CN117313020A - Data processing method of bearing type tension sensor - Google Patents
Data processing method of bearing type tension sensor Download PDFInfo
- Publication number
- CN117313020A CN117313020A CN202311617835.7A CN202311617835A CN117313020A CN 117313020 A CN117313020 A CN 117313020A CN 202311617835 A CN202311617835 A CN 202311617835A CN 117313020 A CN117313020 A CN 117313020A
- Authority
- CN
- China
- Prior art keywords
- data
- tension
- tension data
- subsequence
- sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title abstract description 8
- 238000000034 method Methods 0.000 claims abstract description 60
- 230000005856 abnormality Effects 0.000 claims abstract description 37
- 230000011218 segmentation Effects 0.000 claims description 64
- 230000002159 abnormal effect Effects 0.000 claims description 36
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 4
- 230000010354 integration Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 description 5
- 238000007621 cluster analysis Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000123 paper Substances 0.000 description 1
- 239000002985 plastic film Substances 0.000 description 1
- 229920006255 plastic film Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000004154 testing of material Methods 0.000 description 1
- 239000004753 textile Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE
- G01L5/00—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes
- G01L5/04—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes for measuring tension in flexible members, e.g. ropes, cables, wires, threads, belts or bands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
Abstract
The invention relates to the technical field of data processing, in particular to a data processing method of a bearing type tension sensor. The method comprises the following steps: acquiring a tension data sequence; obtaining the noise degree of each tension data according to the tension data sequence; acquiring a plurality of temporary data segments; obtaining an adjustment sample difference of each temporary data segment according to the noise degree of the tension data in each temporary data segment, and obtaining a plurality of final data segments according to the adjustment sample difference of each temporary data segment; and obtaining the abnormality degree of each final data segment according to the noise degree of each tension data in the final data segment, and carrying out abnormality judgment on the measured object according to the abnormality degree. Thereby, accurate abnormality determination is realized by eliminating interference of noise data on tension data sequence division.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a data processing method of a bearing type tension sensor.
Background
A load-bearing tension sensor is a sensor for measuring tension or pulling force on an object. Such sensors are typically designed to withstand high loads and measure their tension and are therefore suitable for a variety of applications, including engineering, manufacturing, material testing, and automation control. Abnormal tension data is generated when an object measured by the load tension sensor is abnormal. In order to judge the abnormal condition of the object measured by the bearing tension sensor, the tension data acquired by the bearing tension sensor is required to be analyzed, and the abnormal tension data is extracted.
The difference between normal tension data in tension data acquired by the bearing tension sensor is smaller, and the difference between abnormal tension data and normal tension data is larger. Therefore, normal tension data and abnormal tension data can be separated through cluster analysis, and then each data segment obtained through separation is subjected to abnormal analysis.
The Fisher optimal segmentation algorithm is used as a common clustering algorithm, and the principle of the algorithm is as follows: the data segmentation is realized by making the difference between the data in the segmented groups as small as possible and making the difference between the data in the groups as large as possible. When the load-bearing tension sensor is interfered by the outside, noise data can be acquired, and the noise data is not generated by the abnormality of the measured object, so that the noise data cannot be used for abnormality judgment of the measured object. Meanwhile, the difference between the noise data and the normal tension data is also larger, and the normal tension data are easily separated, or the abnormal tension data and the abnormal tension data are easily separated. It can interfere with the accuracy of the cluster analysis. Therefore, how to eliminate the interference of noise data on cluster analysis, so that abnormal tension data can be accurately extracted, and the problem to be solved later becomes.
Disclosure of Invention
In order to solve the technical problems, the invention provides a data processing method of a load-bearing tension sensor, which adopts the following technical scheme:
acquiring data of a measured object by using a bearing type tension sensor to obtain a plurality of tension data, and obtaining a tension data sequence according to the plurality of tension data;
obtaining a reference subsequence of each piece of tension data according to the tension data sequence, and carrying out replacement processing on the tension data in the reference subsequence of each piece of tension data to obtain a comparison subsequence of each piece of tension data; obtaining an adjustment influence area of each piece of tension data according to integration of a difference value of a fitting function of a reference subsequence and a fitting function of a comparison subsequence of each piece of tension data, and obtaining noise degree of each piece of tension data according to the adjustment influence area of each piece of tension data and fluctuation degree difference of the reference subsequence and the comparison subsequence of each piece of tension data;
dividing the tension data sequence, and acquiring a plurality of temporary data segments in the dividing process; obtaining an adjustment sample difference of each temporary data segment according to the noise degree of tension data in each temporary data segment and the difference between the tension data in the temporary data segments, and controlling the segmentation process of the tension data sequence according to the adjustment sample difference of each temporary data segment to obtain a plurality of final data segments;
and obtaining the abnormality degree of each final data segment according to the noise degree of each tension data in the final data segment and the extreme point number of the final data segment, and carrying out abnormality judgment on the measured object according to the abnormality degree.
