Stable field monitoring data analysis method based on laser tracker
By using a stable field monitoring data analysis method based on a laser tracker, the problem of monitoring minute displacements in long-cycle testing of high-precision products has been solved, enabling more accurate displacement change analysis and feedback, and making it suitable for various testing environments.
Patent Information
- Application Number
- CN202510924671.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-04
- Publication Date
- 2025-12-02
AI Technical Summary
Existing technologies are insufficient for effectively monitoring and analyzing minute displacement changes during long-cycle testing of high-precision products, especially in aerospace product testing, where traditional methods suffer from problems such as insufficient fitting, high timeliness, high equipment requirements, or long calculation times.
A steady-field monitoring data analysis method based on a laser tracker is adopted. Data is collected by a laser tracker, and temperature stability is determined by a temperature sensor. A time series or temperature parameter segmentation strategy is used to perform data cleaning and uncertainty analysis. The Hodrick-Prescott filter is used to decompose trend and periodic components to achieve accurate monitoring of multi-dimensional displacement changes.
It achieves more accurate micro-displacement monitoring, is suitable for high-precision scenarios, provides a better data foundation and feedback, and expands application scenarios to a variety of testing environments, including microwave anechoic chambers, vacuum tanks, and mechanical motion and variable temperature performance testing.
Smart Images
Figure CN121048501A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of precision measurement technology, and in particular to a method, device, and storage medium for analyzing stable field monitoring data based on a laser tracker. Background Technology
[0002] Early research on displacement monitoring in stable scenarios primarily focused on the fields of construction and geological engineering. However, with the development of technologies in aerospace, air exploration, and other fields, the demand for precision in product manufacturing and testing is increasing, as is the requirement for stability during long-term product testing. This has led to a further need for stability monitoring of high-precision products during long-term testing. Particularly in long-term radiation performance testing of aerospace detectors conducted in large microwave anechoic chambers, fluctuations in factors such as vibration and temperature over time can cause potential minute displacements in the product, further affecting the test results. Effective monitoring and analysis methods are needed to obtain data on the changes in these minute displacements over time or in relation to related factors. This data is crucial for both controlling these factors and building data foundations for testing and product performance compensation models based on these changes. Therefore, displacement monitoring and analysis in stable scenarios has become an important development direction in this emerging field.
[0003] Current displacement monitoring technologies are derived from the fields of building and geological engineering, and their main methods include: Laser ranging fitting analysis involves fitting the displacement curve based on single-point laser ranging results through control point sampling. However, this method suffers from underfitting or overfitting when dealing with large amounts of data, thus requiring very high precision in control point selection for high-precision monitoring. Sensor array analysis involves deploying acceleration and displacement sensors at different locations and using Kalman filtering to perform real-time estimation. However, this method is primarily for predictive evaluation in process monitoring and requires feedback, demanding high timeliness. It is often used for health monitoring in building and geological engineering, but often falls short in achieving micron- and sub-millimeter-level precision in aerospace products. Digital image analysis uses digital imaging technology to monitor pre-positioned targets or monitoring systems. The target's characteristics are analyzed by estimating displacement based on its changes relative to a reference point. However, this method requires a specific target, i.e., a pre-calibrated reference or multi-target pre-calibration for identification. It is limited by pixel size and is often only applicable to relatively significant displacement changes. The laser beam analysis method is an analysis method that calculates small displacements based on the tiny pixel offsets of the laser beam center in the image. This method can further subdivide pixels to improve accuracy by analyzing the laser beam center. However, it requires processing the subdivided data, which takes time. The sampling and calculation time is often relatively long, and it requires customized equipment and processing systems. Its application scenarios are often used for high-precision systems for timed sampling, rather than for long-term stable monitoring.
