Vibration measurement error determination method and vibration error determination system using the same
The method uses error data selection rules and machine learning to identify and filter out measurement errors in vibration data, improving the reliability of predictive diagnosis by distinguishing between normal and erroneous data.
Patent Information
- Application Number
- JP2023576190
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2021-06-11
- Filing Date
- 2021-11-15
- Publication Date
- 2024-11-13
- Estimated Expiration
- 2041-11-15
AI Technical Summary
Vibration data containing measurement errors can lead to erroneous analysis results, reducing the reliability of predictive diagnosis in structures, particularly due to issues like equipment skill deficiencies, external signal interference, and sensor failures.
A method involving a combination of error data selection rules and machine learning algorithms to determine measurement errors in vibration data, utilizing a low frequency region amplitude threshold and principal component analysis to identify and filter out erroneous data.
Improves the accuracy of vibration data by distinguishing between normal and erroneous data, enhancing the reliability of predictive diagnosis.
Smart Images

Figure 0007583191000003 
Figure 0007583191000004 
Figure 0007583191000005
Abstract
Description
[Technical field]
[0001] The present invention relates to a vibration measurement error determination method and a vibration error discrimination system using the same, and more particularly to a vibration measurement error determination method for detecting whether data regarding a measurement error is included in vibration data and a vibration error discrimination system using the same. [Background technology]
[0002] For structures where vibrations are inevitable, the current state of the structure and the state of the structure in the near future can be predicted by measuring the vibrations generated in the structure. In particular, in nuclear power plants, mobile vibration measuring devices are used to measure the vibrations of facilities for facility management.
[0003] Recently, technology has been developed that uses machine learning to analyze vibration data generated by structures and understand their condition.
[0004] However, problems that occur during the process of measuring the vibration of a structure can result in collected vibration data containing measurement errors, which can ultimately lead to erroneous analysis results for the structure and reduce the reliability of predictive diagnosis.
[0005] For example, during the vibration measurement process, erroneous vibration measurement data may be collected due to the inexperience of the measurer in using the equipment, insufficient management of the measurement equipment, external signal interference, and the like.
[0006] Referring to FIG. 10, (a) and (b) are measurement data containing noise due to residual stress after an impact occurs to a sensor that detects vibration, (c) is measurement data containing shock waveform noise due to a problem such as sensor adhesion, and (d) is measurement data containing noise due to a sensor and / or cable defect.
[0007] In this way, vibration data containing measurement error data generated during the vibration measurement process reduces the reliability of diagnosis even when specialized machine learning is used to analyze the condition of a vibrating structure. Summary of the Invention [Problem to be solved by the invention]
[0008] An object of the present invention is to provide a vibration measurement error determination method for detecting whether data regarding a measurement error is included in vibration data, and a vibration error determination system using the same.
[0009] The object of the present invention is not limited to the above-mentioned objects, and other objects not mentioned will be clearly understood by those skilled in the art from the following description. [Means for solving the problem]
[0010] A vibration measurement error determination method according to an embodiment of the present invention for solving the above problem includes a vibration data acquisition step of measuring vibrations generated in a structure and acquiring vibration data, a first judgment step of determining whether the vibration data is due to a measurement error based on a preset error data selection rule, a second judgment step of determining whether the vibration data is due to a measurement error using a machine learning algorithm, and a final judgment step of determining whether the vibration data is due to a measurement error based on the results determined in the first judgment step and the second judgment step.
[0011] The error data selection rule determines whether the vibration data is due to a measurement error based on a difference between an amplitude value of a low frequency region extracted from the vibration data and a threshold value.
[0012] The low frequency range includes the range below 3 Hz.
[0013] The threshold value is 0.6 mm / s.
[0014] The machine learning algorithm extracts a plurality of sample data from the vibration data, extracts feature information for determining a measurement error from the extracted plurality of sample data, and determines whether the vibration data is due to a measurement error based on the extracted feature information.
[0015] The feature information is extracted from candidate feature information extracted from a plurality of the sample data through principal component analysis.