Preferably, the step of obtaining the reference subsequence of each tension data according to the tension data sequence includes the specific steps of:
for any one tension data, a subsequence formed by M continuous tension data in a tension data sequence centering on the tension data is taken as a reference subsequence of the tension data, and M represents a preset subsequence length.
Preferably, the replacing the tension data in the reference subsequence of each tension data to obtain a comparison subsequence of each tension data includes the following specific steps:
recording any one tension data as target tension data, and recording the previous tension data and the next tension data of the target tension data as adjacent data of the target tension data; performing linear interpolation processing on two adjacent data of the target tension data by using a linear interpolation method to obtain interpolation data of the two adjacent data of the target tension data, and recording the interpolation data as interpolation data of the target tension data; replacing target tension data in a reference subsequence of the target tension data by interpolation data to obtain a comparison subsequence of the target tension data;
a comparison sub-sequence of each tension data is obtained.
Preferably, the integration of the difference between the fitting function of the reference subsequence and the fitting function of the comparison subsequence according to each piece of tension data obtains an adjustment influence area of each piece of tension data, and the method comprises the following specific steps:
for any tension data, a least square method is utilized to fit a polynomial to the reference subsequence to obtain a fitting function of the reference subsequence, and the fitting function of the comparison subsequence is obtained; and carrying out difference processing on the fitting function of the reference subsequence and the fitting function of the comparison subsequence to obtain a difference function, and recording the absolute value of the integral of the difference function as an adjustment influence area.
Preferably, the noise level of each tension data is obtained according to the adjustment influence area of each tension data and the fluctuation level difference between the reference subsequence and the comparison subsequence of each tension data, and the method comprises the following specific steps:
the absolute value of the difference value between each piece of tension data in the reference subsequence and the previous piece of tension data is recorded as the front difference value of the reference subsequence, and the absolute value of the difference value between each piece of tension data in the reference subsequence and the next piece of tension data is recorded as the rear difference value of the reference subsequence; the absolute value of the difference value between each piece of tension data in the comparison sub-sequence and the previous piece of tension data is recorded as the front difference value of the comparison sub-sequence, and the absolute value of the difference value between each piece of tension data in the comparison sub-sequence and the next piece of tension data is recorded as the rear difference value of the comparison sub-sequence;
the method for calculating the noise influence degree of each tension data comprises the following steps:
wherein,indicating the ith tension numberThe j-th preceding difference value of the reference subsequence according,/->The jth post-difference value of the reference sub-sequence representing the ith tension data, ||represents the absolute value sign, |is +.>Representing the number of front differences of the reference sub-sequence of each tension data, and also representing the number of rear differences of the reference sub-sequence of each tension data,/for each tension data>The degree of fluctuation of the reference subsequence representing the ith tension data, +.>The jth front difference of the comparison subsequence representing the ith tension data, +.>The jth post-difference value of the comparison subsequence representing the ith tension data,/i>Representing the number of front differences for the comparison sub-sequence for each tension data, and also representing the number of rear differences for the comparison sub-sequence for each tension data; />Indicating the degree of fluctuation of the comparison subsequence of the ith tension data, +.>Represents the ith tension data,/-)>Representing the average of all tension data in the reference subsequence of the ith tension data; />Represents the adjustment influence area of the ith tension data, < +.>The number of tension data contained in the reference subsequence representing each tension data is also indicative of the preset subsequence length,/->Indicating the noise level of the ith tension data.
