A method and system for full parameter monitoring of intelligent corrugated compensators

Abstract

This invention relates to the field of data monitoring technology, specifically disclosing a method and system for full-parameter monitoring of an intelligent corrugated compensator. The method includes: acquiring axial extension displacement data, radial offset displacement data, and temperature and humidity time-series data of the target compensator; performing unidirectional fusion of the axial extension displacement data to obtain comprehensive axial displacement data; constructing a discrete radial displacement field; performing spatial symmetry decomposition on the discrete radial displacement field, and based on the decomposed symmetric and antisymmetric components, performing distributed force inversion on the torque currently borne by the target compensator to obtain torque data; determining that a leakage event has occurred in the target compensator when the temperature and humidity time-series data exceeds a preset temperature and humidity baseline; and performing a comprehensive health status assessment on the comprehensive axial displacement data, torque data, and leakage event to generate a full-parameter monitoring report for the target compensator. This invention can improve the efficiency of full-parameter monitoring of an intelligent corrugated compensator.

Description

Technical Field

[0001] This invention relates to the field of data monitoring technology, and in particular to a method and system for full parameter monitoring of an intelligent corrugated compensator. Background Technology

[0002] Existing monitoring technologies for corrugated compensators lack the technical design for simultaneous acquisition of multiple parameters. The monitoring of axial expansion and contraction and radial offset displacement suffers from insufficient data analysis. It is impossible to effectively fuse axial displacement data in the same direction, and it is also difficult to construct an accurate discrete radial displacement field based on radial displacement data. Furthermore, there is a lack of scientific technical support for the inversion calculation of the compensator's torsional effect, making it impossible to accurately obtain torsional-related data.

[0003] Existing bellows compensator monitoring technologies rely on a single method for judging temperature and humidity data, failing to incorporate multi-dimensional factors for leak event identification, which can easily lead to judgment biases. Furthermore, they lack the ability to comprehensively assess the health status of compensators, including displacement, torque, and leakage conditions, and cannot generate comprehensive monitoring reports. The overall efficiency and accuracy of monitoring need to be improved. Therefore, how to improve the comprehensiveness of bellows compensator monitoring parameters and the scientific nature of the assessment has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for full parameter monitoring of an intelligent corrugated compensator to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for full parameter monitoring of an intelligent corrugation compensator, comprising: Multi-parameter synchronous measurement was performed on the target compensator to obtain the axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator. The axial expansion and contraction displacement data are fused in the same direction to obtain the comprehensive axial displacement data of the target compensator; Based on the spatial orientation of the radial offset displacement data, a discrete radial displacement field of the target compensator is constructed; The discrete radial displacement field is decomposed into spatial symmetry, and based on the decomposed symmetric and antisymmetric components, the distributed force inversion is performed on the torque currently borne by the target compensator to obtain the torque data of the target compensator. When the temperature and humidity time-series data exceeds the preset temperature and humidity baseline, it is determined that a leakage event has occurred in the target compensator; A comprehensive health status assessment is performed on the axial displacement data, the torque data, and the leakage event to generate a full-parameter monitoring report for the target compensator.

[0006] In a preferred embodiment, the step of performing multi-parameter synchronous measurement on the target compensator to obtain axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator includes: At the same time reference, a trigger signal is sent to the data acquisition channel arranged on the target compensator to acquire the original axial displacement signal, the original radial displacement signal and the original temperature and humidity signal of the target compensator. The original axial displacement signal, the original radial displacement signal, and the original temperature and humidity signal are conditioned to obtain the simulated axial displacement signal, the simulated radial displacement signal, and the simulated temperature and humidity signal of the target compensator. The axial displacement simulation signal, the radial displacement simulation signal, and the temperature and humidity simulation signal are converted from analog to digital to obtain the axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator.

[0007] In a preferred embodiment, the step of performing unidirectional fusion of the axial extension displacement data to obtain the comprehensive axial displacement data of the target compensator includes: Spatial component analysis is performed on the axial expansion displacement data to obtain the multi-directional axial sub-displacements of the target compensator; Based on the directional characteristics of the multi-directional axial sub-displacements, abnormal sub-displacements in the multi-directional axial sub-displacements are eliminated to obtain the effective axial sub-displacements of the target compensator. Based on the deformation contribution of the effective axial sub-displacement, the preliminary axial displacement fusion data of the target compensator is calculated; ; In the formula, This is the initial axial displacement fusion data for the target compensator. This refers to the orientation index of the multi-directional axial sub-displacement. For the first The effective axial sub-displacements in each orientation For the first The contribution weighting coefficients corresponding to the effective sub-displacements of the axial direction in each orientation. The arithmetic mean of the effective sub-displacements along the axis. The standard deviation of the effective sub-displacement along the axis. This is a preset abnormality suppression factor; The preliminary axial displacement fusion data is subjected to time-series smoothing filtering to obtain the comprehensive axial displacement data of the target compensator.

[0008] In a preferred embodiment, constructing the discrete radial displacement field of the target compensator based on the spatial orientation of the radial offset displacement data includes: The radial offset displacement data is analyzed by measurement orientation to obtain the radial displacement metadata of the radial offset displacement data; Based on the spatial label of the radial offset displacement data, the radial displacement metadata is divided into a preset circumferential spatial partition; Spatial consistency verification is performed on the radial displacement metadata within the same circumferential spatial partition, and spatial feature extraction is performed on the valid radial displacement metadata that passes the verification to obtain the radial deformation level data of the circumferential spatial partition. Using the center azimuth angle of the radial deformation horizontal data as the index and the displacement value of the radial deformation horizontal data as the element, an initial spatial distribution mapping table of the target compensator is constructed. The initial spatial distribution mapping table is reconstructed using field data to obtain the discrete radial displacement field of the target compensator.

[0009] In a preferred embodiment, the step of reconstructing the field data of the initial spatial distribution mapping table to obtain the discrete radial displacement field of the target compensator includes: Obtain the center azimuth value corresponding to the azimuth index item in the initial spatial distribution mapping table, and generate the spatial index sequence of the initial spatial distribution mapping table; Based on the spatial index sequence, data retrieval is performed on the initial spatial distribution mapping table to obtain the corresponding feature displacement data of the spatial index sequence; Based on the circumferential adjacency relationship of the center azimuth angle in the spatial index sequence, spatial adjacency topology is constructed for the corresponding feature displacement data to obtain the spatial displacement metadata structure of the corresponding feature displacement data. The spatial displacement metadata structure is encapsulated with attribute associations to obtain the discrete radial displacement field of the target compensator.

[0010] In a preferred embodiment, the spatial symmetry decomposition of the discrete radial displacement field includes: Perform circumferential angle correlation analysis on the radial displacement data in the discrete radial displacement field to obtain the angle-displacement correspondence data of the radial displacement data. According to the circumferential angle order of the angle-displacement corresponding data, the angle-displacement corresponding data are arranged into a radial displacement distribution sequence that is continuously distributed along the circumference. The radial displacement distribution sequence is decomposed by mirror symmetry to obtain the symmetric components of the discrete radial displacement field; By performing odd-symmetric separation on the radial displacement distribution sequence, the antisymmetric component of the discrete radial displacement field is obtained.

[0011] In a preferred embodiment, the step of performing distributed force inversion on the torque currently borne by the target compensator based on the decomposed symmetric and antisymmetric components to obtain the torque data of the target compensator includes: The antisymmetric components are analyzed by alternating positive and negative circumferential analysis to obtain the antisymmetric phase characteristics of the antisymmetric components; Amplitude analysis is performed on the symmetric component, and the analyzed circumferential average amplitude and circumferential fluctuation amplitude are used as background features of the symmetric component. Based on the background features, the antisymmetric feature amplitude of the antisymmetric component is modified by interference suppression to obtain the modified antisymmetric feature amplitude of the antisymmetric component. The modified antisymmetric feature amplitude is matched with the preset torque-antisymmetric feature mapping table to determine the torque value corresponding to the modified antisymmetric feature amplitude. Based on the antisymmetric phase characteristics, the spatial orientation of the torque action of the target compensator is determined, and a torque direction identifier corresponding to the torque value is generated. The torque value and the torque direction identifier are used as the torque data of the target compensator.