[0016] The machine learning algorithm extracts the feature information for each of the sample data, clusters the sample data based on the feature information, and determines whether the vibration data is due to a measurement error based on the clustering results.
[0017] To solve the above problem, a vibration error discrimination system according to an embodiment of the present invention includes a data acquisition unit that measures vibrations generated in a structure and acquires vibration data, a first judgment unit that determines whether the vibration data is due to a measurement error based on preset error data selection rules, a second judgment unit that determines whether the vibration data is due to a measurement error using a machine learning algorithm, and a final judgment unit that determines whether the vibration data is due to a measurement error based on the results determined in the first judgment step and the second judgment step.
[0018] The error data selection rule determines whether the vibration data is due to a measurement error based on a difference between an amplitude value of a low frequency region extracted from the vibration data and a threshold value.
[0019] The low frequency range includes the range below 3 Hz.
[0020] The threshold value is 0.6 mm / s.
[0021] The machine learning algorithm extracts a plurality of sample data from the vibration data, extracts feature information for determining a measurement error from the extracted plurality of sample data, and determines whether the vibration data is due to a measurement error based on the extracted feature information.
[0022] The feature information is extracted from candidate feature information extracted from a plurality of the sample data through principal component analysis.
[0023] The machine learning algorithm extracts the feature information for each of the sample data, clusters the sample data based on the feature information, and determines whether the vibration data is due to a measurement error based on the clustering results.
[0024] Further details of the invention are contained in the detailed description and the drawings. Effect of the Invention
[0025] According to the embodiment of the present invention, at least the following effects are obtained.
[0026] It can be determined whether the measured vibration data includes data for measurement errors.
[0027] In addition, erroneously measured vibration data can be selected to improve the soundness of the vibration data, which is the basic data for predictive diagnosis.
[0028] The effects of the present invention are not limited to the above-mentioned examples, and various other effects are included in the present specification. [Brief description of the drawings]
[0029] [Figure 1] 4 is a flow chart illustrating a method for determining a vibration measurement error according to an embodiment of the present invention. [Diagram 2]2 is a flow chart for explaining a first determination step in FIG. 1. [Diagram 3] 1 is a graph showing an example of normal vibration data. [Figure 4] 1 is a graph showing an example of vibration data including a measurement error. [Diagram 5] 11 is a graph showing expert judgment results for sample vibration data. [Figure 6] 2 is a flow chart for explaining a second determination step in FIG. 1. [Figure 7] 11 is a graph showing a clustering result obtained by judging sample vibration data through a second judging step. [Figure 8] 11 is a graph showing a clustering result obtained by judging sample vibration data through a second judging step. [Figure 9] 1 is a block diagram showing a vibration measurement error discrimination system according to an embodiment of the present invention; [Figure 10] 1 is a graph showing an example of vibration data including a measurement error. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0030] The advantages and features of the present invention, and the method of achieving the same, will become clear from the following detailed description of the embodiments in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, and may be embodied in various different forms, and the embodiments are provided merely to fully disclose the present invention and fully inform those skilled in the art of the scope of the invention, and the present invention is defined by the scope of the claims. The same reference characters refer to the same elements throughout the specification.
[0031] In addition, the embodiments described herein are described with reference to cross-sectional views and / or schematic diagrams that are ideal exemplary views of the present invention. Therefore, the shapes of the exemplary views may be modified depending on manufacturing techniques and / or tolerances. In addition, the components in each drawing shown in the present invention are illustrated slightly enlarged or reduced in size for convenience of explanation. The same reference numerals refer to the same components throughout the specification.
[0032] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described with reference to the drawings illustrating a vibration measurement error determination method and a vibration measurement error determination system using the same according to an embodiment of the present invention.
[0033] FIG. 1 is a flow chart illustrating a method for determining an error in vibration measurement according to an embodiment of the present invention.
[0034] Referring to FIG. 1, a vibration measurement error judgment method according to one embodiment of the present invention includes a vibration data acquisition step (S10), a first judgment step (S20), a second judgment step (S30), and a final judgment step (S40).