Preferably, the splitting process is performed on the tension data sequence, and a plurality of temporary data segments are obtained in the splitting process, including the specific steps of:
analyzing the tension data sequence by using an elbow method to obtain the segmentation number K;
based on the segmentation number K, a Fisher optimal segmentation algorithm is utilized to segment the tension data sequence, and a plurality of temporary data segments are obtained in the segmentation process.
Preferably, the obtaining the adjustment sample difference of each temporary data segment according to the noise degree of the tension data in each temporary data segment and the difference between the tension data in the temporary data segments includes the following specific steps:
wherein,indicating the noise level of the ith tension data in the kth temporary data segment during segmentation, < >>Representing the i-th tension data in the kth temporary data segment during segmentation,/th tension data in the kth temporary data segment during segmentation>Representing the mean value of all tension data in the kth temporary data segment during segmentation,/for the segment>Represents an exponential function based on natural constants, < ->Representing the number of tension data of the kth temporary data segment during segmentation,/for the segment>Representing the adjusted sample differences for the kth temporary data segment during segmentation.
Preferably, the step of controlling the segmentation process of the tension data sequence according to the adjustment sample difference of each temporary data segment to obtain a plurality of final data segments includes the following specific steps:
in the segmentation process, the sample difference of each temporary data segment is replaced by an adjustment sample difference, and the Fisher optimal segmentation algorithm is utilized to complete the segmentation of the tension data sequence so as to obtain a plurality of final data segments.
Preferably, the obtaining the abnormality degree of each final data segment according to the noise degree of each tension data in the final data segment and the number of extreme point numbers of the final data segment includes the following specific steps:
wherein,representing the number of extreme points in the kth final data segment,/->Represents the number of tension data in the kth final data section,/->Noise level of the ith tension data representing the kth final data segment, +.>Represents an exponential function based on natural constants, < ->Representing a linear normalization process,/->Indicating the degree of abnormality of the kth final data segment.
Preferably, the determining of the abnormality of the measurement object according to the abnormality degree includes the specific steps of:
and judging the final data segment with the abnormality degree larger than the preset abnormality degree threshold value R as an abnormal data segment, and judging that the measured object is abnormal.
The invention has the following beneficial effects:
in order to realize the abnormality determination of the measured object, abnormal tension data needs to be extracted, and the abnormal tension data have large differences, so that the abnormal tension data and the normal tension data can be firstly separated, and then the separated tension data can be subjected to abnormality analysis.
The tension data sequence is acquired by using the bearing type tension sensor, and noise data exists in the tension data sequence, so that the noise data can interfere with the data sequence segmentation effect. Therefore, the noise degree of each tension data is obtained by analyzing the condition that each tension data accords with the characteristics of the noise data, the sample difference of each temporary data segment in the data sequence segmentation process is adjusted according to the noise degree of each tension data to obtain the adjusted sample difference of each temporary data segment, and the data sequence segmentation process is controlled by utilizing the adjusted sample difference of each temporary data segment to obtain a final data segment, so that the abnormal tension data and the normal tension data can be segmented in different final data segments. In order to judge which final data segments are data segments in which abnormal tension data exist, the abnormal degree of each final data segment is obtained according to the difference of the tension data in each final data segment and the noise degree of each tension data, the abnormal condition of the tension data in each final data segment is reflected through the abnormal degree of each final data segment, and the abnormal judgment of the measured object is realized according to the abnormal degree of each final data segment.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for processing data of a load-bearing tension sensor according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an area of influence of adjustment of tension data provided by an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description refers to specific embodiments, structures, features and effects of a data processing method for a load-bearing tension sensor according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
An embodiment of a data processing method for a load-bearing tension sensor is provided:
the following specifically describes a specific scheme of a data processing method of a load-bearing tension sensor provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for processing data of a load-bearing tension sensor according to an embodiment of the invention is shown, where the method includes:
s001: and acquiring a plurality of tension data by using a bearing type tension sensor, and obtaining a tension data sequence according to the plurality of tension data.
Specifically, every N seconds, tension data of a measured object is collected by using a bearing type tension sensor, and N times of tension data are collected, so that N tension data are obtained. The types of measurement objects include, but are not limited to, the following: wires, cables, textiles, paper, plastic films. The measuring object in this embodiment is an electric wire, and other measuring objects may be used in other embodiments, and the embodiment is not particularly limited. N represents a preset acquisition interval, N represents a preset acquisition frequency, the embodiment is described by taking N as 2 and N as 1000 as examples, other embodiments can take other values, and the embodiment is not particularly limited.
