of the target precision axis, perform dynamic trend analysis on the fused health characteristics, and obtain a linear offset when the health status of the target precision axis changes linearly; The health degradation monitoring module 300 is used to determine a balanced health index reflecting the dynamic performance degradation of the precision shaft, determine a health response label of the target precision shaft under variable working conditions according to the balanced health index, and further determine a cross-degradation index of the health state of the target precision shaft according to the health response label; The health difference marking module 400 is used to construct a health characteristic curve of the target precision axis throughout its life cycle, mark the health characteristic curve with differences, obtain the health deviation point of the target precision axis at each moment during operation, and then determine the time series health gradient of the target precision axis throughout its life cycle from the health deviation point and the linear offset; The fusion diagnosis module 500 is used to perform fusion diagnosis on the health status of the precision axis according to the cross degradation index and the timing health gradient.

[0056] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0057] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, the storage medium including a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically-erasable programmable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0058] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. A precision shaft health status diagnosis method based on data twin, characterized in that: The health status diagnosis method comprises the following steps: Scan and measure the target precision axis to obtain the running status data of the target precision axis; Adding the operating status data to the health status model of the data twin, outputting the fused health characteristics of the target precision axis, performing dynamic trend analysis on the fused health characteristics, and obtaining the linear offset when the health status of the target precision axis changes linearly; Determine a balanced health index reflecting the dynamic performance degradation of the precision shaft, determine a health response label of the target precision shaft under variable working conditions according to the balanced health index, and then determine a cross-degradation index of the health state of the target precision shaft according to the health response label; Constructing a health characteristic curve of the target precision axis over its entire life cycle, marking the health characteristic curve with differences, obtaining the health deviation point of the target precision axis at each moment during operation, and then determining the temporal health gradient of the target precision axis over its entire life cycle from the health deviation point and the linear offset; A fusion diagnosis is performed on the health status of the precision axis according to the cross-degradation index and the timing health gradient.

2. A precision shaft health status diagnosis method based on data twinning as claimed in claim 1, characterized in that: The operating status data is added to the health status model of the data twin, and the fusion health characteristics of the output target precision axis specifically include: Based on the characteristics of data twins, a model of the health status of the precision shaft is constructed to obtain the health status model of the data twins; The operating status data is input into the analysis layer of the health status model of the data twin to obtain the fused health characteristics of the target precision axis.

3. A precision shaft health status diagnosis method based on data twinning as claimed in claim 1, characterized in that: The dynamic trend analysis of the fusion health feature is performed to obtain the linear offset when the health state of the target precision axis changes linearly, which specifically includes: Determine the local discrete characteristics of the target precision axis health state when it changes linearly according to the fused health characteristics; Determine the steady-state trend information of the target precision axis operation status; A linear offset when a health state of a target precision axis changes linearly is determined according to the local discrete characteristics and the steady-state trend information.

4. A method for diagnosing the health status of a precision shaft based on data twins as claimed in claim 1, characterized in that: Determining the balanced health index reflecting the dynamic performance degradation of precision shafts specifically includes: Obtain dynamic performance characteristics of precision shafts; determining a balance adjustment amount reflecting dynamic performance degradation of the precision shaft according to the dynamic performance characteristics; A balance health index reflecting dynamic performance degradation of the precision shaft is determined by the dynamic performance characteristics and the balance adjustment amount.

5. A precision shaft health status diagnosis method based on data twinning as claimed in claim 1, characterized in that: The health response label for determining the operating state of the target precision axis under variable working conditions according to the balance health index specifically includes: Determine the state attenuation information of the target precision axis's operating state under variable working conditions; Determining the distribution adaptation characteristics of the precision shaft when the running state changes according to the state attenuation information; Determine the coupling identification amount of the health state of the target precision shaft through the distributed adaptation feature; A health response label of the operating state of the target precision shaft under variable working conditions is determined according to the balance health index and the coupling identification amount.

6. A method for diagnosing the health status of a precision shaft based on data twins as claimed in claim 1, characterized in that: The cross-degradation indicators of the target precision axis health status determined by the health response tag specifically include: Determine a health status category of a target precision axis health status according to the health response tag; Determine the performance constraint level when the target precision axis performance changes through the health status category; A cross-degradation indicator of a target precision axis health state is determined according to the performance constraint hierarchy.

7. A precision shaft health status diagnosis method based on data twinning as claimed in claim 1, characterized in that: Determining the temporal health gradient of the target precision axis throughout its life cycle based on the health deviation point and the linear offset specifically includes: Determining the state affiliation characteristics of the target precision axis during operation according to the health deviation point; Determining the timing operation properties of the target precision axis throughout its life cycle according to the linear offset; The timing health gradient of the target precision axis throughout its life cycle is determined through the state affiliation characteristics and the timing operation attributes.

8. A method for diagnosing the health status of a precision shaft based on data twins as claimed in claim 1, characterized in that: The health characteristic curve refers to a curve reflecting the health change trend of a target precision shaft obtained by analyzing the health status data of the target precision shaft throughout its entire life cycle.

9. A precision shaft health status diagnosis method based on data twinning as claimed in claim 1, characterized in that: The operating status data refers to the multi-dimensional dynamic feature information collected by high-precision sensors and scanning measurement equipment during the actual operation of the precision shaft.

10. A precision shaft health status diagnosis system based on data twin, used to execute a precision shaft health status diagnosis method based on data twin as claimed in any one of claims 1 to 9, characterized in that: The health status diagnosis system comprises: A scanning measurement module is used to scan and measure the target precision axis to obtain the running status data of the target precision axis; A dynamic trend analysis module is used to add the operating status data to the health status model of the data twin, output the fused health characteristics of the target precision axis, perform dynamic trend analysis on the fused health characteristics, and obtain the linear offset when the health status of the target precision axis changes linearly; A health degradation monitoring module, used to determine a balanced health index reflecting the dynamic performance degradation of the precision shaft, determine a health response label of the target precision shaft under variable working conditions according to the balanced health index, and further determine a cross-degradation index of the health state of the target precision shaft from the health response label; A health difference marking module is used to construct a health characteristic curve of the target precision axis throughout its life cycle, to mark the health characteristic curve with differences, to obtain the health deviation point of the target precision axis at each moment during operation, and then to determine the temporal health gradient of the target precision axis throughout its life cycle from the health deviation point and the linear offset; A fusion diagnosis module is used to perform fusion diagnosis on the health status of the precision axis according to the cross-degradation index and the timing health gradient.

Images