[0004] Against this backdrop, this invention provides a method for analyzing stable field monitoring data based on a laser tracker. This method segments and cleans existing displacement monitoring data under various influencing factors and incorporates uncertainty analysis of the measurement system itself, thereby more effectively extracting its expectation and variance curves. Furthermore, for overall data evaluation after cleaning, the Hodrick-Prescott filtering method is introduced to decompose periodic and trend factors in displacement changes, ultimately obtaining accurate multi-dimensional displacement change regions in real-world scenarios. This method can be applied to long-term system testing and monitoring of large aerospace products and their testing equipment, and is adaptable to the monitoring, analysis, and calculation of various measurement and sensing systems. Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for analyzing stable field monitoring data based on a laser tracker, comprising the following steps: S1: The laser tracker collects the absolute position coordinates of the monitoring points at fixed time intervals to obtain the raw dataset; S2: Determine the temperature stability based on the temperature data obtained from the temperature sensor. If the temperature is stable, use the time-series segmentation strategy; if the temperature is unstable, use the temperature parameter segmentation strategy. Divide the original dataset into several sub-datasets at equal time intervals according to the time-series segmentation strategy or the temperature parameter segmentation strategy. S3: Perform data cleaning and uncertainty analysis on the subset of data to obtain cleaned data and subdivided data; S4: Calculate and analyze the expected vector and variance vector of the subdivided data, reverse the reorganization of the cleaned data according to the original segmentation strategy, and decompose it into trend components and periodic components through Hodrick-Prescott filtering.
[0006] Preferably, the original dataset is divided into several sub-datasets at equal time intervals according to the time-series segmentation strategy, including: Set the equal time interval to The original dataset is coarsely divided into several sub-datasets according to the time-series segmentation strategy, wherein the last sub-dataset is allowed to be non-equal length data. Number of segments satisfy ,in Total duration The interval is used for segmentation.
[0007] Preferably, the original dataset is divided into several sub-datasets according to the temperature parameter segmentation strategy, including: Set the temperature parameter range to ={ , The original dataset is coarsely divided into several sub-datasets based on the temperature parameter range over time. Each sub-dataset is allowed to contain data of varying lengths. The number of sub-datasets is [number missing]. .
[0008] Preferably, in step S3, data cleaning and uncertainty analysis are performed on the subset of data to obtain cleaned data and subdivided data, including: S31: The isolated forest algorithm is used to clean the outliers in the subset of data to obtain cleaned data; S32: The cleaning data is used to construct an uncertainty ellipsoid model based on the characteristics of the laser tracker; S33: Determine whether iterative subdivision is needed based on the uncertainty ellipsoid model. If subdivision is needed, obtain the subdivided data.
[0009] Preferably, in step S31, the isolated forest algorithm is used to clean the outlier data of the subset of data to obtain cleaned data, which further includes: S311: Input the data of the aforementioned subset into the isolated forest model, set the node depth, and generate isolated trees through recursive random partitioning; S312: Calculate the anomaly score s(x) for each data point based on the isolation tree: in For data points Path length in an isolated tree Where H is the subsample size, and H() is the harmonic number. for The average of all isolated trees; S313: If the abnormal score is greater than the set threshold, the data point is determined to be an abnormal value and removed to obtain the cleaned data.
[0010] Preferably, in step S32, the cleaning data is used to construct an uncertainty ellipsoid model based on the characteristics of the laser tracker, further including: S321: The cleaning data is used to calculate the covariance matrix based on the anisotropic error characteristics of the laser tracker; S322: Based on the covariance matrix, establish a vector field coordinate system with the shortest axis of the ellipsoid as the Z-axis and the other two axes as the X-axis and Y-axis, respectively, and transform the cleaned data into this coordinate system to construct an uncertainty ellipsoid model.
[0011] Preferably, in step S33, determining whether iterative subdivision is needed based on the uncertainty ellipsoid model, and if subdivision is needed, obtaining subdivided data, further includes: S331: Extract the eigenvalues of the three axes of the uncertainty ellipsoid model; S332: If at least one axis exceeds the uncertainty model threshold and meets the set requirements, then locate the axis with the largest deviation, and split the original dataset in the time dimension. Then, recursively perform steps including data cleaning, constructing an uncertainty ellipsoid model, and deviation determination on the newly added subset. S333: Subdivision stops when all group uncertainty ellipsoidal models meet the requirements or the amount of data in a single group meets the requirements.
[0012] Preferably, in step S4, the Hodrick-Prescott filter decomposes the components into trend and periodic components, including: The reconstructed data is projected onto the target direction to obtain the projected data. The Hodrick-Prescott filter decomposes the projected data into the trend component and the periodic component according to the objective function; The objective function is as follows: in, , for the first Monitoring data in the group, This represents the number of time series data in this group. It is a second-order difference operator. For smoothing coefficients, For the trend component, the smoothing coefficient is calculated as follows: in, This represents the sampling frequency coefficient.
[0013] Based on the same concept, the present invention also provides a computer device, characterized in that it includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor causes the processor to perform the steps of the stable field monitoring data analysis method based on a laser tracker as described in any one of the embodiments.