[0035] In the vibration data acquisition step (S10), vibrations occurring in the structure are measured to acquire vibration data. The vibration data may be acquired using a mobile vibration measuring device or from a database (DB) in which vibration data is stored.
[0036] The structure includes rotating equipment that rotates while operating. Vibrations occur in the structure as the rotating equipment operates, and the current states of the structure and the rotating equipment can be diagnosed by measuring the vibrations, and based on the diagnosis, a predictive diagnosis can be performed for the structure and the rotating equipment, i.e., a near-future state can be predicted and diagnosed. In particular, the vibration measurement error determination method according to an embodiment of the present invention can be used for equipment management of nuclear power plants by acquiring vibration data from various equipment used in nuclear power plants.
[0037] In the first determination step (S20), it can be determined whether the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error based on a preset error data selection rule, which will be described in more detail later.
[0038] In the second determination step (S30), it is determined whether the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error using a determination criterion different from the determination criterion in the first determination step (S20). For example, in the second determination step (S30), it may be determined whether the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error using a machine learning algorithm. This will be described in more detail later.
[0039] In the final judgment step (S40), a final judgment is made as to whether or not the vibration data acquired in the vibration data acquisition step (S10) includes data due to measurement errors, based on the results of the judgments made in the first judgment step (S20) and the second judgment step (S30).
[0040] For example, in the final judgment step (S40), it can be judged that there is no measurement error only for vibration data that has been judged to be free of any measurement error in the first judgment step (S20) and the second judgment step (S30).
[0041] Alternatively, in the final determination step (S40), the vibration data that is determined to have no measurement error in either the first determination step (S20) or the second determination step (S30) can be determined to have no measurement error.
[0042] FIG. 2 is a flow chart for explaining the first decision step in FIG.
[0043] Referring to FIG. 2, the first judgment step (S20) may include a time waveform data extraction step (S21), a fast Fourier transform step (S22), a vibration value extraction step in the low frequency domain (S23), a step of comparing the vibration value with a threshold value (S24), and a measurement error judgment step (S25).
[0044] In the time waveform data extraction step (S21), vibration data of a time waveform (Time Wave Form) is extracted from the vibration data acquired in the vibration data acquisition step (S10). If the vibration data acquired in the vibration data acquisition step (S10) is data of a time waveform (Time Wave Form), the time waveform data extraction step (S21) can be omitted.
[0045] In the Fast Fourier Transform step (S22), the vibration data of the time waveform is transformed into the frequency domain. In the Fast Fourier Transform step (S22), the vibration data of the time waveform can be sampled and then transformed into the frequency domain. In the process of transforming the vibration data of the time waveform into the frequency domain, FFT (Fast Fourier transform) can be used using a Rectangular Window that does not use a window function.
[0046] In the low-frequency region vibration value extraction step (S23), vibration values (amplitude values) for the low-frequency region are calculated and extracted from the vibration data converted into the frequency domain.
[0047] FIG. 3 is a graph showing an example of normal vibration data, and FIG. 4 is a graph showing an example of vibration data including a measurement error.
[0048] 3 and 4, it can be seen that the vibration data including the measurement error has a large amplitude in the low frequency region of 3 Hz or less compared to the normal vibration data, and it can be seen that the ski-slope phenomenon is observed in the low frequency band. Therefore, in the step of extracting vibration values in the low frequency region (S23), the low frequency region can be set to a frequency region of 3 Hz or less.
[0049] In the step (S24) of comparing the vibration value with a threshold value, the vibration value in the low frequency region extracted in the step (S23) of extracting vibration value in the low frequency region is compared with the threshold value.
[0050] In order to find a threshold value for determining measurement errors, the applicant measured vibration data at 1,811 points for 248 pieces of equipment at Hanbit-3 Nuclear Power Plant Units 5 and 6, and had experts discriminate between vibration data containing measurement errors for a total of 10,634 sample vibration data.
[0051] As a result, of the 10,634 sample vibration data, the experts determined that 10,175 were normal vibration data, 231 were vibration data containing measurement errors, and the judgement was withheld for 228.