[0012] In a preferred embodiment, determining that the target compensator has experienced a leakage event when the temperature and humidity time-series data exceeds a preset temperature and humidity baseline includes: Statistical regression processing was performed on the ambient temperature and humidity data of the target compensator under leak-free operating conditions to obtain the temperature and humidity baseline of the target compensator. The temperature and humidity time series data are compared with the temperature baseline range and humidity baseline range of the temperature and humidity baseline, and abnormal points that exceed the temperature baseline range and humidity baseline range are marked. The extent and duration of the exceedance of the abnormal points are recorded. When the number of abnormal points reaches a preset spatial redundancy threshold, the out-of-limit directional features of the abnormal points are extracted, and the spatial consistency result of the abnormal points is determined. The change rate analysis is performed on the temperature and humidity time series data at the abnormal points to obtain the time series abrupt change information of the temperature and humidity time series data; When the spatial consistency result indicates that the abnormal points exhibit the same over-limit direction and the temporal mutation information indicates a synchronous rapid mutation in temperature and humidity, a multi-factor fusion judgment is performed on the over-limit amplitude, the over-limit duration, and the spatial distribution location of the abnormal points to obtain the judgment result of the target compensator leaking.

[0013] In a preferred embodiment, the step of performing a comprehensive health status assessment on the axial displacement data, the torque data, and the leakage event to generate a full-parameter monitoring report for the target compensator includes: The severity of deformation is determined by analyzing the comprehensive axial displacement data to obtain the axial displacement state level of the target compensator. Based on the relative deviation of the torque data within the preset torque safety threshold range, the torque state level of the target compensator is determined. The overall operating status level of the target compensator is obtained by comprehensively analyzing the axial displacement status level, the torque status level, and the status results of the leakage event. The axial displacement data, torque data, leakage events, axial displacement status level, torque status level, status results, and overall operating status level are integrated into a full-parameter monitoring report for the target compensator.

[0014] To address the aforementioned problems, the present invention also provides a full-parameter monitoring system for an intelligent corrugation compensator, the system comprising: The multi-parameter synchronous measurement module is used to perform multi-parameter synchronous measurement on the target compensator to obtain the axial extension displacement data, radial offset displacement data and temperature and humidity time series data of the target compensator. An axial displacement homogeneous fusion module is used to fuse the axial extension displacement data in the same direction to obtain the comprehensive axial displacement data of the target compensator. The discrete radial displacement field construction module is used to construct the discrete radial displacement field of the target compensator based on the spatial orientation of the radial offset displacement data. The torque inversion solution module is used to perform spatial symmetry decomposition on the discrete radial displacement field, and based on the decomposed symmetric and antisymmetric components, to perform distributed force inversion on the torque currently borne by the target compensator, and obtain the torque data of the target compensator. The leakage event identification module is used to determine that a leakage event has occurred in the target compensator when the temperature and humidity time series data exceeds the preset temperature and humidity baseline; The comprehensive parameter evaluation module is used to conduct a comprehensive health status evaluation of the axial displacement data, the torque data, and the leakage event, and generate a comprehensive parameter monitoring report of the target compensator.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention enables simultaneous measurement of multiple parameters, including axial and radial displacement, temperature, and humidity, of a corrugated compensator. The accuracy of the original data is ensured through signal conditioning and analog-to-digital conversion. Accurate axial composite displacement data and discrete radial displacement field are obtained through unidirectional fusion and spatial orientation analysis, respectively. Combined with symmetry decomposition, accurate inversion of torque data is completed, which greatly improves the accuracy of compensator displacement and force monitoring.

[0016] 2. This invention achieves accurate identification of leakage events through multi-factor fusion judgment, effectively improving the accuracy of leakage judgment. At the same time, it classifies axial displacement and torsional state and conducts comprehensive health status assessment, generates full-parameter monitoring reports, improves the presentation of monitoring results, enhances the systematicness and overall efficiency of full-parameter monitoring of bellows compensators, and provides comprehensive and reliable data support for equipment operation and maintenance. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for monitoring all parameters of an intelligent corrugated compensator according to an embodiment of the present invention. Figure 2 This is a functional block diagram of a full-parameter monitoring system for an intelligent corrugated compensator provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for full-parameter monitoring of an intelligent corrugated compensator. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for full-parameter monitoring of an intelligent corrugated compensator can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0020] Reference Figure 1The diagram shown is a flowchart illustrating a method for monitoring all parameters of an intelligent corrugated compensator according to an embodiment of the present invention. In this embodiment, the method for monitoring all parameters of an intelligent corrugated compensator includes: Multi-parameter synchronous measurement was performed on the target compensator to obtain the axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator. In this embodiment of the invention, the step of performing multi-parameter synchronous measurement on the target compensator to obtain the axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator includes: At the same time reference, a trigger signal is sent to the data acquisition channel arranged on the target compensator to acquire the original axial displacement signal, the original radial displacement signal and the original temperature and humidity signal of the target compensator. The original axial displacement signal, the original radial displacement signal, and the original temperature and humidity signal are conditioned to obtain the simulated axial displacement signal, the simulated radial displacement signal, and the simulated temperature and humidity signal of the target compensator. The axial displacement simulation signal, the radial displacement simulation signal, and the temperature and humidity simulation signal are converted from analog to digital to obtain the axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator.

[0021] Using a high-precision crystal oscillator to provide a unified millisecond-level timestamp as the global time reference, standard-level trigger signals are simultaneously sent to the axial displacement data acquisition channels, radial displacement data acquisition channels, and temperature and humidity data acquisition channels, which are pre-deployed at preset monitoring points such as the corrugation monitoring section and flange connection section of the target compensator. This triggers the sensing elements in each data acquisition channel to synchronously enter the working state. The sensing elements in the axial displacement data acquisition channel capture the physical change signals generated by the axial expansion and contraction of the target compensator and convert them into the original axial displacement signal. The sensing elements in the radial displacement data acquisition channel capture the physical change signals generated by the radial offset of the target compensator and convert them into the original radial displacement signal. The sensing elements in the temperature and humidity data acquisition channel capture the physical change signals of temperature and humidity of the target compensator body and the surrounding monitoring environment and convert them into the original temperature and humidity signals. Each data acquisition channel completes the synchronous acquisition of the corresponding original signals under a unified millisecond-level time reference, and the acquired original axial displacement signals, original radial displacement signals, and original temperature and humidity signals are temporarily stored in their respective signal storage modules in real time.

[0022] The raw axial displacement signal, raw radial displacement signal, and raw temperature and humidity signal temporarily stored in the signal storage module are subjected to continuous signal conditioning operations of filtering, amplification, and linearization calibration in sequence. The filtering operation uses a passive RC filter circuit to process the raw signal and filter out high-frequency interference noise mixed in the raw signal. The amplification operation uses an operational amplifier to adjust the amplitude of the filtered signal to the preset signal acquisition range of the analog-to-digital converter. The linearization calibration operation corrects the amplified signal point by point according to the signal-physical quantity correspondence curve calibrated by each sensing element at the factory, eliminating the nonlinear deviation of the raw signal caused by the sensing characteristics. After the above complete signal conditioning operations, axial displacement analog signal, radial displacement analog signal, and temperature and humidity analog signal that meet the analog-to-digital conversion acquisition standard are obtained respectively.

[0023] A high-speed analog-to-digital converter (ADC) chip with 16-bit quantization precision is used to perform synchronous ADC operations on axial displacement analog signals, radial displacement analog signals, and temperature and humidity analog signals. The ADC chip discretizes the continuously changing axial displacement analog signals, radial displacement analog signals, and temperature and humidity analog signals at a preset fixed sampling frequency, converts the voltage amplitude information of the analog signals into binary digital signals, and then converts the binary digital signals into decimal values ​​that can represent the actual physical quantities one by one according to the preset signal calibration conversion relationship of each data acquisition channel. After completing the ADC conversion of the axial displacement analog signal, axial extension displacement data is obtained; after completing the ADC conversion of the radial displacement analog signal, radial offset displacement data is obtained; and after completing the ADC conversion of the temperature and humidity analog signal, it is continuously arranged and organized according to the timestamp of a unified time base, finally obtaining temperature and humidity time series data. All converted data are structured and stored by associating them with the millisecond-level timestamp of the global time base.

[0024] The beneficial effects are as follows: By employing a high-precision crystal oscillator to provide a millisecond-level unified time reference, synchronous triggering of each data acquisition channel is achieved, ensuring strict time consistency in the acquisition process of the original axial displacement, radial displacement, and temperature / humidity signals. This avoids data deviation problems caused by asynchronous acquisition time from the source of data acquisition. Through a series of signal conditioning operations consisting of a passive RC filter circuit, operational amplifier, and calibration curve correction of the sensing element, effective filtering of interference noise in the original signal, standardized adjustment of signal amplitude, and complete elimination of nonlinear deviation are achieved. This ensures that the obtained axial displacement analog signal, radial displacement analog signal, and temperature / humidity analog signal fully match the analog-to-digital conversion acquisition standard, guaranteeing the accuracy and standardization of the analog signal. A high-speed analog signal with 16-bit quantization precision is used. The digital-to-digital converter chip performs analog-to-digital conversion and completes the accurate conversion of analog signals to physical quantity values ​​according to the preset signal calibration conversion relationship. At the same time, it provides structured storage of all final data with unified millisecond-level timestamps, realizing high-quality conversion of analog signals to digital data. The resulting axial extension displacement data, radial offset displacement data, and temperature and humidity time series data can truly and accurately reflect the actual operating status of the target compensator. The entire implementation process adopts clear hardware equipment operation and quantifiable preset standards, and has a high degree of reproducibility. It provides a synchronous, accurate, and standardized raw data foundation for all subsequent data processing operations such as axial displacement fusion, discrete radial displacement field construction, and torsional data inversion, fundamentally ensuring the effectiveness, accuracy, and reliability of data processing in each subsequent step.