Further, the acquired N tension data are arranged in time sequence to obtain a tension data sequence.
S002: and obtaining the noise degree of each tension data according to the tension data sequence.
Since the noise data and the normal tension data are greatly different, some data are required as references for noise analysis.
Specifically, regarding any one tension data, a sub-sequence formed by M continuous tension data in a tension data sequence centering on the tension data is taken as a reference sub-sequence of the tension data, M represents a preset sub-sequence length, the embodiment is described by taking M as 11, other embodiments may take other values, and the embodiment is not particularly limited.
It should be noted that, the noise data is generally random, that is, the noise data is not generally continuously generated, and the difference between the noise data and the normal tension data is large. Thus, the noise data only affects the variation law in a small range, and the reference subsequence can reflect the data variation in a small range of each tension data, so that the noise data affects the variation law of the reference subsequence. Thus, the variation law of the reference subsequence is greatly varied after the noise data is removed, so that the noise degree of each tension data can be estimated based on the variation law.
Further, any one tension data is recorded as target tension data, and the previous tension data and the next tension data of the target tension data are recorded as adjacent data of the target tension data; performing linear interpolation processing on two adjacent data of the target tension data by using a linear interpolation method to obtain interpolation data of the two adjacent data of the target tension data, and recording the interpolation data as interpolation data of the target tension data; and replacing the target tension data in the reference subsequence of the target tension data by interpolation data to obtain a comparison subsequence of the target tension data. And similarly, obtaining a comparison subsequence of each tension data.
And for any tension data, using a least square method to fit a polynomial to the reference subsequence to obtain a fitting function of the reference subsequence, and similarly obtaining a fitting function of the comparison subsequence. Performing difference processing on the fitting function of the reference subsequence and the fitting function of the comparison subsequence to obtain a difference function, and recording the absolute value of the integral of the difference function as the adjustment influence area, as shown in schematic diagram 2And->Fitting functions representing the reference subsequence and the comparison subsequence, respectively,/for each reference subsequence>I-th tension data representing reference subsequence, < ->The i-1 th tension data representing the reference subsequence,>the (i+1) th tension data representing the reference subsequence,>interpolation data of two adjacent data representing the i-th tension data. The absolute value of the difference between each tension data in the reference sub-sequence and the previous tension data is recorded as the front difference of the reference sub-sequence, and the absolute value of the difference between each tension data in the reference sub-sequence and the next tension data is recorded as the rear difference of the reference sub-sequence. The difference between each tension data in the comparison sub-sequence and the previous tension dataThe absolute value of the value is marked as the front difference value of the comparison sub-sequence, and the absolute value of the difference value between each piece of tension data and the next piece of tension data in the comparison sub-sequence is marked as the rear difference value of the comparison sub-sequence.
It should be noted that, the interpolation data of two adjacent data is used to replace the original tension data, instead of directly removing the original tension data, to prevent the influence of different data quantity on the result.
The method for calculating the noise influence degree of each tension data comprises the following steps:
wherein,the j-th front difference value of the reference subsequence representing the i-th tension data,/th front difference value of the reference subsequence>The jth post-difference value of the reference sub-sequence representing the ith tension data, ||represents the absolute value sign, |is +.>Representing the number of front differences of the reference sub-sequence of each tension data, and also representing the number of rear differences of the reference sub-sequence of each tension data,/for each tension data>The degree of fluctuation of the reference subsequence representing the ith tension data, +.>The jth front difference of the comparison subsequence representing the ith tension data, +.>The jth post-difference value of the comparison subsequence representing the ith tension data,/i>Front difference number representing the comparison subsequence of each tension dataThe amount, also representing the number of post-differences for the comparison sub-sequences for each tension data; />Indicating the degree of fluctuation of the comparison subsequence of the ith tension data, +.>The difference between the fluctuation degree of the comparison sub-sequence of the ith tension data and the fluctuation degree of the reference sub-sequence is reflected, and the larger the value is, the larger fluctuation of the ith tension data before and after replacement is indicated, so that the difference between the ith tension data and other data is larger, and the larger the noise degree of the ith tension data is. />Represents the ith tension data,/-)>The mean of all tension data in the reference subsequence of i tension data represented; />Represents the adjustment influence area of the ith tension data, < +.>The number of tension data contained in the reference subsequence representing each tension data is also indicative of the preset subsequence length. />The variation of the projection area of the reference sub-sequence before and after replacement of the ith tension data is reflected, and the greater the value, the greater the noise level of the ith tension data. />Noise level of the i-th tension data, < +.>An exponential function based on a natural constant is represented.