[0014] Based on the same concept, the present invention also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the stable field monitoring data analysis method based on a laser tracker as described in the embodiments.
[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention determines temperature stability based on temperature data obtained from temperature sensors. If the temperature is stable, a time-series segmentation strategy is used; if the temperature is unstable, a temperature parameter segmentation strategy is used. The original dataset is divided into several sub-data sets at equal time intervals. Data cleaning and uncertainty analysis are performed on the sub-data sets to obtain cleaned data and subdivided data, thus enabling the analysis of monitoring data. Compared with direct analysis of monitoring data, this invention further considers the uncertainty of the measurement system, resulting in more accurate results and better characterizing displacement changes during the test process. This provides the possibility for equipment with limited precision to be applied to monitoring scenarios with higher precision. This invention further considers the spatial coordinate system of the displacement vector field by establishing an uncertainty ellipsoid model, and decomposes and analyzes the trend of displacement change in multiple degrees of freedom. It can better describe the trend of micro displacement in three-dimensional space, and provides a better data foundation for the practical application of monitoring and analysis data in feedback. Compared with other analysis methods, it achieves this through active data segmentation. This invention expands the data analysis factors from time series and temperature series to include excitation series and mechanical motion parameters, thus broadening its application scenarios.
[0016] This invention is applicable to long-term radiation performance testing of products in microwave anechoic chambers. It can also be extended to monitoring other stable testing environments by changing the factors in data analysis, such as motion testing of mechanisms inside vacuum tanks, variable temperature performance testing, microwave calibration testing, and other scenarios. Attached Figure Description
[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention.
[0018] Figure 1 This is a flowchart of the stable field monitoring data analysis method based on a laser tracker according to the present invention; Figure 2 This is another flowchart of the stable field monitoring data analysis method based on a laser tracker according to the present invention; Figure 3 This is a schematic diagram of the device monitoring and a point cloud diagram of the measured data of the monitoring points of the present invention; Figure 4 This is a schematic diagram of the coarse segmentation of the monitoring point measurement data of the present invention; Figure 5 This invention compares grouped data before and after cleaning. Figure 6 This invention is an uncertainty model based on grouped data. Figure 7This is a schematic diagram of the vector field coordinate system established by the present invention based on the uncertainty model of grouped data; Figure 8 This is the expected analysis result of the monitoring point measurement data of this invention; Figure 9 This is the result of filtering analysis of the monitoring point measurement data after cleaning and integration according to the present invention; Figure 10 This is the temperature change curve of the monitoring point in this embodiment of the invention. Detailed Implementation
[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. Obviously, the described embodiments are only some, not all, of the embodiments described in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without creative effort are within the scope of protection of this application.
[0020] Those skilled in the art will understand that, unless otherwise stated, the singular forms “a” and “an” used herein, and “the”, may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.
[0021] First Embodiment This embodiment focuses on the radiation performance testing of a large microwave product. It requires displacement monitoring and analysis throughout the entire testing process. The selected monitoring equipment includes a Leica laser tracker and a 1.5-foot target sphere. The testing site is a large microwave anechoic chamber.
[0022] The product under test and the laser tracker are fixed in the microwave anechoic chamber. Target balls are arranged on the test mounting surface of the product under test as monitoring points, and temperature sensors are deployed simultaneously to monitor the ambient temperature field, so as to realize the deployment of monitoring devices in an electromagnetically shielded environment.
[0023] Please see Figure 1 and Figure 2 As shown in the figure, this embodiment provides a method for analyzing stable field monitoring data based on a laser tracker, which includes the following steps: S1: The laser tracker collects the absolute position coordinates of the monitoring points at fixed time intervals to obtain the raw dataset. Specifically, in this embodiment, during the test, the laser tracker collects data at 1.5-second time intervals, acquiring the absolute position coordinates of the monitoring points. The obtained measurement data is as follows: Figure 3 As shown in the figure, the data indicates that the monitored object has a certain slight displacement. A total of 29,410 data points were monitored, and the monitoring time was 12 hours and 5 minutes.
[0024] S2: Determine temperature stability based on temperature data obtained from the temperature sensor. If the temperature is stable, use a time-series segmentation strategy; if the temperature is unstable, use a temperature parameter segmentation strategy. Divide the original dataset into several sub-datasets at equal time intervals using either the time-series segmentation strategy or the temperature parameter segmentation strategy. Specifically, in this embodiment, based on the changes in the main influencing factors of the absolute position change of the monitored object points, the data is segmented... Divided into a series of datasets ,in The number of data sets is specified. The collected data includes time series t, temperature series T, excitation series F, and mechanical motion parameters.