[0025] The axial expansion and contraction displacement data are fused in the same direction to obtain the comprehensive axial displacement data of the target compensator; In this embodiment of the invention, the step of performing unidirectional fusion of the axial extension displacement data to obtain the comprehensive axial displacement data of the target compensator includes: Spatial component analysis is performed on the axial expansion displacement data to obtain the multi-directional axial sub-displacements of the target compensator; Based on the directional characteristics of the multi-directional axial sub-displacements, abnormal sub-displacements in the multi-directional axial sub-displacements are eliminated to obtain the effective axial sub-displacements of the target compensator. Based on the deformation contribution of the effective axial sub-displacement, the preliminary axial displacement fusion data of the target compensator is calculated; ; In the formula, This is the initial axial displacement fusion data for the target compensator. This refers to the orientation index of the multi-directional axial sub-displacement. For the first The effective axial sub-displacements in each orientation For the first The contribution weighting coefficients corresponding to the effective sub-displacements of the axial direction in each orientation. The arithmetic mean of the effective sub-displacements along the axis. The standard deviation of the effective sub-displacement along the axis. This is a preset abnormality suppression factor; The preliminary axial displacement fusion data is subjected to time-series smoothing filtering to obtain the comprehensive axial displacement data of the target compensator.

[0026] The axial expansion displacement data that has been converted from analog to digital and stored in a structured manner is retrieved. This data is associated with a millisecond-level global timestamp and a preset monitoring azimuth identifier. The monitoring azimuth is set according to eight fixed azimuths evenly distributed around the target compensator, namely 0 degrees, 90 degrees, 180 degrees, 270 degrees, 45 degrees, 135 degrees, 225 degrees, and 315 degrees. For the axial expansion displacement data under each timestamp, the data is split one by one according to the preset monitoring azimuth identifier. The axial expansion displacement data collected independently in each azimuth is extracted separately and structured and classified. The independent structured data group corresponding to each azimuth is a single axial sub-displacement. The axial sub-displacements corresponding to all monitoring azimuths are integrated. The integrated structured data set is the multi-azimuth axial sub-displacement of the target compensator. Each axial sub-displacement retains the corresponding monitoring azimuth identifier and millisecond-level global timestamp information.

[0027] The displacement direction and value information of each axial sub-displacement in the multi-directional axial sub-displacement are extracted. The displacement direction is based on the positive direction of the axial extension and contraction of the target compensator as a unified reference. This positive direction is set as the extension direction of the compensator along the central axis, and the negative direction is the contraction direction of the compensator along the central axis. The mainstream direction of the overall axial deformation of the compensator is determined by statistically analyzing the proportion of displacement directions of all axial sub-displacements under the same millisecond-level timestamp. The displacement direction with a proportion of 90% or more is the mainstream direction under the current timestamp. Axial sub-displacements whose displacement direction is inconsistent with the mainstream direction are marked as abnormal sub-displacements. At the same time, the arithmetic mean of the values ​​of all axial sub-displacements under the same timestamp is calculated. The displacement value deviation threshold is set to ±20% of the arithmetic mean. Axial sub-displacements whose displacement values ​​exceed the threshold range are also marked as abnormal sub-displacements. All data in the multi-directional axial sub-displacements are checked one by one. All marked abnormal sub-displacements are completely removed. The structured data set of the remaining unmarked axial sub-displacements is the effective axial sub-displacement of the target compensator. The effective axial sub-displacements still retain the corresponding monitoring orientation identifier and millisecond-level global timestamp information.

[0028] Based on the actual influence of the monitoring azimuth corresponding to each axial effective sub-displacement on the overall axial deformation of the target compensator, a fixed deformation contribution weight coefficient is set for each axial effective sub-displacement. Specifically, the contribution weight coefficient for the 0-degree and 180-degree monitoring azimuths at the compensator flange connection is set to 0.2; the contribution weight coefficient for the 90-degree and 270-degree monitoring azimuths in the middle of the corrugated section is set to 0.1; and the contribution weight coefficient for the remaining 45-degree, 135-degree, 225-degree, and 315-degree monitoring azimuths is set to 0.15. First, the arithmetic mean of all axial effective sub-displacements at the same time point is calculated by summing the values ​​of all axial effective sub-displacements and then dividing the sum by the total number of axial effective sub-displacements. Next, the standard deviation of all axial effective sub-displacements at the same time point is calculated by first calculating the difference between each axial effective sub-displacement and the arithmetic mean, squaring each difference, summing the squares, and then dividing the sum by the total number of axial effective sub-displacements. The result is then square-rooted. Based on the target compensator model and a fixed anomaly suppression factor of three, amplitude suppression is applied to each axial effective sub-displacement. This is done by calculating the absolute value of the difference between a single axial effective sub-displacement and its arithmetic mean, dividing this absolute value by the standard deviation to obtain the deviation of the single axial effective sub-displacement, and then performing amplitude attenuation on this deviation using exponential calculation. The value of the single axial effective sub-displacement is multiplied by the attenuation result to obtain the suppressed value of that axial effective sub-displacement. Subsequently, the contribution weight coefficient of each axial effective sub-displacement is multiplied by the corresponding suppressed value. All multiplication results are summed, and the sum of all axial effective sub-displacement contribution weight coefficients is calculated. The sum of the former is divided by the latter sum to obtain the preliminary axial displacement fusion data of the target compensator at the same millisecond-level timestamp. This complete calculation operation is performed on the axial effective sub-displacement at each timestamp to finally obtain the full-time-series preliminary axial displacement fusion data.

[0029] A sliding window filtering method is used to perform time-series smoothing filtering on the preliminary axial displacement fusion data of the entire time series. The window length of the sliding window is set to fifteen consecutive millisecond-level timestamps. The window slides along the time axis with a fixed step size of a single timestamp. For all the preliminary axial displacement fusion data in each sliding window, their arithmetic mean is calculated. This average value is used as the filtered data corresponding to the center timestamp of the window. For the parts where the first seven timestamps at the beginning of the time axis and the last seven timestamps at the end cannot form a complete window, the data at the beginning and end are supplemented by extending the data from the beginning and end. That is, the filtered data corresponding to the timestamps of the incomplete windows at the beginning and end directly adopts the filtered data of the center timestamp of the adjacent complete window. The filtered data corresponding to all millisecond-level timestamps are integrated and arranged in a structured manner according to the chronological order of the time axis. The resulting structured time series data is the axial comprehensive displacement data of the target compensator.

[0030] The beneficial effects include spatial component analysis of axial expansion and contraction displacement data according to preset fixed monitoring orientations, obtaining multi-directional axial sub-displacements through clear orientational decomposition and structured classification, enabling precise extraction of the spatial characteristics of axial displacement data, laying a clear spatial data foundation for subsequent unidirectional fusion operations, and constructing a dual judgment standard based on a threshold of 90% mainstream displacement direction and a threshold of ±20% deviation of displacement value from the arithmetic mean, achieving accurate identification and complete elimination of abnormal sub-displacements. The obtained effective axial sub-displacements can truly reflect the actual axial deformation state of the target compensator, avoiding abnormal displacements from the data source. To avoid interference from outlier data in subsequent fusion results, a fixed deformation contribution weighting coefficient is set based on the actual impact of each monitoring orientation on the overall axial deformation of the compensator. Combined with the specific calculation methods of the arithmetic mean and standard deviation, and a preset fixed anomaly suppression factor, amplitude suppression processing is applied to the effective axial sub-displacements. The preliminary axial displacement fusion data is then calculated by weighted summation and divided by the sum of the weighting coefficients. This approach fully considers the actual impact of effective axial sub-displacements from different orientations on the overall axial deformation, and also allows for targeted amplitude attenuation of effective axial sub-displacements that deviate from the overall average level, effectively reducing the impact of outlier data on the fusion process. To minimize interference and ensure accurate alignment of the preliminary axial displacement fusion data with the actual axial deformation of the compensator, a time-series smoothing filter was applied to the preliminary axial displacement fusion data using a sliding window filtering method with fixed window length and fixed step size. The axial composite displacement data was obtained by calculating the arithmetic mean within the window and supplementing the data from the beginning and end, effectively eliminating local fluctuations in the time-series data. This resulted in a more stable time-series characteristic of the axial composite displacement data, more accurately reflecting the true time-series changes in the axial expansion and contraction of the compensator. Throughout the entire implementation process, all monitoring orientations, judgment thresholds, weighting coefficients, anomaly suppression factors, and filtering window parameters were... With fixed values ​​set, all operation steps have clear execution methods and specific calculation processes, exhibiting high reproducibility. The products of each step strictly adhere to technical requirements, and the correlation between data is clear and explicit. From spatial component analysis to abnormal sub-displacement removal, and then to preliminary data fusion calculation and temporal smoothing filtering, the entire process ensures that the final axial composite displacement data possesses both spatial accuracy and temporal stability. It can comprehensively, realistically, and accurately reflect the axial expansion and contraction displacement state of the target compensator, providing high-quality and highly reliable axial displacement data support for subsequent comprehensive health status assessments of the target compensator.