Thus, the noise level of each tension data is obtained, and the possibility that each tension data is noise data can be reflected by the noise level.
S003: and dividing the tension data sequence according to the noise degree of each tension data to obtain a plurality of final data segments.
It should be noted that, when the tension data sequence is segmented by using the Fisher optimal segmentation algorithm, the number of segments needs to be determined first.
Specifically, the tension data sequence is analyzed by an elbow method to obtain the segmentation number K.
In order to prevent the noise data from interfering with the segmentation process, the segmentation parameters in the segmentation process should be adjusted according to the noise level of each piece of tension data. The segmentation parameters in the segmentation process in the Fisher optimal segmentation algorithm are sample differences of each temporary data segment.
Furthermore, based on the number K of segments, a Fisher optimal segmentation algorithm is utilized to segment the tension data sequence, and a plurality of temporary data segments are obtained in the segmentation process.
Further, the calculation formula of the sample difference of each temporary data segment in the segmentation process of the traditional Fisher optimal segmentation algorithm is as follows:
wherein,representing the i-th tension data in the kth temporary data segment during segmentation,/th tension data in the kth temporary data segment during segmentation>Representing the mean value of all tension data in the kth temporary data segment during segmentation,/for the segment>Representing the number of tension data for the kth temporary data segment during the segmentation process. />Representing the sample difference of the kth temporary data segment during segmentation.
In this embodiment, the sample difference of each temporary data segment in the segmentation process of the traditional Fisher optimal segmentation algorithm is adjusted, and the adjusted sample difference of each temporary data segment is recorded as the adjusted sample difference of each temporary data segment.
The calculation method of the adjustment sample difference of each temporary data segment comprises the following steps:
wherein,representing the noise degree of the ith tension data in the kth temporary data segment in the segmentation process, reducing the influence of data with large noise degree on the segmentation analysis in order to reduce the interference of the noise data on the segmentation analysis, and reducing the influence of +_>Representing the i-th tension data in the kth temporary data segment during segmentation,/th tension data in the kth temporary data segment during segmentation>Representing the mean value of all tension data in the kth temporary data segment during segmentation,/for the segment>Represents an exponential function based on natural constants, < ->Representing the number of tension data for the kth temporary data segment during the segmentation process. />Representing the adjusted sample differences for the kth temporary data segment during segmentation.
Further, in the segmentation process, the sample difference of each temporary data segment is replaced by an adjustment sample difference, and the tension data sequence is segmented by using a Fisher optimal segmentation algorithm to obtain a plurality of final data segments.
S004: and calculating the abnormality degree of each final data segment, and carrying out abnormality judgment according to the abnormality degree of each final data segment.
In order to determine that the measured object is abnormal, abnormal tension data needs to be acquired, and the abnormal tension data and normal tension data are already divided into a plurality of final data segments through data division in the previous process, so that only the abnormal condition of each final data segment needs to be analyzed at the moment.
It should be further noted that when the tension data has a high noise level, the tension data is more likely to be noise data, and thus the tension data is less likely to be abnormal tension data. The abnormal tension data described herein refers to abnormal data generated due to an abnormality of the measurement object, not noise data. Meanwhile, when the measured object is free from abnormality and noise interference is not present, the obtained tension data generally has small fluctuation. When there is an abnormality in the measured object, the measured tension data varies greatly. And thus determine the degree of abnormality of each final data segment based thereon.
Specifically, the method for calculating the abnormality degree of each final data segment includes:
wherein,representing the number of extreme points in the kth final data segment,/->Represents the number of tension data in the kth final data section,/->The value reflects the frequency of variation of the kth final data segment, and a larger value indicates that the final data segment varies more frequentlyPropagation, thus the final data segment is more abnormal, < >>Noise level of the ith tension data representing the kth final data segment, +.>Reflecting the noise of the tension data in the final data segment, the larger the value is, the less likely the tension data in the final data segment is abnormal tension data, +.>An exponential function based on a natural constant is represented. />A linear normalization process is represented. />Indicating the degree of abnormality of the kth final data segment.