[0052] FIG. 5 is a graph showing the expert judgment results for the sample vibration data.
[0053] Referring to FIG. 5, it can be seen that there is a large amount of normal vibration data in the low amplitude low frequency region (region of 0.006 in / s or less), normal vibration data overlaps with vibration data containing measurement errors in the region where the amplitude of the low frequency region is 0.007 to 0.026 in / s, and vibration data containing measurement errors appears in the region where the amplitude of the low frequency region is 0.013 in / s or more.
[0054] Based on FIG. 5, in order to set the amplitude threshold, the number of cases in which vibration data including measurement errors were judged to be normal vibration data and, conversely, the number of cases in which normal vibration data were judged to be vibration data including measurement errors were compared when the thresholds were set to 0.4 mm / s (= 0.0157 in / s), 0.5 mm / s (= 0.0197 in / s), 0.6 mm / s (= 0.2362 in / s), 0.7 mm / s (= 0.0276 in / s), and 0.8 mm / s (= 0.0315 in / s), respectively.
[0055] [Table 1]
[0056] When the threshold was set to 0.8 mm / s, normal vibration data was not judged as data containing measurement errors, but data containing measurement errors were often judged as normal vibration data. Conversely, when the threshold was set to 0.4 mm / s, the rate at which data containing measurement errors were judged as normal vibration data decreased, but the rate at which normal vibration data was judged as data containing measurement errors increased.
[0057] It can be seen that there is a trade-off between the ratio at which data including measurement errors are determined to be normal vibration data based on the threshold value and the ratio at which normal vibration data are determined to be data including measurement errors.
[0058] In this embodiment, in order to minimize the occurrence of determining that normal vibration data is data containing measurement errors and also to minimize overall erroneous determinations, the threshold value (T, see FIG. 5) was set to 0.6 mm / s.
[0059] However, depending on requirements such as the precision of the measurement error discrimination, the low-frequency amplitude threshold for selecting vibration data including measurement errors may be selected differently within a range of 0.4 to 0.8 mm / s, and if necessary, the threshold may be set to a value equal to or less than 0.4 mm / s or equal to or more than 0.8 mm / s.
[0060] In the step of comparing the vibration value with a threshold (S24), the vibration value in the low frequency region extracted in the step of extracting vibration values in the low frequency region (S23) is compared with a set threshold to determine whether the vibration value is below (or less than) the threshold or exceeds (or greater than) the threshold.
[0061] In the measurement error determination step (S25), it is determined whether the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error based on the difference between the vibration value determined in the step (S24) of comparing the vibration value with a threshold value and the threshold value.
[0062] That is, if the vibration value is equal to or less than the threshold value (or less than the threshold value), it is determined that the vibration data is normal, and if the vibration value is greater than or equal to the threshold value, it is determined that the vibration data includes data due to a measurement error.
[0063] As described above, in the first judgment step (S20), when judging whether the vibration data includes data due to a measurement error based on the error data selection rule, i.e., the difference between the vibration value in the low-frequency region and a threshold value, there are cases where normal vibration data is judged as data including a measurement error, or data including a measurement error is judged as normal vibration data, depending on the threshold value.
[0064] To compensate for this, the vibration measurement error determination method according to this embodiment includes a second determination step (S30).
[0065] FIG. 6 is a flow chart for explaining the second decision step in FIG.
[0066] In a second decision step (S30), a machine learning algorithm can be used to determine whether the vibration data includes data due to measurement errors.
[0067] To this end, referring to FIG. 6, the second judgment step (S30) may include a time waveform data extraction step (S31), a sample data extraction step (S32), a candidate feature information extraction step (S33), a principal component analysis step (S34), a clustering analysis step (S35), and a measurement error judgment step (S36).
[0068] In the time waveform data extraction step (S31), vibration data of a time waveform (Time Wave Form) is extracted from the vibration data acquired in the vibration data acquisition step (S10). If the vibration data acquired in the vibration data acquisition step (S10) is data of a time waveform (Time Wave Form), the time waveform data extraction step (S31) can be omitted.