[0025] Preferably, in step S2, the original dataset is divided into several sub-datasets at equal time intervals according to a time-series segmentation strategy, including: Set the time interval to =60 minutes, the original dataset is coarsely segmented into several subsets, with the last subset allowed to contain data of varying lengths. Specifically, in this embodiment, the original dataset is first coarsely segmented. Based on synchronously monitored temperature data, its temperature field is relatively stable; therefore, only time series is selected as a factor for segmentation and subsequent analysis. Measurement sampling is performed at equal intervals. Please refer to... Figure 4 As shown, the coarse segmentation uses equal time intervals, i.e., 60 minutes per group, for a total of 13 groups, each with 2400 sampling data points, and the last group has 610 points. ; Number of segments satisfy ,in Total duration The interval is used for segmentation.
[0026] Please see Figure 10 As shown, the original dataset is divided into several subsets based on a temperature parameter splitting strategy, including: Set the temperature parameter range as follows: ={ , The original dataset is coarsely divided into several subsets based on temperature parameter ranges over time. Each subset is allowed to contain data of varying lengths. The number of subsets is [number missing]. Specifically, in this embodiment, if temperature change is the main factor affecting the absolute position change of the monitoring point, a temperature parameter segmentation strategy is used to set the equal temperature change interval as... ={22.5, 23.2, 23.5, 23.2, 23.5, 23.7, 23.9, 24.1, 24.4, 24.2}, that is, according to the time series, the original dataset is coarsely divided into several sub-data sets based on the temperature values, where each data set is of non-equal length, and the number of sub-sets k=10.
[0027] S3: Perform data cleaning and uncertainty analysis on the subset to obtain cleaned data and segmented data.
[0028] Preferably, in step S3, data cleaning and uncertainty analysis are performed on the subset of data to obtain cleaned data and subdivided data, including: S31: The isolated forest algorithm is used to clean the outlier values in the subset of data to obtain cleaned data. Specifically, in this embodiment, please refer to the first set of cleaning results. Figure 5 As shown, by cleaning outliers with large deviations, the data retention rate was 98.54%. S32: An uncertainty ellipsoid model is constructed based on the characteristics of the laser tracker using the cleaned data. Specifically, in this embodiment, please refer to the modeling results of the first group. Figure 6 As shown; S33: Determine whether iterative subdivision is needed based on the uncertainty ellipsoid model. If subdivision is needed, subdivided data is obtained. Specifically, in this embodiment, the uncertainty extreme value range of the uncertainty model established based on each group of data in the vector field coordinate system is compared to determine whether to perform fine subdivision. Since the ellipsoid distribution of each group is consistent in this embodiment, no further fine subdivision is performed.
[0029] Preferably, in step S31, the isolated forest algorithm is used to clean the outlier data of the subset of data to obtain cleaned data, which further includes: S311: Input the data from the subset into the isolated forest model and set the node depth. Generate isolated trees through recursive random partitioning. Specifically, in this embodiment, the node depth parameter is set to 7. S312: Calculate the outlier score s(x) for each data point based on the isolation tree: in For data points Path length in an isolated tree Where H is the subsample size, and H() is the harmonic number. for The average of all isolated trees; S313: If the abnormal score is greater than the set threshold, the data point is determined to be an outlier and removed to obtain cleaned data.
[0030] Preferably, in step S42, the cleaning data is used to construct an uncertainty ellipsoid model based on the characteristics of the laser tracker, further including: S321: The covariance matrix is calculated based on the anisotropic error characteristics of the laser tracker after cleaning the data; S322: Based on the covariance matrix, establish a vector field coordinate system with the shortest axis of the ellipsoid as the Z-axis and the other two axes as the X and Y axes. Transform the cleaned data into this coordinate system and construct an uncertainty ellipsoid model. Specifically, in this embodiment, please refer to... Figure 7 As shown, the established model conforms to the ellipsoidal distribution of the uncertainty anisotropy of the laser tracker, and the ellipsoidal distribution of each group is consistent. The shortest axis of the ellipsoid is taken as the Z-axis, and the other two axes are the X and Y axes, respectively. The vector field coordinate system of the monitoring is established, and the monitoring data is transformed into this coordinate system.
[0031] Preferably, in step S43, it is determined whether iterative subdivision is needed based on the uncertainty ellipsoid model. If subdivision is needed, subdivided data is obtained, further including: S331: Extract eigenvalues of the three axes of the uncertainty ellipsoid model; S332: If at least one axis exceeds the threshold setting requirement of the uncertainty model, locate the axis with the largest deviation, split the original dataset in the time dimension, and recursively perform steps including data cleaning, constructing the uncertainty ellipsoid model, and deviation determination on the newly added subset; S333: Subdivision stops when all group uncertainty ellipsoidal models meet the requirements or the amount of data in a single group meets the requirements.
[0032] Preferably, the expected vector and variance vector of the subdivided data are calculated, and the cleaned data is reorganized according to the time series, that is, the global sequence is reorganized according to the timestamp order of the subdivided data and the cleaned data.
[0033] S5: Calculate the expected vector and variance vector of the subdivided data, reorganize the cleaned data according to the time series, and decompose it into trend component c and periodic component g through Hodrick-Prescott filtering. Specifically, in this embodiment, calculate and group the data... The expected vector of the uncertainty model in spatial distribution With variance vector And the components of the expectation vector and variance vector in each direction of the vector field coordinate system. and For analysis of the expected value and variance curves to show how factors change with data segmentation, please refer to [link to relevant documentation]. Figure 8 As shown, this represents the uncertainty ellipsoidal models for each group. curve.
[0034] Preferably, the Hodrick-Prescott filter decomposes the component into a trend component c and a periodic component g, including: The reconstructed data is projected onto the target direction to obtain the projected data; The Hodrick-Prescott filter decomposes the projected data into a trend component c and a periodic component g according to the objective function. Specifically, in this embodiment, the cleaned data are recombined into a sequence based on the segmentation parameters, and the Hodrick-Prescott filter is performed in the direction of the selected change vector based on the initial point position to extract the data decomposed into a periodic component g and a trend component c. The objective function is as follows: in, , for the first Monitoring data in the group, This represents the number of time series data in this group. It is a second-order difference operator. For smoothing coefficients, For the trend component, the smoothing coefficient is calculated as follows: in, For the sampling frequency coefficient, specifically, in this embodiment, the data integration filtering analysis involves recombining the cleaned data into a sequence based on the segmentation parameters, and using the initial point position as a reference, performing Hodrick-Prescott filtering in the direction of the selected analysis change vector.
[0035] The fluctuations of the measurement and monitoring system over a short period are analyzed using the periodic component g, while the changes and trends of the product over the entire time series of influencing factors are analyzed using the trend component c. This embodiment primarily analyzes the minute displacements mainly along the Z-axis of the vector field coordinate system; therefore, an overall combined filtering evaluation is performed. Please refer to the curves for the periodic component g and the trend analysis c. Figure 9 As shown in the figure, the trend analysis curves indicate that the overall product showed an upward trend during the test, which stabilized after reaching a critical point. This is consistent with the expected curves of each group. The periodic component curves show that the monitoring system remained stable overall during the test.
[0036] Example 2 Based on the same concept, this embodiment also provides a computer device, including a memory and a processor. The memory stores computer-readable instructions, which, when executed by the processor, cause the processor to perform the steps of a stable field monitoring data analysis method based on a laser tracker in the first embodiment of the present invention.
[0037] This embodiment also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of a stable field monitoring data analysis method based on a laser tracker according to an embodiment of the present invention.
[0038] It is understood that, for the aforementioned method for analyzing stable field monitoring data based on a laser tracker, if all components are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer server or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.
[0039] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable storage medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.
[0040] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.
Claims
1. A method for analyzing stable field monitoring data based on a laser tracker, characterized in that, Includes the following steps: S1: The laser tracker collects the absolute position coordinates of the monitoring points at fixed time intervals to obtain the raw dataset; S2: Determine the temperature stability based on the temperature data obtained from the temperature sensor. If the temperature is stable, use the time-series segmentation strategy; if the temperature is unstable, use the temperature parameter segmentation strategy. Divide the original dataset into several sub-datasets at equal time intervals according to the time-series segmentation strategy or the temperature parameter segmentation strategy. S3: Perform data cleaning and uncertainty analysis on the subset of data to obtain cleaned data and subdivided data; S4: Calculate and analyze the expected vector and variance vector of the subdivided data, reverse the reorganization of the cleaned data according to the original segmentation strategy, and decompose it into trend components and periodic components through Hodrick-Prescott filtering.
2. The method for analyzing stable field monitoring data based on a laser tracker according to claim 1, characterized in that, The original dataset is divided into several sub-datasets at equal time intervals according to the time-series segmentation strategy, including: Set the equal time interval to The original dataset is coarsely divided into several sub-datasets according to the time-series segmentation strategy, wherein the last sub-dataset is allowed to be non-equal length data. Number of segments satisfy ,in Total duration The interval is used for segmentation.
3. The method for analyzing stable field monitoring data based on a laser tracker according to claim 2, characterized in that, The original dataset is divided into several sub-datasets according to the temperature parameter segmentation strategy, including: Set the temperature parameter range to ={ , The original dataset is coarsely divided into several sub-datasets based on the temperature parameter range over time. Each sub-dataset is allowed to contain data of varying lengths. The number of sub-datasets is [number missing]. .
4. The method for analyzing stable field monitoring data based on a laser tracker according to claim 3, characterized in that, In step S3, data cleaning and uncertainty analysis are performed on the subset of data to obtain cleaned data and subdivided data, including: S31: The isolated forest algorithm is used to clean the outliers in the subset of data to obtain cleaned data; S32: The cleaning data is used to construct an uncertainty ellipsoid model based on the characteristics of the laser tracker; S33: Determine whether iterative subdivision is needed based on the uncertainty ellipsoid model. If subdivision is needed, obtain the subdivided data.
5. The method for analyzing stable field monitoring data based on a laser tracker according to claim 4, characterized in that, In step S31, the isolated forest algorithm is used to clean the outlier data of the subset of data to obtain cleaned data, which further includes: S311: Input the data of the aforementioned subset into the isolated forest model, set the node depth, and generate isolated trees through recursive random partitioning; S312: Calculate the anomaly score s(x) for each data point based on the isolation tree: in For data points Path length in an isolated tree Where H is the subsample size, and H() is the harmonic number. for The average of all isolated trees; S313: If the abnormal score is greater than the set threshold, the data point is determined to be an abnormal value and removed to obtain the cleaned data.
6. The method for analyzing stable field monitoring data based on a laser tracker according to claim 5, characterized in that, In step S32, the cleaning data is used to construct an uncertainty ellipsoid model based on the characteristics of the laser tracker, which further includes: S321: The cleaning data is used to calculate the covariance matrix based on the anisotropic error characteristics of the laser tracker; S322: Based on the covariance matrix, establish a vector field coordinate system with the shortest axis of the ellipsoid as the Z-axis and the other two axes as the X-axis and Y-axis, respectively, and transform the cleaned data into this coordinate system to construct an uncertainty ellipsoid model.
7. The method for analyzing stable field monitoring data based on a laser tracker according to claim 6, characterized in that, In step S33, it is determined whether iterative subdivision is needed based on the uncertainty ellipsoid model. If subdivision is needed, subdivided data is obtained, further including: S331: Extract the eigenvalues of the three axes of the uncertainty ellipsoid model; S332: If at least one axis exceeds the threshold setting requirement of the uncertainty model, the axis with the largest positioning deviation is located. The original dataset is bisected in the time dimension, and the steps including data cleaning, construction of uncertainty ellipsoid model, and deviation determination are recursively performed on the newly added subset. S333: Subdivision stops when all group uncertainty ellipsoidal models meet the requirements or the amount of data in a single group meets the requirements.
8. The method for analyzing stable field monitoring data based on a laser tracker according to claim 7, characterized in that, In step S4, the Hodrick-Prescott filter decomposes the components into trend and periodic components, including: The reconstructed data is projected onto the target direction to obtain the projected data. The Hodrick-Prescott filter decomposes the projected data into the trend component and the periodic component according to the objective function; The objective function is as follows: in, , for the first Monitoring data in the group, This represents the number of time series data points in this group. It is a second-order difference operator. For smoothing coefficients, For the trend component, the smoothing coefficient is calculated as follows: in, This represents the sampling frequency coefficient.
9. A computer device, characterized in that, The method includes a memory and a processor, wherein the memory stores computer-readable instructions that, when executed by the processor, cause the processor to perform the steps of the stable field monitoring data analysis method based on a laser tracker as described in any one of claims 1 to 8.
10. A storage medium storing computer-readable instructions, characterized in that, When the computer-readable instructions are executed by one or more processors, the one or more processors perform the steps of the stable field monitoring data analysis method based on a laser tracker as described in any one of claims 1 to 8.