Further, the final data segment with the abnormality degree larger than the preset abnormality degree threshold value R is judged to be an abnormal data segment, when the abnormal data segment exists, the measured object is judged to be abnormal, and early warning is sent out. In this embodiment, R is taken as 0.42 as an example, and other values may be taken in other embodiments, and the embodiment is not particularly limited.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (10)
1. A method for processing data of a load-bearing tension sensor, the method comprising:
acquiring data of a measured object by using a bearing type tension sensor to obtain a plurality of tension data, and obtaining a tension data sequence according to the plurality of tension data;
obtaining a reference subsequence of each piece of tension data according to the tension data sequence, and carrying out replacement processing on the tension data in the reference subsequence of each piece of tension data to obtain a comparison subsequence of each piece of tension data; obtaining an adjustment influence area of each piece of tension data according to integration of a difference value of a fitting function of a reference subsequence and a fitting function of a comparison subsequence of each piece of tension data, and obtaining noise degree of each piece of tension data according to the adjustment influence area of each piece of tension data and fluctuation degree difference of the reference subsequence and the comparison subsequence of each piece of tension data;
dividing the tension data sequence, and acquiring a plurality of temporary data segments in the dividing process; obtaining an adjustment sample difference of each temporary data segment according to the noise degree of tension data in each temporary data segment and the difference between the tension data in the temporary data segments, and controlling the segmentation process of the tension data sequence according to the adjustment sample difference of each temporary data segment to obtain a plurality of final data segments;
and obtaining the abnormality degree of each final data segment according to the noise degree of each tension data in the final data segment and the extreme point number of the final data segment, and carrying out abnormality judgment on the measured object according to the abnormality degree.
2. The method for processing data of a load-bearing tension sensor according to claim 1, wherein the step of obtaining a reference subsequence of each tension data according to the tension data sequence comprises the steps of:
for any one tension data, a subsequence formed by M continuous tension data in a tension data sequence centering on the tension data is taken as a reference subsequence of the tension data, and M represents a preset subsequence length.
3. The method for processing load-bearing tension sensor data according to claim 1, wherein the replacing the tension data in the reference subsequence of each tension data to obtain the comparison subsequence of each tension data comprises the following specific steps:
recording any one tension data as target tension data, and recording the previous tension data and the next tension data of the target tension data as adjacent data of the target tension data; performing linear interpolation processing on two adjacent data of the target tension data by using a linear interpolation method to obtain interpolation data of the two adjacent data of the target tension data, and recording the interpolation data as interpolation data of the target tension data; replacing target tension data in a reference subsequence of the target tension data by interpolation data to obtain a comparison subsequence of the target tension data;
a comparison sub-sequence of each tension data is obtained.
4. The method for processing load-bearing tension sensor data according to claim 1, wherein the integrating of the difference between the fitting function of the reference sub-sequence and the fitting function of the comparison sub-sequence according to each tension data to obtain the adjustment influence area of each tension data comprises the following specific steps:
for any tension data, a least square method is utilized to fit a polynomial to the reference subsequence to obtain a fitting function of the reference subsequence, and the fitting function of the comparison subsequence is obtained; and carrying out difference processing on the fitting function of the reference subsequence and the fitting function of the comparison subsequence to obtain a difference function, and recording the absolute value of the integral of the difference function as an adjustment influence area.