[0069] In the sample data extraction step (S32), at least one sample data may be extracted from the vibration data of the time waveform. The number of sample data to be extracted may be selected in various ways depending on the embodiment, but it is preferable to extract a large number of sample data from the vibration data in order to ensure robustness of the machine learning algorithm for determining measurement errors from the vibration data. However, as the number of sample data increases, considerable resources are required for calculations for extracting feature information of the machine learning algorithm, so it is necessary to select an appropriate number of sample data. In consideration of this, the number of sample data is preferably several tens, more preferably about 20. The criteria for extracting sample data may be set in various ways depending on the embodiment.
[0070] In the candidate characteristic information extraction step (S33), candidate characteristic information is extracted from the sample data extracted in the sample data extraction step (S32).
[0071] The candidate feature information can be extracted from first statistical information calculated from the sample data extracted in the sample data extraction step (S32) and / or second statistical information extracted in the time waveform data extraction step (S31) or calculated from the time waveform vibration data acquired in the vibration data acquisition step (S10).
[0072] The first statistical information may include the RMS (Root Mean Square), mean, variance, and standard deviation of the sample data. When a plurality of sample data are extracted in the sample data extraction step (S12), the first statistical information may include the RMS, mean, variance, and standard deviation of each sample data.
[0073] The second statistical information may include the RMS, mean, variance, and standard deviation of the vibration data of the time waveform.
[0074] Normal vibration data exhibits a pattern of periodic changes between + and -. However, vibration data containing measurement errors exhibits a pattern of low periodicity, unlike normal vibration data.
[0075] Therefore, the average value among the statistical information of the data can be an index capable of distinguishing between normal vibration data and vibration data including measurement errors.
[0076] Furthermore, the RMS value of the statistical information of the data can be an index showing information related to the amplitude of the data.
[0077] Meanwhile, the variance and standard deviation of data can be indicators capable of distinguishing between normal vibration data having periodicity and vibration data having non-periodic measurement errors. In particular, among a plurality of sample data extracted from one vibration data, sample data having measurement errors has a higher variance and / or standard deviation than other sample data. Therefore, the variance and standard deviation between sample data can be indicators capable of distinguishing between data related to measurement errors included in the vibration data. In addition, the difference in variance and / or standard deviation between vibration data and sample data can also be an indicator capable of distinguishing between data related to measurement errors included in the vibration data.
[0078] The candidate feature information may include abnormal features of signal shapes that appear relatively frequently in vibration data including measurement errors, features that are highly sensitive to dispersion, features that are less sensitive to signals generated by defects or periodic components, and the like.
[0079] For ease of explanation, the first statistical information, that is, the RMS of the sample data is referred to as s_rms, the mean of the sample data is referred to as s_mean, the variance of the sample data is referred to as s_var, and the standard deviation of the sample data is referred to as s_std.
[0080] Moreover, the second statistical information is represented by d_rms, d_mean, d_var, and d_std, respectively, as the RMS of the vibration data, the average of the vibration data, the variance of the vibration data, and the standard deviation of the vibration data.
[0081] The candidate feature information proposed in this specification is as shown in Table 2 below.
[0082] [Table 2]
[0083] Referring to Table 2, the first candidate feature information may be the ratio (s_mean / s_rms) of the mean (s_mean) of the sample data to the RMS (s_rms) of the sample data.
[0084] The second candidate feature information may be the ratio (s_std / d_rms) of the standard deviation (s_std) of the sample data to the RMS (d_rms) of the vibration data.
[0085] The third candidate feature information may be the ratio (s_std / d_std) of the standard deviation (s_std) of the sample data to the standard deviation (d_std) of the vibration data.
[0086] A fourth candidate feature information may be the ratio (s_mean / d_var) of the mean of the sample data (s_mean) to the variance of the vibration data (d_var).
[0087] The fifth candidate feature information may be the ratio (s_mean / d_std) of the mean (s_mean) of the sample data to the standard deviation (d_std) of the vibration data.