[0031] Based on the spatial orientation of the radial offset displacement data, a discrete radial displacement field of the target compensator is constructed; In this embodiment of the invention, constructing the discrete radial displacement field of the target compensator based on the spatial orientation of the radial offset displacement data includes: The radial offset displacement data is analyzed by measurement orientation to obtain the radial displacement metadata of the radial offset displacement data; Based on the spatial label of the radial offset displacement data, the radial displacement metadata is divided into a preset circumferential spatial partition; Spatial consistency verification is performed on the radial displacement metadata within the same circumferential spatial partition, and spatial feature extraction is performed on the valid radial displacement metadata that passes the verification to obtain the radial deformation level data of the circumferential spatial partition. Using the center azimuth angle of the radial deformation horizontal data as the index and the displacement value of the radial deformation horizontal data as the element, an initial spatial distribution mapping table of the target compensator is constructed. The initial spatial distribution mapping table is reconstructed using field data to obtain the discrete radial displacement field of the target compensator.

[0032] The process of reconstructing the field data from the initial spatial distribution mapping table to obtain the discrete radial displacement field of the target compensator includes: Obtain the center azimuth value corresponding to the azimuth index item in the initial spatial distribution mapping table, and generate the spatial index sequence of the initial spatial distribution mapping table; Based on the spatial index sequence, data retrieval is performed on the initial spatial distribution mapping table to obtain the corresponding feature displacement data of the spatial index sequence; Based on the circumferential adjacency relationship of the center azimuth angle in the spatial index sequence, spatial adjacency topology is constructed for the corresponding feature displacement data to obtain the spatial displacement metadata structure of the corresponding feature displacement data. The spatial displacement metadata structure is encapsulated with attribute associations to obtain the discrete radial displacement field of the target compensator.

[0033] The radial offset displacement data that has been converted from analog to digital and stored in a structured manner is retrieved. This data is associated with a millisecond-level global timestamp, a preset measurement azimuth code, and a spatial label. The measurement azimuth is evenly distributed with 24 fixed monitoring points within a 360-degree circumference of the target compensator. Each point corresponds to a unique measurement azimuth code and a spatial label containing a specific azimuth angle range. The radial offset displacement data under each timestamp is extracted point by point, and the core information contained in the data, such as the radial displacement value, millisecond-level global timestamp, measurement azimuth code, and spatial label, is separated one by one. The extracted core information is then structured and integrated with each monitoring point as the basic unit. The structured information set corresponding to each monitoring point is the single radial displacement metadata. The radial displacement metadata corresponding to all 24 monitoring points is integrated as a whole and stored according to timestamp classification to obtain the radial displacement metadata of the target compensator. All radial displacement metadata retains complete core information and forms a precise one-to-one correspondence with the original radial offset displacement data.

[0034] The 360-degree circumferential range of the target compensator is equally divided into eight pre-defined circumferential spatial zones. Each circumferential spatial zone corresponds to a 45-degree azimuth range, specifically: 0-45 degrees, 45-90 degrees, 90-135 degrees, 135-180 degrees, 180-225 degrees, 225-270 degrees, 270-315 degrees, and 315-360 degrees. Each circumferential spatial zone is assigned a unique numerical zone code, and a fixed center azimuth angle is also assigned to each zone: 22.5 degrees for the 0-45 degree zone, 67.5 degrees for the 45-90 degree zone, 112.5 degrees for the 90-135 degree zone, and 112.5 degrees for the 135-180 degree zone. The 80-degree zone is 157.5 degrees, the 180-225-degree zone is 222.5 degrees, the 225-270-degree zone is 267.5 degrees, the 270-315-degree zone is 312.5 degrees, and the 315-360-degree zone is 337.5 degrees. The spatial tag of the radial displacement metadata contains the specific azimuth range information to which the data belongs. Based on the azimuth range information, each radial displacement metadata is accurately matched to the corresponding circumferential spatial zone. All radial displacement metadata in the same circumferential spatial zone are classified and integrated and stored separately according to the zone code. The radial displacement metadata of different zones maintains an independent storage state and retains complete core information.

[0035] Spatial consistency verification is performed on all radial displacement metadata within the same circumferential spatial partition under the same millisecond-level global timestamp. First, the radial displacement values ​​of all radial displacement metadata within the partition are summed. Then, the sum is divided by the total number of radial displacement metadata within the partition to obtain the arithmetic mean of the displacement values ​​of the radial displacement metadata within the partition. The spatial consistency verification threshold is set to ±15% of this arithmetic mean. Radial displacement metadata with displacement values ​​exceeding this threshold is marked as inconsistent data and completely removed from the metadata set of the partition. The remaining unmarked radial displacement metadata is the valid radial displacement metadata for the partition. Spatial features are extracted from the effective radial displacement metadata. The displacement values ​​of all effective radial displacement metadata in the partition are summed and divided by the total number of effective radial displacement metadata. The resulting arithmetic mean is used as the displacement value of the circumferential spatial partition. This displacement value is integrated with the center azimuth angle and millisecond-level global timestamp of the corresponding circumferential spatial partition in a structured manner. The resulting single structured data unit is the radial deformation level data of the circumferential spatial partition. The eight circumferential spatial partitions under the same timestamp generate corresponding radial deformation level data independently. The full-time radial displacement metadata is used to generate full-time radial deformation level data in sequence according to the timestamp.

[0036] Using the center azimuth angle of each circumferential spatial partition as the unique index item in the initial spatial distribution mapping table, the center azimuth angles of the eight circumferential spatial partitions are arranged in ascending circumferential order of 22.5 degrees, 67.5 degrees, 112.5 degrees, 157.5 degrees, 222.5 degrees, 267.5 degrees, 312.5 degrees, and 337.5 degrees to determine the fixed position of each index item in the mapping table. The displacement value corresponding to the radial deformation horizontal data of each circumferential spatial partition is used as the mapping element for the corresponding center azimuth angle index item. This ensures that the data under the same millisecond-level global timestamp are mapped... The index entries of the center azimuth angle of the circumferential spatial partition and the corresponding displacement value elements are structurally integrated to form a key-value pair mapping structure with the center azimuth angle as the horizontal axis and the displacement value as the vertical axis. Each index entry in this mapping structure corresponds to a unique element value, and all index entries are strictly arranged in circumferential order. This standardized key-value pair mapping structure is the initial spatial distribution mapping table of the target compensator at the corresponding timestamp. The radial deformation horizontal data of the whole time series are used to generate the corresponding initial spatial distribution mapping table according to the millisecond timestamp, forming a set of initial spatial distribution mapping tables of the whole time series.

[0037] Extract the central azimuth values ​​corresponding to all index entries in the initial spatial distribution mapping table under a single millisecond-level global timestamp, and rearrange these central azimuth values ​​in an ascending circumferential order from 22.5 degrees to 337.5 degrees. During the sorting process, the circumferential spatial partition code and displacement value association information corresponding to each central azimuth value are fully preserved. The sorted central azimuth values ​​form a continuous sequence structure in a fixed order. This sequence structure is the spatial index sequence of the initial spatial distribution mapping table. Each central azimuth value in the spatial index sequence is a unique value and forms a precise one-to-one association with the original index entry in the initial spatial distribution mapping table. The initial spatial distribution mapping table of the entire time series generates the corresponding spatial index sequence independently in this way.

[0038] Based on the fixed arrangement order of the center azimuth angle values ​​in the spatial index sequence, and using a single center azimuth angle value as the precise retrieval basis, directional data retrieval is performed sequentially in the corresponding initial spatial distribution mapping table. The displacement value element corresponding to each center azimuth angle index item is extracted, and all retrieved displacement value elements are continuously and structurally organized according to the arrangement order of the center azimuth angles in the spatial index sequence to form an ordered displacement value sequence structure. In this sequence structure, each displacement value has a unique one-to-one correspondence with the corresponding center azimuth angle value in the spatial index sequence. This standardized displacement value sequence structure is the corresponding feature displacement data of the spatial index sequence. Each spatial index sequence generates a unique corresponding feature displacement data, and the two maintain a precise correlation and matching relationship throughout the process.

[0039] Based on the circumferential adjacency of the central azimuth angle values ​​in the spatial index sequence, a spatial adjacency topology is constructed for the corresponding feature displacement data. The central azimuth angles of 337.5 degrees and 22.5 degrees are set as circumferentially adjacent. The remaining central azimuth angles are sequentially paired in ascending circumferential order from 22.5 degrees to 337.5 degrees. Each central azimuth angle value in the spatial index sequence and its corresponding displacement value in the feature displacement data are taken as an independent basic topology node. Each basic topology node fully contains the two core pieces of information: central azimuth angle and displacement value. Adjacent basic topology nodes are seamlessly connected through fixed topology connections according to the preset circumferential adjacency relationship to form a closed circumferential spatial topology structure. Then, the central azimuth angle, displacement value, and corresponding topology connection relationship of each basic topology node are systematically and structurally integrated. The structured data set of all basic topology nodes is the spatial displacement metadata structure of the corresponding feature displacement data. This structure has a complete and clear circumferential spatial topology association relationship.

[0040] The core attributes such as the center azimuth, displacement value, and topological connection relationship of all basic topological nodes in the spatial displacement metadata structure are extracted. At the same time, the corresponding millisecond-level global timestamp, the circumferential spatial partition code, and the number of valid radial displacement metadata in the corresponding partition are associated. The core attributes and auxiliary attributes are encapsulated in an integrated attribute association. The encapsulation adopts a standardized structured attribute key-value pair form, assigning a unique attribute name to each attribute, and each attribute name corresponds to a unique attribute value. All encapsulated attribute key-value pairs are integrated by basic topological nodes. Then, the encapsulated data of all basic topological nodes are arranged in an orderly manner according to the circumferential topological order. The resulting standardized spatial data field with complete spatial topological structure, core attributes, and auxiliary attributes is the discrete radial displacement field of the target compensator. The full-time spatial displacement metadata structure generates the full-time discrete radial displacement field according to the timestamp, and all discrete radial displacement fields retain complete spatiotemporal information and attribute information.

[0041] The beneficial effects include obtaining radial displacement metadata from the radial offset displacement data, accurately decomposing its spatiotemporal characteristics, and laying a standardized data foundation for subsequent processing. The compensator is divided into eight equidistant partitions along its circumference with fixed center azimuth angles, achieving spatially ordered classification of the radial displacement data and making subsequent processing more targeted. Spatial consistency verification is performed using a threshold of ±15% of the arithmetic mean. The extracted radial deformation level data after removing invalid data accurately reflects the actual radial deformation state of each partition. By constructing an initial spatial distribution mapping table, generating a spatial index sequence, building a closed spatial adjacency topology, and integrating core and auxiliary attributes for encapsulation, the resulting discrete radial displacement field possesses a complete spatial topological structure, temporal sequence, and attribute information. The entire implementation process has fixed parameters, standardized steps, high reproducibility, and accurate correlation of the product data of each step. The final discrete radial displacement field comprehensively and accurately reflects the spatial distribution and temporal changes of the compensator's radial offset, providing a high-quality spatial data foundation for subsequent spatial symmetry decomposition and torsional data inversion, ensuring the accuracy and scientific rigor of subsequent torsional inversion calculations.

[0042] The discrete radial displacement field is decomposed into spatial symmetry, and based on the decomposed symmetric and antisymmetric components, the distributed force inversion is performed on the torque currently borne by the target compensator to obtain the torque data of the target compensator. In this embodiment of the invention, the spatial symmetry decomposition of the discrete radial displacement field includes: Perform circumferential angle correlation analysis on the radial displacement data in the discrete radial displacement field to obtain the angle-displacement correspondence data of the radial displacement data. According to the circumferential angle order of the angle-displacement corresponding data, the angle-displacement corresponding data are arranged into a radial displacement distribution sequence that is continuously distributed along the circumference. The radial displacement distribution sequence is decomposed by mirror symmetry to obtain the symmetric components of the discrete radial displacement field; By performing odd-symmetric separation on the radial displacement distribution sequence, the antisymmetric component of the discrete radial displacement field is obtained.

[0043] The distribution force inversion of the torque currently borne by the target compensator is performed based on the decomposed symmetric and antisymmetric components to obtain the torque data of the target compensator, including: The antisymmetric components are analyzed by alternating positive and negative circumferential analysis to obtain the antisymmetric phase characteristics of the antisymmetric components; Amplitude analysis is performed on the symmetric component, and the analyzed circumferential average amplitude and circumferential fluctuation amplitude are used as background features of the symmetric component. Based on the background features, the antisymmetric feature amplitude of the antisymmetric component is modified by interference suppression to obtain the modified antisymmetric feature amplitude of the antisymmetric component. The modified antisymmetric feature amplitude is matched with the preset torque-antisymmetric feature mapping table to determine the torque value corresponding to the modified antisymmetric feature amplitude. Based on the antisymmetric phase characteristics, the spatial orientation of the torque action of the target compensator is determined, and a torque direction identifier corresponding to the torque value is generated. The torque value and the torque direction identifier are used as the torque data of the target compensator.

[0044] Retrieve the discrete radial displacement field of the constructed target compensator, extract the center azimuth angle values ​​and corresponding radial displacement values ​​of all basic topology nodes in the data field, take a single center azimuth angle as an independent unit, and structurally associate each center azimuth angle with its unique corresponding radial displacement value. During the association process, retain the corresponding millisecond-level global timestamp. After integrating all the structured data units that have completed the azimuth and displacement association, the angle-displacement correspondence data of the radial displacement data is obtained.

[0045] The angle-displacement data are arranged continuously in circumferential order of 22.5 degrees, 67.5 degrees, 112.5 degrees, 157.5 degrees, 222.5 degrees, 267.5 degrees, 312.5 degrees, and 337.5 degrees, according to the central azimuth angle. After arrangement, an ordered data sequence is formed, which is distributed sequentially along the circumference. Each position in the sequence corresponds to a fixed circumferential angle and a unique radial displacement value. This ordered data sequence is the radial displacement distribution sequence that is continuously distributed along the circumference.

[0046] Using 180 degrees as a fixed mirror symmetry axis, determine the mirror symmetry angle for each circumferential angle in the radial displacement distribution sequence. The mirror symmetry angle is calculated by adding 180 degrees to the original angle value. If the result exceeds 360 degrees, subtract 360 degrees. Summate the radial displacement value of each angle with the radial displacement value of its mirror symmetry angle, and then divide the sum by 2 to obtain the mirror symmetry displacement value of that angle. Arrange the mirror symmetry displacement values ​​corresponding to all circumferential angles in the original circumferential order of the radial displacement distribution sequence. The resulting ordered data sequence is the symmetry component of the discrete radial displacement field.

[0047] Using 180 degrees as a fixed odd-symmetric separation axis, the mirror symmetry angle of each circumferential angle in the radial displacement distribution sequence is determined in the same way. The radial displacement value of each angle is subtracted from the radial displacement value of its mirror symmetry angle, and the result of the subtraction is divided by 2 to obtain the odd-symmetric displacement value of that angle. All the odd-symmetric displacement values ​​corresponding to the circumferential angles are arranged in the original circumferential order of the radial displacement distribution sequence. The resulting ordered data sequence is the antisymmetric component of the discrete radial displacement field.

[0048] Extract the positive and negative numerical attributes of the odd symmetric displacement values ​​corresponding to all circumferential angles in the antisymmetric component. Organize the alternating pattern of the positive and negative attributes of the odd symmetric displacement values ​​in circumferential order from 22.5 degrees to 337.5 degrees. At the same time, determine the starting circumferential angle of this alternating pattern. The obtained alternating pattern and the corresponding starting circumferential angle are structurally integrated. The integrated structured information is the antisymmetric phase feature of the antisymmetric component.

[0049] Extract the mirror symmetric displacement values ​​corresponding to all circumferential angles in the symmetric component, sum all displacement values, and then divide the sum by the total number of data in the symmetric component. The resulting value is the circumferential average amplitude of the symmetric component. Calculate the difference between each mirror symmetric displacement value and the circumferential average amplitude, sum the absolute values ​​of all differences, and then divide the sum by the total number of data. The resulting value is the circumferential fluctuation amplitude of the symmetric component. Integrate the circumferential average amplitude and the circumferential fluctuation amplitude, and use the two integrated values ​​together as the background feature of the symmetric component.