5. The method for processing load-bearing tension sensor data according to claim 1, wherein the noise level of each tension data is obtained according to the adjustment influence area of each tension data and the fluctuation level difference between the reference subsequence and the comparison subsequence of each tension data, comprising the following specific steps:
the absolute value of the difference value between each piece of tension data in the reference subsequence and the previous piece of tension data is recorded as the front difference value of the reference subsequence, and the absolute value of the difference value between each piece of tension data in the reference subsequence and the next piece of tension data is recorded as the rear difference value of the reference subsequence; the absolute value of the difference value between each piece of tension data in the comparison sub-sequence and the previous piece of tension data is recorded as the front difference value of the comparison sub-sequence, and the absolute value of the difference value between each piece of tension data in the comparison sub-sequence and the next piece of tension data is recorded as the rear difference value of the comparison sub-sequence;
the method for calculating the noise influence degree of each tension data comprises the following steps:
wherein,the j-th front difference value of the reference subsequence representing the i-th tension data,/th front difference value of the reference subsequence>The jth post-difference value of the reference sub-sequence representing the ith tension data, ||represents the absolute value sign, |is +.>Representing the number of front differences of the reference sub-sequence of each tension data, and also representing the number of rear differences of the reference sub-sequence of each tension data,/for each tension data>The degree of fluctuation of the reference subsequence representing the ith tension data, +.>The jth front difference of the comparison subsequence representing the ith tension data, +.>The jth post-difference value of the comparison subsequence representing the ith tension data,/i>Representing the number of front differences for the comparison sub-sequence for each tension data, and also representing the number of rear differences for the comparison sub-sequence for each tension data; />Indicating the degree of fluctuation of the comparison subsequence of the ith tension data, +.>Represents the ith tension data,/-)>Representing the average of all tension data in the reference subsequence of the ith tension data; />Represents the adjustment influence area of the ith tension data, < +.>The number of tension data contained in the reference subsequence representing each tension data is also indicative of the preset subsequence length,/->Noise level of the i-th tension data, < +.>An exponential function based on a natural constant is represented.
6. The method for processing data of a load-bearing tension sensor according to claim 1, wherein the step of dividing the tension data sequence to obtain a plurality of temporary data segments during the dividing process comprises the following specific steps:
analyzing the tension data sequence by using an elbow method to obtain the segmentation number K;
based on the segmentation number K, a Fisher optimal segmentation algorithm is utilized to segment the tension data sequence, and a plurality of temporary data segments are obtained in the segmentation process.
7. The method for processing load-bearing tension sensor data according to claim 1, wherein the step of obtaining the adjustment sample difference for each temporary data segment according to the noise level of the tension data in each temporary data segment and the difference between the tension data in the temporary data segments comprises the following specific steps:
wherein,indicating the noise level of the ith tension data in the kth temporary data segment during segmentation, < >>Representing the i-th tension data in the kth temporary data segment during segmentation,/th tension data in the kth temporary data segment during segmentation>Representing the mean value of all tension data in the kth temporary data segment during segmentation,/for the segment>Represents an exponential function based on natural constants, < ->Representing the number of tension data of the kth temporary data segment during segmentation,/for the segment>Representing the adjusted sample differences for the kth temporary data segment during segmentation.
8. The method for processing load-bearing tension sensor data according to claim 1, wherein the step of controlling the segmentation process of the tension data sequence according to the adjustment sample difference of each temporary data segment to obtain a plurality of final data segments comprises the following specific steps:
in the segmentation process, the sample difference of each temporary data segment is replaced by an adjustment sample difference, and the Fisher optimal segmentation algorithm is utilized to complete the segmentation of the tension data sequence so as to obtain a plurality of final data segments.
9. The method for processing data of a load-bearing tension sensor according to claim 1, wherein the obtaining the degree of abnormality of each final data segment according to the noise degree of each tension data in the final data segment and the number of extreme points of the final data segment comprises the following specific steps:
wherein,representing the number of extreme points in the kth final data segment,/->Represents the number of tension data in the kth final data section,/->Noise level of the ith tension data representing the kth final data segment, +.>Represents an exponential function based on natural constants, < ->Representing a linear normalization process,/->Indicating the degree of abnormality of the kth final data segment.