[0088] The sixth candidate feature information may be the ratio (s_mean / d_rms) of the mean of the sample data (s_mean) to the RMS of the vibration data (d_rms).
[0089] The seventh candidate feature information may be the ratio (d_mean / s_rms) of the mean (d_mean) of the vibration data to the RMS (s_rms) of the sample data.
[0090] The eighth candidate feature information may be the ratio (s_std / s_rms) of the standard deviation (s_std) of the sample data to the RMS (s_rms) of the sample data.
[0091] The ninth candidate feature information may be the ratio (s_var / d_var) of the variance of the sample data (s_var) to the variance of the vibration data (d_var).
[0092] The tenth candidate feature information may be the ratio (s_std / s_var) of the standard deviation (s_std) of the sample data to the variance (s_var) of the sample data.
[0093] The eleventh candidate feature information may be the ratio (s_var / s_std) of the variance (s_var) of the sample data to the standard deviation (s_std) of the sample data.
[0094] The twelfth candidate feature information may be the ratio (s_var / s_rms) of the variance of the sample data (s_var) to the RMS of the sample data (s_rms).
[0095] Of the four first statistical information (s_rms, s_mean, s_var, s_std) and four second statistical information (d_rms, d_mean, d_var, d_std), 56 pieces of ratio information can be derived for two pieces of information. Of these, it was confirmed that the aforementioned 12 candidate feature information are specific information that are sensitive to measurement errors in vibration data.
[0096] In the principal component analysis step (S34), the dimensions of the 12 candidate features are reduced through principal component analysis (PCA) to extract feature information. This is for efficient analysis using machine learning. However, the principal component analysis step (S34) may be omitted depending on the embodiment.
[0097] In the clustering analysis step (S35), the sampling data is clustered based on the feature information. Clustering is a type of unsupervised learning, and evaluates clustering based on the similarity between unlabeled data.
[0098] Since the method of this embodiment aims to distinguish between normal vibration data and vibration data containing measurement errors, the structural conditions to be determined through clustering are simple, and a small number of measurement error data must be found among a large amount of data. Therefore, a clustering technique is suitable as a machine learning method for distinguishing between normal vibration data and vibration data containing measurement errors.
[0099] Among clustering techniques, hierarchical clustering can be divided into agglomerative clustering (combined hierarchical clustering), which is a bottom-up method, and disjunctive clustering (split hierarchical clustering), which is a top-down method. Agglomerative clustering is a method that starts with each data and clusters similar data sequentially.
[0100] When applying agglomerative clustering to this embodiment, a method is performed in which N clusters, each having one sample data, are started, a similarity matrix is calculated for each of them, and based on this, the most similar pair of clusters is found and merged into one cluster, and the process of updating the similarity matrix again is repeated, so that all sample data becomes one cluster.
[0101] In addition, agglomerative clustering algorithms include Single linkage, Complete linkage, Centroid linkage, Average linkage, Ward's linkage, etc.
[0102] Among these, while other algorithms merge crowds based on the distance between crowds, Ward's linkage merges crowds in a manner that minimizes deviations within the crowds. In particular, Ward's linkage shows lower sensitivity to noise and abnormal values than Single linkage, and other algorithms have a tendency to cluster crowds of similar size, so it is considered to be more suitable for this embodiment that distinguishes between normal vibration data and vibration data containing measurement errors.
[0103] 7 and 8 are graphs showing the clustering results obtained by judging the existing vibration data in the second judging step.
[0104] 7 and 8, it can be seen that the normal vibration data (C1) and the vibration data including measurement errors (C2) are clustered into different clusters. In the second judgment step, 1,188 data out of 10,634 sample vibration data were determined to be vibration data including measurement errors (C2).
[0105] In the measurement error determination step (S36), it can be determined whether the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error based on the analysis results derived in the clustering analysis step (S35).
[0106] As shown in FIG. 5, the measurement error judgment for vibration data in the second judgment step (S30) is an area where even experts have difficulty making an accurate judgment. However, it is possible to provide an accurate judgment basis in an area where normal vibration data that cannot be judged accurately in the first judgment step (S20) is mixed with vibration data containing measurement errors, thereby complementing the judgment error in the first judgment step (S20).
[0107] As described above, the vibration measurement error determination method according to one embodiment of the present invention determines whether the vibration data includes data due to a measurement error by using two different determination methods, and combines the determination results of each determination method to finally determine whether the vibration data includes data due to a measurement error, thereby improving the accuracy of the determination.
[0108] In addition, erroneously measured vibration data can be selected to improve the soundness of the vibration data, which is the basic data for predictive diagnosis.
[0109] A vibration measurement error determination system using the above-described vibration measurement error determination method will now be described.
[0110] FIG. 9 is a block diagram showing a vibration measurement error discrimination system according to one embodiment of the present invention.
[0111] Referring to FIG. 9 , a vibration measurement error discrimination system according to one embodiment of the present invention may include a data acquisition unit 10, a first judgment unit 20, a second judgment unit 30, a final judgment unit 40, and a machine learning analysis database 50.
[0112] The data acquiring unit 10 executes the vibration data acquiring step (S10) described above. That is, the data acquiring unit 10 can acquire vibration data from a mobile vibration measuring device, or can acquire vibration data from a database (DB) in which the vibration data is stored.
[0113] The first determination unit 20 executes the above-mentioned first determination step (S20). That is, the first determination unit 20 can determine whether the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error based on a preset error data selection rule.
[0114] The second determination unit 30 executes the second determination step (S30) described above. That is, the second determination unit 30 can determine whether the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error based on a determination criterion different from the determination criterion of the first determination step (S20). For example, the second determination unit 30 can determine whether the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error by using a machine learning algorithm.
[0115] The final judgment unit 40 executes the above-mentioned final judgment step (S40). That is, the final judgment unit 40 compiles the judgment results of the first judgment unit 20 and the second judgment unit 30, and makes a final judgment as to whether or not the vibration data acquired in the vibration data acquisition step (S10) includes data due to a measurement error.
[0116] The final decision unit 40 may determine that there is no measurement error only for vibration data that is determined to be free of measurement errors by both the first decision unit 20 and the second decision unit 30. Alternatively, the final decision unit 40 may determine that there is no measurement error only for vibration data that is determined to be free of measurement errors by either the first decision unit 20 or the second decision unit 30.
[0117] For vibration data that is determined to have a measurement error, the final decision unit 40 can provide a notification to the user that a measurement error exists.
[0118] In this embodiment, the first judgment unit 20, the second judgment unit 30, and the final judgment unit 40 have been described as being configured separately from one another. However, this is for the purpose of describing the first judgment unit 20, the second judgment unit 30, and the final judgment unit 40 by functionally dividing them, and the first judgment unit 20, the second judgment unit 30, and the final judgment unit 40 may be physically configured as one or a plurality of arithmetic units.
[0119] The machine learning analytics database 50 may include machine learning algorithms that perform analytics on the vibration data.
[0120] The final decision unit 40 can provide the vibration data that is determined to be free of measurement errors to the machine learning analysis database 50, and the machine learning analysis database 50 can update the vibration data received from the final decision unit 40.
[0121] As described above, the vibration measurement error discrimination system according to one embodiment of the present invention determines whether the vibration data includes data due to a measurement error by using two different judgment methods, and combines the judgment results of each judgment method to finally determine whether the vibration data includes data due to a measurement error, thereby improving the accuracy of the judgment.
[0122] In addition, erroneously measured vibration data can be selected to improve the soundness of the vibration data, which is the basic data for predictive diagnosis.
[0123] Those skilled in the art will understand that the present invention can be embodied in other specific forms without changing the technical idea or essential features of the present invention. Therefore, the above-described embodiments are merely illustrative and should not be construed as limiting. The scope of the present invention is defined by the claims below rather than the above detailed description, and all modifications and variations derived from the meaning and scope of the claims and their equivalents should be interpreted as being included in the scope of the present invention. [Explanation of symbols]
[0124] 10 Data Acquisition Section 20 First Judgment Section 30 Second Judgment Section 40 Final Judgment Department 50 Machine Learning Analytics Database S10 Vibration data acquisition step S12 Sample data extraction step S20 First decision step S21 Time waveform data extraction step S22 Fast Fourier Transform step S23 Vibration value extraction step S24: Step of comparing the vibration value with a threshold value S25 Measurement error judgment step S30 Second decision step S31 Time waveform data extraction step S32 Sample Data Extraction Step S33: Candidate feature information extraction step S34 Principal component analysis step S35 Clustering analysis step S36 Measurement error judgment step S40 Final decision step
Claims
1. a vibration data acquisition step of measuring vibrations occurring in the structure and acquiring vibration data; a first judgment step of judging whether the vibration data is due to a measurement error based on a preset error data selection rule; a second decision step utilizing a machine learning algorithm to determine whether the vibration data is due to measurement error; and a final judgment step of determining whether the vibration data is due to a measurement error based on the results of the judgments made in the first judgment step and the second judgment step.
2. The error data selection rule is 2. The vibration measurement error determination method according to claim 1, further comprising the step of determining whether the vibration data is due to a measurement error based on a difference between an amplitude value of a low frequency region extracted from the vibration data and a threshold value.
3. The vibration measurement error determination method according to claim 2 , wherein the low frequency region includes a region of 3 Hz or less.
4. The vibration measurement error determination method according to claim 2 , wherein the threshold value is 0.6 mm / s.
5. The machine learning algorithm comprises:
2. The vibration measurement error judgment method according to claim 1, further comprising the steps of: extracting a plurality of sample data from the vibration data; extracting feature information for discriminating a measurement error from the extracted plurality of sample data; and judging whether the vibration data is due to a measurement error based on the extracted feature information.
6. The vibration measurement error determination method according to claim 5 , wherein the feature information is extracted from candidate feature information extracted from a plurality of the sample data through principal component analysis.
7. The machine learning algorithm comprises:
6. The vibration measurement error determination method according to claim 5, further comprising the steps of: extracting the feature information for each of the sample data; clustering the sample data based on the feature information; and determining whether the vibration data is due to a measurement error based on a clustering result.
8. a data acquisition unit that measures vibrations occurring in the structure and acquires vibration data; a first determination unit that determines whether the vibration data is due to a measurement error based on a preset error data selection rule; a second determination unit that utilizes a machine learning algorithm to determine whether the vibration data is due to a measurement error; and a final judgment unit that judges whether the vibration data is due to a measurement error based on the results of judgments made in the first judgment step and the second judgment step.
9. The error data selection rule is 9. The vibration measurement error discrimination system according to claim 8, wherein whether or not the vibration data is due to a measurement error is determined based on a difference between an amplitude value of a low frequency region extracted from the vibration data and a threshold value.
10. The vibration measurement error discrimination system according to claim 9 , wherein the low frequency region includes a region below 3 Hz.
11. 10. The vibration measurement error discrimination system of claim 9, wherein the threshold is 0.6 mm / s.
12. The machine learning algorithm comprises:
9. The vibration measurement error discrimination system according to claim 8, further comprising: a vibration measurement error discrimination system for discriminating a measurement error by extracting a plurality of sample data from the vibration data; and a vibration measurement error discrimination system for discriminating a measurement error by extracting feature information from the plurality of sample data.
13. The vibration measurement error discrimination system according to claim 12 , wherein the feature information is extracted from candidate feature information extracted from a plurality of the sample data through principal component analysis.
14. The machine learning algorithm comprises: The vibration measurement error discrimination system according to claim 12, wherein the feature information is extracted for each of the sample data, the sample data is clustered based on the feature information, and it is determined whether the vibration data is due to a measurement error based on a clustering result.
Citation Information
Patent Citations
Diaphragm for loudspeaker
JP1984000296A
Abnormality detection system and abnormality detection method
JP2019079356A
Information processing system
JP2020192015A
Vibration monitoring and diagnosing system for wind power generator
US20170363072A1