[0050] Extract the odd-symmetric displacement values ​​corresponding to all circumferential angles in the antisymmetric component, take the absolute value of each displacement value and sum them, then divide the sum by the total number of data in the antisymmetric component. The resulting value is the original antisymmetric characteristic amplitude of the antisymmetric component. Divide the circumferential fluctuation amplitude in the background feature by the circumferential average amplitude to obtain the interference correction coefficient. If the result of this coefficient is negative, it is taken as 0. Multiply the original antisymmetric characteristic amplitude by 1 and subtract the interference correction coefficient to obtain the final value, which is the corrected antisymmetric characteristic amplitude of the antisymmetric component.

[0051] Retrieve the preset torque-antisymmetric feature mapping table. This mapping table is a standardized key-value pair structure. The table contains continuous and non-overlapping antisymmetric feature amplitude ranges. Each amplitude range corresponds to a unique torque value. The corrected antisymmetric feature amplitude is precisely matched with the amplitude range in the mapping table to determine the specific range to which the amplitude belongs. The unique torque value corresponding to this range is the torque value corresponding to the corrected antisymmetric feature amplitude.

[0052] Based on the initial circumferential angle in the antisymmetric phase characteristics, the circumferential spatial partition to which this angle belongs is determined. This circumferential spatial partition is the spatial orientation of the torque action of the target compensator. The preset torque direction determination rule is that positive values ​​of odd-symmetric displacement increase along the circumference in the counterclockwise direction, and negative values ​​increase along the circumferential direction in the clockwise direction. According to the alternating positive and negative change pattern in the antisymmetric phase characteristics, the torque action direction of the target compensator is determined according to this rule, and the corresponding clockwise or counterclockwise torque direction identifier is generated.

[0053] The determined torque value, the spatial orientation of the torque action, and the generated torque direction identifier are integrated into a structured whole. During the integration process, the corresponding millisecond-level global timestamp is associated to form complete structured data containing torque value, torque direction identifier, spatial orientation of the torque action, and timestamp. This structured data is the torque data of the target compensator. The discrete radial displacement field of the whole time series is processed in sequence according to the above steps to generate the whole time series torque data.

[0054] The beneficial effects are as follows: by fixing the 180-degree axis of symmetry, the discrete radial displacement field is decomposed into mirror symmetry and separated into odd symmetry, achieving precise separation of symmetric and antisymmetric components. This allows for the effective isolation of displacement characteristics corresponding to torsional action. Based on the positive and negative attributes of the antisymmetric components, the phase characteristics are analyzed, and the amplitude of the symmetric components is analyzed to obtain background characteristics. Interference suppression and correction of the antisymmetric characteristic amplitude are also completed, effectively reducing the impact of radial deformation background interference on torsional inversion. A pre-defined standardized torsional-antisymmetric characteristic mapping table is used to achieve precise matching between amplitude and torsional values. Combined with phase characteristics, the orientation and direction of torsional action are determined, making the torsional data inversion results more comprehensive. The entire implementation process has standardized operation, fixed judgment rules, and clear calculation methods, resulting in high reproducibility. The data of each step are accurately correlated, and the final torsional data can truly reflect the torsional action state of the target compensator, providing accurate and reliable torsional monitoring data support for subsequent comprehensive health status assessment.

[0055] When the temperature and humidity time-series data exceeds the preset temperature and humidity baseline, it is determined that a leakage event has occurred in the target compensator; In this embodiment of the invention, determining that the target compensator has experienced a leakage event when the temperature and humidity time-series data exceeds a preset temperature and humidity baseline includes: Statistical regression processing was performed on the ambient temperature and humidity data of the target compensator under leak-free operating conditions to obtain the temperature and humidity baseline of the target compensator. The temperature and humidity time series data are compared with the temperature baseline range and humidity baseline range of the temperature and humidity baseline, and abnormal points that exceed the temperature baseline range and humidity baseline range are marked. The extent and duration of the exceedance of the abnormal points are recorded. When the number of abnormal points reaches a preset spatial redundancy threshold, the out-of-limit directional features of the abnormal points are extracted, and the spatial consistency result of the abnormal points is determined. The change rate analysis is performed on the temperature and humidity time series data at the abnormal points to obtain the time series abrupt change information of the temperature and humidity time series data; When the spatial consistency result indicates that the abnormal points exhibit the same over-limit direction and the temporal mutation information indicates a synchronous rapid mutation in temperature and humidity, a multi-factor fusion judgment is performed on the over-limit amplitude, the over-limit duration, and the spatial distribution location of the abnormal points to obtain the judgment result of the target compensator leaking.

[0056] The ambient temperature and humidity data of the target compensator under leak-free operating conditions were retrieved. The leak-free operating condition is a continuous and stable operating condition in which the target compensator operates at full load and no medium leakage is detected. Raw ambient temperature and humidity monitoring data were collected continuously for 72 hours under this condition. The sampling interval was consistent with the temperature and humidity time series data at the millisecond level. Statistical regression processing was performed on this batch of temperature and humidity data to calculate the arithmetic mean and standard deviation of the temperature data. The range of the temperature arithmetic mean plus or minus 2 standard deviations was set as the temperature baseline interval. The arithmetic mean and standard deviation of the humidity data were calculated. The range of the humidity arithmetic mean plus or minus 2 standard deviations was set as the humidity baseline interval. The temperature baseline interval and the humidity baseline interval together constitute the temperature and humidity baseline of the target compensator. All baseline values ​​were retained to three decimal places and stored in a structured manner.

[0057] The target compensator's temperature and humidity time-series data, which has undergone analog-to-digital conversion, is retrieved. This data is associated with a unique monitoring point code, a millisecond-level global timestamp, and the corresponding temperature and humidity values. The temperature and humidity time-series data is split point by point according to the monitoring point and timestamp. The temperature value at each timestamp of each monitoring point is compared with the temperature baseline interval in the temperature and humidity baseline. The humidity value is also compared with the humidity baseline interval. If the temperature value exceeds the temperature baseline interval or the humidity value exceeds the humidity baseline interval, the data point at that timestamp of that monitoring point is marked as an abnormal point. At the same time, the exceedance range of the abnormal point is calculated. The exceedance range of temperature is the absolute value of the difference between the measured temperature value and the boundary value of the temperature baseline interval, and the exceedance range of humidity is the absolute value of the difference between the measured humidity value and the boundary value of the humidity baseline interval. The number of consecutive timestamps from the first time the abnormal point exceeds the baseline to the time when it returns to the baseline range is counted. The actual duration of the exceedance of the abnormal point is converted according to the sampling interval and used as the exceedance duration of the abnormal point. The exceedance range and exceedance duration of all marked abnormal points are recorded and structurally associated.

[0058] The preset spatial redundancy threshold for abnormal points is 70% of the total number of temperature and humidity monitoring points of the target compensator. The total number of abnormal points marked at the same timestamp is counted. When this number reaches the spatial redundancy threshold, the out-of-limit directional features of each abnormal point are extracted. Temperature out-of-limit directional features are divided into two categories: temperature value above the upper limit of the temperature baseline interval and temperature value below the lower limit of the temperature baseline interval. Humidity out-of-limit directional features are divided into two categories: humidity value above the upper limit of the humidity baseline interval and humidity value below the lower limit of the humidity baseline interval. The temperature and humidity out-of-limit directional features of all abnormal points are checked one by one. If the temperature out-of-limit directional features of all abnormal points are completely consistent and the humidity out-of-limit directional features are completely consistent, the spatial consistency result is judged as the abnormal points showing the same out-of-limit direction. If the temperature or humidity out-of-limit directional features of any abnormal point are inconsistent with those of other points, the spatial consistency result is judged as the abnormal points not showing the same out-of-limit direction.

[0059] Extract all temperature and humidity time-series data marked as anomalies. Perform continuous numerical analysis on the temperature and humidity data of each individual point in millisecond-level timestamp order. Calculate the rate of temperature change and the rate of humidity change between adjacent timestamps. The temperature rate of change is the ratio of the temperature difference between adjacent timestamps to the time interval, and the humidity rate of change is the ratio of the humidity difference between adjacent timestamps to the time interval. Preset thresholds for temperature and humidity abrupt changes are: temperature rate of change ≥ 0.5℃ / second and humidity rate of change ≥ 3%RH / second. If, within the same time interval, the temperature and humidity rates of change at an anomaly point simultaneously reach the corresponding abrupt change thresholds, it is determined that a synchronous rapid temperature and humidity abrupt change exists within that time period. The determination result and the corresponding time interval information are then structurally integrated to form the structured data, which is the time-series abrupt change information of the temperature and humidity time-series data.

[0060] When the spatial consistency result shows that the abnormal points exhibit the same over-limit direction and the temporal mutation information indicates a synchronous rapid change in temperature and humidity, a multi-factor fusion judgment is performed on the over-limit amplitude, over-limit duration, and spatial distribution location of the abnormal points. The preset multi-factor judgment criteria are: temperature over-limit amplitude ≥2℃ and humidity over-limit amplitude ≥5%RH, over-limit duration ≥3 seconds, and the spatial distribution of the abnormal points covering 3 or more preset circumferential spatial partitions of the target compensator. If the relevant data of the abnormal points simultaneously meet the above three judgment criteria, the judgment result of the target compensator having a leakage event is obtained. If any judgment criterion is not met, the judgment result of the target compensator not having a leakage event is obtained. The judgment results are all associated with the corresponding timestamp and the relevant data of the abnormal points to complete the structured storage.

[0061] The beneficial effects are as follows: a temperature and humidity baseline is constructed based on continuous temperature and humidity data under leak-free operating conditions through statistical regression, ensuring that the baseline is supported by real operating data and guaranteeing the scientific basis of leakage judgment. By comparing the time-series temperature and humidity data with the baseline, marking abnormal points and recording relevant parameters, accurate identification of abnormal data and complete information retention are achieved. A fixed spatial redundancy threshold is set as the trigger condition for subsequent judgments. Combined with the characteristics of the over-limit direction, spatial consistency is judged, effectively avoiding misjudgments caused by single-point data anomalies. By calculating the rate of change and setting a fixed mutation threshold to analyze the time-series mutation information, accurate judgment of synchronous rapid mutations in temperature and humidity is achieved. Finally, the fixed standard for multi-factor fusion judgment is used to obtain the leakage event judgment result. Comprehensive verification is carried out from multiple dimensions such as amplitude, duration, and spatial distribution, which greatly improves the accuracy and reliability of leakage event judgment, effectively avoids the occurrence of missed judgments and misjudgments, and provides accurate leakage status data for compensator health status assessment.

[0062] A comprehensive health status assessment is performed on the axial displacement data, the torque data, and the leakage event to generate a full-parameter monitoring report for the target compensator.

[0063] In this embodiment of the invention, the step of performing a comprehensive health status assessment on the axial displacement data, the torque data, and the leakage event to generate a full-parameter monitoring report for the target compensator includes: The severity of deformation is determined by analyzing the comprehensive axial displacement data to obtain the axial displacement state level of the target compensator. Based on the relative deviation of the torque data within the preset torque safety threshold range, the torque state level of the target compensator is determined. The overall operating status level of the target compensator is obtained by comprehensively analyzing the axial displacement status level, the torque status level, and the status results of the leakage event. The axial displacement data, torque data, leakage events, axial displacement status level, torque status level, status results, and overall operating status level are integrated into a full-parameter monitoring report for the target compensator.

[0064] The axial comprehensive displacement data of the target compensator is retrieved. This data is time-series structured data associated with millisecond-level global timestamps. Preset numerical thresholds for judging the axial displacement status level are set: the safe operation threshold is 0-5mm, the warning threshold is 5-8mm, and the danger threshold is ≥8mm. The displacement values ​​in the axial comprehensive displacement data are verified point by point according to the millisecond-level timestamps. If the displacement value is in the 0-5mm range, the axial displacement status level at that timestamp is judged as Level 1 Normal. If the displacement value is in the 5-8mm range, the axial displacement status level at that timestamp is judged as Level 2 Warning. If the displacement value is ≥8mm, the axial displacement status level at that timestamp is judged as Level 3 Danger. The displacement values ​​at all timestamps are structured and associated with the corresponding judged axial displacement status levels to form full-time axial displacement status level data.

[0065] The torque data of the target compensator is retrieved, and the torque value associated with the millisecond-level global timestamp is extracted. The preset torque safety threshold range is 0-100 N·m. The relative deviation of the torque value is calculated based on the midpoint of this range, 50 N·m. The calculation method is to divide the absolute value of the difference between the measured torque value and 50 N·m by 50 N·m, and then convert the result into a percentage. The preset deviation threshold for torque status level judgment is set: 0% relative deviation is Level 1 normal, relative deviation greater than 0% and ≤20% is Level 2 warning, and relative deviation >20% is Level 3 danger. The relative deviation of each torque value is calculated according to the millisecond-level timestamp and matched with the corresponding judgment threshold to determine the torque status level at each timestamp. The torque value, relative deviation and corresponding torque status level are structurally associated to form full-time torque status level data.

[0066] The system retrieves full-time axial displacement status level data, torque status level data, and leakage event judgment results. Leakage event status results are categorized as either "occurred" or "not occurred." A pre-defined parameter comprehensive judgment rule is used: if the axial displacement status level is Level 1, the torque status level is Level 1, and the leakage event status result is "not occurred," the overall operating status level is determined to be Level 1 (normal). If either the axial displacement status level or the torque status level is Level 2, and the others are Level 1, and the leakage event status result is "not occurred," the overall operating status level is determined to be Level 2 (early warning). If either the axial displacement status level or the torque status level is Level 3, or the leakage event status result is "occurred," the overall operating status level is determined to be Level 3 (dangerous). The judgment results of the three parameters are substituted into the judgment rule for point-by-point judgment using millisecond-level timestamps to obtain the overall operating status level of the target compensator at each timestamp, forming full-time overall operating status level data.

[0067] The system collects comprehensive axial displacement data, torque data, and leakage event judgment results of the target compensator throughout the entire time series, along with corresponding axial displacement status level data, torque status level data, and overall operational status level data. All data is integrated in a time series using millisecond-level timestamps as the core index. Key statistical information for each status level is extracted, including the duration of Level 1 (normal), Level 2 (warning), and Level 3 (danger). If a leakage event has not occurred, details such as the occurrence time, duration, and distribution of abnormal points are added. The integrated time series data and key statistical information are arranged in a standardized format, which includes four sections: data overview, time series monitoring details, status level statistics, and abnormal event details. All data within each section retains complete correlation information and accurate numerical records, ultimately forming a structured and standardized full-parameter monitoring report for the target compensator.

[0068] The beneficial effects include: accurately determining the axial displacement status level by setting fixed numerical thresholds; determining the torque status level based on preset safety threshold ranges and clear calculation methods; ensuring that the status assessment of a single parameter has a unified standard and high reproducibility; establishing clear multi-parameter comprehensive judgment rules; and combining axial and torque status levels with leakage event results to derive the overall operating status level. This achieves a comprehensive assessment of the compensator's health status, avoiding the limitations of single-parameter assessment. Finally, all monitoring data, status levels, and statistical information are collected in a standardized format to generate a full-parameter monitoring report. The report covers raw monitoring data, grading judgment results, anomaly details, and statistical information. The data correlation is clear, the content is complete and comprehensive, and it can intuitively reflect the full-parameter operating status and health status of the target compensator, providing accurate and comprehensive data analysis basis for daily operation and maintenance, fault diagnosis, and risk warning of the equipment.

[0069] like Figure 2The diagram shown is a functional block diagram of a full-parameter monitoring system for an intelligent corrugated compensator provided in an embodiment of the present invention.

[0070] The intelligent corrugated compensator full-parameter monitoring system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the intelligent corrugated compensator full-parameter monitoring system 100 may include a multi-parameter synchronous measurement module 101, an axial displacement unidirectional fusion module 102, a discrete radial displacement field construction module 103, a torsional inversion calculation module 104, a leakage event identification module 105, and a full-parameter comprehensive evaluation module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0071] In this embodiment, the functions of each module / unit are as follows: The multi-parameter synchronous measurement module 101 is used to perform multi-parameter synchronous measurement on the target compensator to obtain the axial extension displacement data, radial offset displacement data and temperature and humidity time series data of the target compensator. The axial displacement unidirectional fusion module 102 is used to fuse the axial extension displacement data in the same direction to obtain the axial comprehensive displacement data of the target compensator. The discrete radial displacement field construction module 103 is used to construct the discrete radial displacement field of the target compensator based on the spatial orientation of the radial offset displacement data. The torque inversion solution module 104 is used to perform spatial symmetry decomposition on the discrete radial displacement field, and based on the decomposed symmetric and antisymmetric components, to perform distributed force inversion on the torque currently borne by the target compensator, and obtain the torque data of the target compensator. The leakage event identification module 105 is used to determine that a leakage event has occurred in the target compensator when the temperature and humidity time series data exceeds the preset temperature and humidity baseline. The comprehensive evaluation module 106 is used to conduct a comprehensive health status evaluation of the axial displacement data, the torque data, and the leakage event, and generate a comprehensive parameter monitoring report of the target compensator.

[0072] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0073] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0074] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0075] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0076] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for full parameter monitoring of an intelligent corrugated compensator, characterized in that, The method includes: Multi-parameter synchronous measurement was performed on the target compensator to obtain the axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator. The axial expansion and contraction displacement data are fused in the same direction to obtain the comprehensive axial displacement data of the target compensator, including: Spatial component analysis is performed on the axial expansion displacement data to obtain the multi-directional axial sub-displacements of the target compensator; Based on the directional characteristics of the multi-directional axial sub-displacements, abnormal sub-displacements in the multi-directional axial sub-displacements are eliminated to obtain the effective axial sub-displacements of the target compensator. Based on the deformation contribution of the effective axial sub-displacement, the preliminary axial displacement fusion data of the target compensator is calculated; ； In the formula, This is the initial axial displacement fusion data for the target compensator. This refers to the orientation index of the multi-directional axial sub-displacement. For the first The effective axial sub-displacements in each orientation For the first The contribution weighting coefficients corresponding to the effective sub-displacements of the axial direction in each orientation. The arithmetic mean of the effective sub-displacements along the axis. The standard deviation of the effective sub-displacement along the axis. This is a preset abnormality suppression factor; The preliminary axial displacement fusion data is subjected to time-series smoothing filtering to obtain the comprehensive axial displacement data of the target compensator. Based on the spatial orientation of the radial offset displacement data, a discrete radial displacement field of the target compensator is constructed, including: The radial offset displacement data is analyzed by measurement orientation to obtain the radial displacement metadata of the radial offset displacement data; Based on the spatial label of the radial offset displacement data, the radial displacement metadata is divided into a preset circumferential spatial partition; Spatial consistency verification is performed on the radial displacement metadata within the same circumferential spatial partition, and spatial feature extraction is performed on the valid radial displacement metadata that passes the verification to obtain the radial deformation level data of the circumferential spatial partition. Using the center azimuth angle of the radial deformation horizontal data as the index and the displacement value of the radial deformation horizontal data as the element, an initial spatial distribution mapping table of the target compensator is constructed. The initial spatial distribution mapping table is reconstructed using field data to obtain the discrete radial displacement field of the target compensator, including: Obtain the center azimuth value corresponding to the azimuth index item in the initial spatial distribution mapping table, and generate the spatial index sequence of the initial spatial distribution mapping table; Based on the spatial index sequence, data retrieval is performed on the initial spatial distribution mapping table to obtain the corresponding feature displacement data of the spatial index sequence; Based on the circumferential adjacency relationship of the center azimuth angle in the spatial index sequence, spatial adjacency topology is constructed for the corresponding feature displacement data to obtain the spatial displacement metadata structure of the corresponding feature displacement data. The spatial displacement metadata structure is encapsulated with attribute associations to obtain the discrete radial displacement field of the target compensator; The discrete radial displacement field is spatially symmetrically decomposed, and based on the decomposed symmetric and antisymmetric components, the distributed force inversion is performed on the torsional load currently borne by the target compensator to obtain the torsional data of the target compensator, including: Perform circumferential angle correlation analysis on the radial displacement data in the discrete radial displacement field to obtain the angle-displacement correspondence data of the radial displacement data. According to the circumferential angle order of the angle-displacement corresponding data, the angle-displacement corresponding data are arranged into a radial displacement distribution sequence that is continuously distributed along the circumference. The radial displacement distribution sequence is decomposed by mirror symmetry to obtain the symmetric components of the discrete radial displacement field; The radial displacement distribution sequence is subjected to odd-symmetric separation to obtain the antisymmetric component of the discrete radial displacement field. The antisymmetric component is subjected to circumferential positive and negative alternating analysis to obtain the antisymmetric phase characteristics of the antisymmetric component. Amplitude analysis is performed on the symmetric component, and the analyzed circumferential average amplitude and circumferential fluctuation amplitude are used as background features of the symmetric component. Based on the background features, the antisymmetric feature amplitude of the antisymmetric component is modified by interference suppression to obtain the modified antisymmetric feature amplitude of the antisymmetric component. The modified antisymmetric feature amplitude is matched with the preset torque-antisymmetric feature mapping table to determine the torque value corresponding to the modified antisymmetric feature amplitude. Based on the antisymmetric phase characteristics, the spatial orientation of the torque action of the target compensator is determined, and a torque direction identifier corresponding to the torque value is generated. The torque value and the torque direction identifier are used as the torque data of the target compensator. When the temperature and humidity time-series data exceeds the preset temperature and humidity baseline, it is determined that a leakage event has occurred in the target compensator; A comprehensive health status assessment is performed on the axial displacement data, the torque data, and the leakage event to generate a full-parameter monitoring report for the target compensator.

2. The method for full parameter monitoring of an intelligent corrugated compensator as described in claim 1, characterized in that, The method of performing multi-parameter synchronous measurement on the target compensator to obtain axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator includes: At the same time reference, a trigger signal is sent to the data acquisition channel arranged on the target compensator to acquire the original axial displacement signal, the original radial displacement signal and the original temperature and humidity signal of the target compensator. The original axial displacement signal, the original radial displacement signal, and the original temperature and humidity signal are conditioned to obtain the simulated axial displacement signal, the simulated radial displacement signal, and the simulated temperature and humidity signal of the target compensator. The axial displacement simulation signal, the radial displacement simulation signal, and the temperature and humidity simulation signal are converted from analog to digital to obtain the axial extension displacement data, radial offset displacement data, and temperature and humidity time series data of the target compensator.

3. The method for full parameter monitoring of an intelligent corrugated compensator as described in claim 1, characterized in that, The step of determining a leakage event in the target compensator when the temperature and humidity time-series data exceeds a preset temperature and humidity baseline includes: Statistical regression processing was performed on the ambient temperature and humidity data of the target compensator under leak-free operating conditions to obtain the temperature and humidity baseline of the target compensator. The temperature and humidity time series data are compared with the temperature baseline range and humidity baseline range of the temperature and humidity baseline, and abnormal points that exceed the temperature baseline range and humidity baseline range are marked. The extent and duration of the exceedance of the abnormal points are recorded. When the number of abnormal points reaches a preset spatial redundancy threshold, the out-of-limit directional features of the abnormal points are extracted, and the spatial consistency result of the abnormal points is determined. The change rate analysis is performed on the temperature and humidity time series data at the abnormal points to obtain the time series abrupt change information of the temperature and humidity time series data; When the spatial consistency result indicates that the abnormal points exhibit the same over-limit direction and the temporal mutation information indicates a synchronous rapid mutation in temperature and humidity, a multi-factor fusion judgment is performed on the over-limit amplitude, the over-limit duration, and the spatial distribution location of the abnormal points to obtain the judgment result of the target compensator leaking.

4. The method for full parameter monitoring of an intelligent corrugated compensator as described in claim 1, characterized in that, The comprehensive health status assessment of the axial displacement data, torque data, and leakage events is performed to generate a full-parameter monitoring report for the target compensator, including: The severity of deformation is determined by analyzing the comprehensive axial displacement data to obtain the axial displacement state level of the target compensator. Based on the relative deviation of the torque data within the preset torque safety threshold range, the torque state level of the target compensator is determined. The overall operating status level of the target compensator is obtained by comprehensively analyzing the axial displacement status level, the torque status level, and the status results of the leakage event. The axial displacement data, torque data, leakage events, axial displacement status level, torque status level, status results, and overall operating status level are integrated into a full-parameter monitoring report for the target compensator.

5. A full-parameter monitoring system for an intelligent corrugated compensator, characterized in that, The system for implementing the full parameter monitoring method of an intelligent corrugated compensator as described in claim 1 includes: The multi-parameter synchronous measurement module is used to perform multi-parameter synchronous measurement on the target compensator to obtain the axial extension displacement data, radial offset displacement data and temperature and humidity time series data of the target compensator. An axial displacement homogeneous fusion module is used to fuse the axial extension displacement data in the same direction to obtain the comprehensive axial displacement data of the target compensator. The discrete radial displacement field construction module is used to construct the discrete radial displacement field of the target compensator based on the spatial orientation of the radial offset displacement data. The torque inversion solution module is used to perform spatial symmetry decomposition on the discrete radial displacement field, and based on the decomposed symmetric and antisymmetric components, to perform distributed force inversion on the torque currently borne by the target compensator, and obtain the torque data of the target compensator. The leakage event identification module is used to determine that a leakage event has occurred in the target compensator when the temperature and humidity time series data exceeds the preset temperature and humidity baseline; The comprehensive parameter evaluation module is used to conduct a comprehensive health status evaluation of the axial displacement data, the torque data, and the leakage event, and generate a comprehensive parameter monitoring report of the target compensator.

Images