10. The method for processing data of a load-bearing tension sensor according to claim 1, wherein the abnormality determination of the measurement object according to the abnormality degree comprises the specific steps of:
and judging the final data segment with the abnormality degree larger than the preset abnormality degree threshold value R as an abnormal data segment, and judging that the measured object is abnormal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311617835.7A CN117313020B (en) | 2023-11-30 | 2023-11-30 | Data processing method of bearing type tension sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311617835.7A CN117313020B (en) | 2023-11-30 | 2023-11-30 | Data processing method of bearing type tension sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117313020A true CN117313020A (en) | 2023-12-29 |
| CN117313020B CN117313020B (en) | 2024-01-26 |
Family
ID=89287017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311617835.7A Active CN117313020B (en) | 2023-11-30 | 2023-11-30 | Data processing method of bearing type tension sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117313020B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117540220A (en) * | 2024-01-09 | 2024-02-09 | 巨洋神州科技集团有限公司 | Near-zero carbon park source network load matching method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101319944A (en) * | 2007-06-08 | 2008-12-10 | 昆山市康达电子仪表有限公司 | Force transducer for on-line detecting tension force of thread type object |
| CN110826642A (en) * | 2019-11-15 | 2020-02-21 | 安徽农业大学 | Unsupervised anomaly detection method for sensor data |
| CN115858894A (en) * | 2023-02-14 | 2023-03-28 | 温州众成科技有限公司 | Visual big data analysis method |
| US20230288918A1 (en) * | 2022-03-09 | 2023-09-14 | The Boeing Company | Outlier detection and management |
-
2023
- 2023-11-30 CN CN202311617835.7A patent/CN117313020B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101319944A (en) * | 2007-06-08 | 2008-12-10 | 昆山市康达电子仪表有限公司 | Force transducer for on-line detecting tension force of thread type object |
| CN110826642A (en) * | 2019-11-15 | 2020-02-21 | 安徽农业大学 | Unsupervised anomaly detection method for sensor data |
| US20230288918A1 (en) * | 2022-03-09 | 2023-09-14 | The Boeing Company | Outlier detection and management |
| CN115858894A (en) * | 2023-02-14 | 2023-03-28 | 温州众成科技有限公司 | Visual big data analysis method |
Non-Patent Citations (1)
| Title |
|---|
| 庞向坤;黄越;王振;余彦;高嵩;: "基于相关系数的多变量异常数据段的检测", 控制工程, no. 01, pages 196 - 202 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117540220A (en) * | 2024-01-09 | 2024-02-09 | 巨洋神州科技集团有限公司 | Near-zero carbon park source network load matching method and system |
| CN117540220B (en) * | 2024-01-09 | 2024-04-05 | 巨洋神州科技集团有限公司 | Near-zero carbon park source network load matching method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117313020B (en) | 2024-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117313020B (en) | Data processing method of bearing type tension sensor | |
| CN111562108A (en) | Rolling bearing intelligent fault diagnosis method based on CNN and FCMC | |
| JP2015075821A (en) | State diagnostic method and state diagnostic device | |
| CN110716500B (en) | Method and system for determining segmented modeling points of temperature sensitive interval | |
| CN108107086B (en) | Gas detection method based on array gas sensor and gas sensor | |
| CN110569566B (en) | Method for predicting mechanical property of plate strip | |
| CN117167903B (en) | Artificial intelligence-based foreign matter fault detection method for heating ventilation equipment | |
| CN117196353B (en) | Environmental pollution assessment and monitoring method and system based on big data | |
| CN113858566B (en) | Injection molding machine energy consumption prediction method and system based on machine learning | |
| CN116451014A (en) | Real-time monitoring method for temperature of movable part in wireless passive equipment | |
| CN117235462B (en) | Intelligent fault prediction method for bag type packaging machine based on time sequence data analysis | |
| CN116772285B (en) | Intelligent building heating load safety real-time monitoring method | |
| CN112329272B (en) | Residual life fitting prediction method integrating photovoltaic module degradation and life data | |
| CN117148784B (en) | Operation fault analysis method for multi-axis multi-channel numerical control system | |
| CN116595403B (en) | Real-time monitoring and managing system for gas state of exhaust pipeline | |
| CN116738221B (en) | Pressurized pipeline gas analysis method and system | |
| CN116494493A (en) | Intelligent monitoring method for injection molding centralized feeding system | |
| CN109584225B (en) | Unsupervised defect detection method based on self-encoder | |
| CN112700039B (en) | Steady state detection and extraction method for load operation data of thermal power plant | |
| KR20150034482A (en) | Controling method of very small fraction of nonconforming | |
| CN110162853B (en) | Mechanical experiment data preprocessing method based on mechanical model | |
| CN113642183A (en) | Design method of EWMA control chart with variable sample volume residual error | |
| CN113312587A (en) | Sensor acquisition data missing value processing method based on ARIMA prediction and regression prediction | |
| CN117591986B (en) | Real-time automobile data processing method based on artificial intelligence | |
| CN117473345B (en) | Flange forging defect detection method based on data analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |