CN116881745A - Pressure transmitter abnormality monitoring method based on big data - Google Patents
Pressure transmitter abnormality monitoring method based on big data Download PDFInfo
- Publication number
- CN116881745A CN116881745A CN202311150132.8A CN202311150132A CN116881745A CN 116881745 A CN116881745 A CN 116881745A CN 202311150132 A CN202311150132 A CN 202311150132A CN 116881745 A CN116881745 A CN 116881745A
- Authority
- CN
- China
- Prior art keywords
- data
- pressure
- pressure data
- abnormal
- 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
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012544 monitoring process Methods 0.000 title claims abstract description 35
- 230000005856 abnormality Effects 0.000 title claims abstract description 15
- 230000002159 abnormal effect Effects 0.000 claims abstract description 101
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 abstract description 8
- 238000012545 processing Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 12
- 239000004973 liquid crystal related substance Substances 0.000 description 8
- 230000035945 sensitivity Effects 0.000 description 5
- 238000007621 cluster analysis Methods 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000000126 substance 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
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Measuring Fluid Pressure (AREA)
Abstract
The invention relates to the technical field of data processing, and provides a pressure transmitter abnormality monitoring method based on big data, which comprises the following steps: acquiring a pressure standard data sequence, and acquiring a plurality of pressure data of a pressure transmitter to form a pressure data sequence; clustering the pressure standard data sequences to obtain an abnormal pressure judgment threshold value, and obtaining vibration interference factors of each pressure data according to a standard pressure Gaussian model of the pressure standard data; acquiring abnormal probability of each pressure data according to vibration interference factors and distribution of the pressure data; correcting the variance model according to the abnormal probability and the vibration interference factor, and obtaining a classification result according to the variance result and an abnormal pressure judgment threshold; judging whether the pressure data is abnormal according to the classification result, and acquiring the pressure data in real time to perform abnormality monitoring. The invention aims to solve the problem that abnormal data cannot be accurately distinguished when clustering analysis is carried out on pressure data.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a pressure transmitter abnormality monitoring method based on big data.
Background
Pressure transmitters play an important role in many applications in today's high-tech society; pressure transmitters are widely used in chemical, petroleum, electrical, and other industries for detecting the pressure of liquids, gases, and the like; however, since the pressure transmitter works in a severe environment, faults are easy to occur, so that the problems of larger deviation of measurement results and the like are caused, and further, the safety risk in the related production process is caused, and even serious damage and personnel injury of equipment are possibly caused, so that the faults of the pressure transmitter must be found and solved in time, and abnormal monitoring of pressure data is required.
In the existing method, a clustering analysis method is generally adopted for abnormal monitoring of pressure data, the pressure data of the pressure transmitter is subjected to clustering analysis through an isodata algorithm, and abnormal data monitoring is realized according to a clustering result; however, the isodata has a merging and classifying process when being classified, the process depends on the variance value of the data, the variance value fluctuation is large due to the fact that the data volume is too small, and abnormal data cannot be found in time; the variance value change caused by the abnormal data is not obvious enough due to the overlarge data volume, and the abnormal data cannot be found, so that the weight needs to be set for the abnormal data, the variance value is more sensitive to the abnormal data, meanwhile, due to the existence of noise data, the influence of the noise data on the variance value is reduced, and further, the result obtained by clustering analysis of the pressure data is more accurate.
Disclosure of Invention
The invention provides a pressure transmitter abnormality monitoring method based on big data, which solves the problem that abnormal data cannot be accurately distinguished when the existing pressure data is subjected to cluster analysis, and adopts the following technical scheme:
one embodiment of the present invention provides a method for monitoring anomalies in a pressure transmitter based on big data, the method comprising the steps of:
acquiring a pressure standard data sequence, and acquiring a plurality of pressure data of a pressure transmitter to form a pressure data sequence;
clustering the pressure standard data sequences to obtain an abnormal pressure judgment threshold value, and obtaining vibration interference factors of each pressure data according to a standard pressure Gaussian model of the pressure standard data;
acquiring abnormal probability of each pressure data according to vibration interference factors and distribution of the pressure data; correcting the variance model according to the abnormal probability and the vibration interference factor, and obtaining a classification result according to the variance result and an abnormal pressure judgment threshold;
judging whether the pressure data is abnormal according to the classification result, and acquiring the pressure data in real time to perform abnormality monitoring.
Further, the clustering of the pressure standard data sequence to obtain the abnormal pressure judgment threshold comprises the following specific steps:
clustering the pressure standard data sequence through an isodata algorithm, wherein the distance measurement between elements in the sequence adopts the absolute value of the difference value of the two elements, and the clustering result is a category; and calculating variances for all elements in the pressure standard data sequence, and taking the obtained variances as abnormal pressure judgment thresholds of single categories.
Further, the specific method for obtaining the vibration interference factor of each pressure data includes:
obtaining a standard pressure Gaussian model according to a clustering result of the pressure standard data sequence; multiplying each data in the standard pressure Gaussian model by the inverse of the peak value in the model, and marking the obtained result as the Gaussian value of each data in the standard pressure Gaussian model;
taking any one pressure data in the pressure data sequence as target pressure data, inputting the target pressure data into a standard Gaussian distribution model to obtain a Gaussian value of the target pressure data, and marking a difference obtained by subtracting the Gaussian value from 1 as a vibration interference factor of the target pressure data;
a vibration disturbance factor is obtained for each pressure data in the sequence of pressure data.
Further, the standard pressure Gaussian model is obtained by the specific method:
acquiring a Gaussian function distribution model according to the pressure standard data sequence, and marking the Gaussian function distribution model as a standard pressure Gaussian model; the peak in the standard pressure gaussian model corresponds to the center point of the category in the clustered result.
Further, the abnormal probability of each pressure data is obtained by the specific method comprising the following steps:
the pressure data with the vibration interference factor larger than the vibration interference threshold value in the pressure data sequence is recorded as possible vibration data; obtaining reference pressure data and reference segment data of each pressure data according to the data value and vibration interference factor of each pressure data; first, theAbnormal probability of individual pressure data->The calculation method of (1) is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,represent the firstVibration disturbance factors of the individual pressure data,represent the firstThe time difference between the individual pressure data and the reference pressure data,represent the firstThe number of vibration possible data in the reference segment data of the individual pressure data,represent the firstReference segment data of the pressure dataVibration disturbance factors of the individual vibration possible data,an exponential function that is based on a natural constant;
the anomaly probability of each pressure data is acquired.
Further, the specific method for obtaining the reference pressure data and the reference segment data of each pressure data includes:
acquiring a plurality of pieces of data in the pressure data sequence according to the vibration possible data; for the firstPressure data, obtain->Absolute value of difference between each pressure data and each pressure data in the pressure data sequence, and marking the pressure data with the smallest absolute value of difference as +.>Reference pressure data for the individual pressure data;
for the firstA plurality of pieces of data before the time sequence of the pressure data, calculating the acquisition time and the +.>Absolute value of difference between the acquisition time of each pressure data is recorded as each segment of data and +.>Time differences of the individual pressure data; for->A plurality of pieces of data after the time sequence of the pressure data, calculating the acquisition time and the +.>Absolute value of difference between the acquisition time of each pressure data is recorded as each segment of data and +.>Time difference of the pressure data, the data with the smallest time difference is regarded as +.>Reference segment data for the individual pressure data;
and acquiring reference pressure data and reference segment data of each pressure data.
Further, the specific method for acquiring the segments of data in the pressure data sequence according to the vibration possible data includes:
and forming a section of data by adjacent vibration possible data in the pressure data sequence, wherein the vibration possible data of which the left and right adjacent pressure data are not the vibration possible data are independently used as a section of data, so as to obtain a plurality of sections of data in the pressure data sequence.
Further, the method for correcting the variance model according to the anomaly probability and the vibration interference factor comprises the following specific steps:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the result of the pressure data sequence output by means of the modified variance model,/for>Representing the number of pressure data in the sequence of pressure data, respectively>Indicate->Weight of individual pressure data, +.>Indicate->Data value of individual pressure data, +.>Representing the average of all pressure data in the sequence of pressure data; />Indicate->Vibration disturbance factor of individual pressure data, +.>Indicate->Abnormal probability of individual pressure data, +.>An exponential function based on a natural constant is represented.
Further, the classification result is obtained according to the variance result and the abnormal pressure judgment threshold, and the specific method comprises the following steps:
comparing the result output by the variance model after the correction of the pressure data sequence with an abnormal pressure judgment threshold value, and clustering all the pressure data into the existing categories of the pressure standard data sequence through an isodata algorithm if the result is smaller than or equal to the abnormal pressure judgment threshold value; if the pressure data is larger than the abnormal pressure judgment threshold, classifying the existing pressure standard data sequence and all pressure data in the pressure data sequence through an isodata algorithm, and obtaining a classification result after classification.
Further, the specific method for judging whether the pressure data is abnormal according to the classification result includes the following steps:
if the classification number in the classification result is more than or equal to 2, in the pressure data acquisition process in the pressure data sequence, the pressure transmitter is abnormal; if the classification number in the classification result is 1, the pressure transmitter is not abnormal, and the pressure data are normal.
The beneficial effects of the invention are as follows: according to the pressure sensor, the pressure data are clustered through an isodata algorithm, and the abnormal monitoring of the pressure transmitter is realized through a clustering result; the clustering analysis process of the isodata algorithm depends on the variance value of the data, and the variance value can not accurately reflect abnormal data due to too small data quantity, so that vibration interference factors are quantized through numerical distribution of the pressure data, noise data randomly generated by mechanical vibration or vibration interference factors of the abnormal data are large, the noise data randomly and independently appear and the characteristic that the abnormal data is continuously distributed is combined to obtain abnormal probability, the variance model is corrected through the abnormal probability and the vibration interference factors, so that the variance calculation reduces the perception of the noise data, the sensitivity to the abnormal data is improved, the abnormal data can not be larger due to larger data quantity, the combined classification result of the clustering analysis according to the variance result is more accurate, the accuracy of abnormal monitoring of the pressure transmitter is improved, faults of the pressure transmitter are timely found, and normal operation of the pressure transmitter is guaranteed.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a flow chart of a method for monitoring anomalies in a pressure transmitter based on big data according to one embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to FIG. 1, a flowchart of a method for monitoring anomalies in a pressure transmitter based on big data according to one embodiment of the present invention is shown, the method comprising the steps of:
and S001, acquiring a pressure standard data sequence, and acquiring a plurality of pressure data of the pressure transmitter to form the pressure data sequence.
The purpose of this embodiment is to perform anomaly monitoring on pressure data of a pressure transmitter, obtain an abnormal pressure judgment threshold value by cluster analysis by obtaining pressure data under normal conditions for a period of time, perform cluster analysis on pressure data obtained in real time, and obtain a classification result according to the abnormal pressure judgment threshold value, so that the pressure data under normal conditions and the pressure data collected in real time need to be obtained first.
Specifically, the pressure transmitter is used for monitoring pressure data of liquid or gas in a conveying pipeline, the pressure data in the conveying pipeline is relatively stable within a period of time when the equipment starts to operate, and the pressure data has small change, so that the pressure data of the period of time are collected to form a pressure standard data sequence, the pressure standard data sequence is used for providing an abnormal pressure judgment threshold value for subsequent cluster analysis, the embodiment collects pressure data for 5 minutes after the equipment starts to operate for 1 minute, wherein the sampling time interval is set to be 1 second, and the pressure standard data sequence is formed according to the collection sequence of the pressure data; after the pressure standard data sequence is obtained, the pressure transmitter is used for carrying out anomaly monitoring on the pressure data, 100 pressure data are collected for the first time to form the pressure data sequence for carrying out subsequent analysis, wherein the sampling time interval is set to be 1 second, and the pressure data sequence to be detected and the pressure standard data sequence are obtained; meanwhile, for subsequent processing, the embodiment sends the pressure data sequence to be detected and the pressure standard data sequence acquired in real time to a data center in a wireless transmission mode for subsequent analysis and anomaly monitoring.
Thus, a pressure standard data sequence and a plurality of pressure data are obtained.
And step S002, clustering the pressure standard data sequences to obtain an abnormal pressure judgment threshold value, and obtaining vibration interference factors of each pressure data according to a standard pressure Gaussian model of the pressure standard data.
It should be noted that, in this embodiment, the pressure standard data sequence and the pressure data sequence are continuously and circularly classified by using the isodata algorithm, so that no clustering is finished, the merging and classification are always in a state of merging and classifying, and the merging and classifying depend on variance values in a cluster, so that firstly, an abnormal pressure judgment threshold, namely, a variance threshold, is obtained based on the pressure standard data sequence under normal conditions, and a standard pressure gaussian model is constructed, and vibration interference factors of each pressure data are determined in the pressure data sequence by using the standard pressure gaussian model, so that sensitivity of noise data to the variance values is reduced by using the vibration interference factors, and further influence of noise data generated by inherent mechanical vibration in a conveying pipeline monitored by a pressure transmitter on classification results is reduced.
It should be further noted that, because the difference between the pressure data in the normal condition in the conveying pipeline is smaller, a stable value tends to be reached, and the noise data generated by the mechanical vibration is randomly and independently generated, the noise data points are far away from the center of the data distribution, and the pressure data after time sequence is closed to the data distribution center again; and meanwhile, determining an abnormal pressure judgment threshold according to the pressure standard data sequence, determining a clustering result of the pressure standard data sequence as a category through the abnormal pressure judgment threshold, and preventing normal pressure data from being divided into a plurality of categories through setting the lower limit of a single category.
Specifically, firstly, the variances are calculated for all elements in the pressure standard data sequence, the obtained variances are used as the abnormal pressure judgment threshold value of a single class in the isodata algorithm, and an implementer can set the abnormal pressure judgment threshold value, namely the variance threshold value, according to the actual situation, but the variances cannot be smaller than the variance value obtained by the pressure standard data sequence.
Further, clustering the pressure standard data sequence through an isodata algorithm, wherein the distance measurement between elements in the sequence adopts the absolute value of the difference value of two elements, namely the absolute value of the difference value of the pressure data, and the clustering result is a category; and obtaining the center point of the category, obtaining the absolute value of the difference value between each element in the pressure standard data sequence and the element corresponding to the center point, and recording the absolute value as the outlier distance of each element.
It should be further noted that, according to the central limit theorem, when the data volume is large enough, the data will show gaussian distribution, so a gaussian model can be constructed through outlier distances, and vibration interference factors can be calculated through the gaussian model; meanwhile, as noise data randomly and independently appear, the corresponding outlier distance of the noise data is larger, and the outlier distance corresponding to abnormal data is also larger, so that the vibration interference factor is quantified by combining the outlier distance with a Gaussian model.
Specifically, a gaussian function distribution model is obtained according to a pressure standard data sequence and is recorded as a standard pressure gaussian model, the gaussian function distribution model is obtained as a known technology, and the embodiment is not repeated, wherein a peak value in the standard pressure gaussian model corresponds to a center point of a class in a clustering result, corresponding data of each element in the pressure standard data sequence in the model can be obtained according to the standard pressure gaussian model, each data in the standard pressure gaussian model is multiplied by the reciprocal of the peak value in the model, and the obtained result is recorded as a gaussian value of each data in the standard pressure gaussian model; inputting any one of the pressure data in the pressure data sequence to be detected into a standard Gaussian distribution model, obtaining a Gaussian value of the pressure data, marking a difference obtained by subtracting the Gaussian value from 1 as a vibration interference factor of the pressure data, and obtaining the vibration interference factor of each pressure data in the pressure data sequence according to the method; since the larger the outlier distance, the larger the vibration disturbance factor should be, and the farther from the peak in the gaussian distribution model, the smaller the data will be, and the smaller the corresponding gaussian value will be, the vibration disturbance factor is subtracted by 1.
Thus, the vibration interference factor of each pressure data is obtained, and the variance threshold for the isodata algorithm is obtained at the same time, namely, the abnormal pressure judgment threshold for judging the abnormal pressure data later.
Step S003, according to vibration interference factors and distribution of the pressure data, obtaining abnormal probability of each pressure data; correcting the variance model according to the abnormal probability and the vibration interference factor, and obtaining a classification result according to the variance result and the abnormal pressure judgment threshold.
It should be noted that, noise data generated by mechanical vibration occurs randomly and independently, and abnormal data generated by the pressure transmitter continuously occurs in time sequence, i.e. a plurality of continuous pressure data are abnormal and slowly fall back to normal data, meanwhile, the outlier distance of the abnormal data is larger, and the abnormal probability of the pressure data is obtained according to the characteristics of outlier distance and continuous distribution; and meanwhile, the variance model is corrected according to the vibration interference factors and the abnormal probability, so that pressure data with larger abnormal probability is more sensitive in the variance calculation process, the influence of the pressure data on the variance is larger, and the data with larger vibration interference factors but smaller abnormal probability is noise data randomly generated by mechanical vibration, so that the influence of the noise data on the variance calculation and the abnormal pressure judgment is reduced.
Specifically, a vibration interference threshold is preset, the vibration interference threshold is described by using 0.7, for a pressure data sequence to be detected, pressure data in which all vibration interference factors are greater than the vibration interference threshold is obtained and is recorded as vibration possible data, adjacent vibration possible data form a section of data, and vibration possible data which are not continuously distributed, namely, neither left nor right adjacent pressure data are vibration possible data, are independently used as a section of data, and a plurality of sections of data in the pressure data sequence can be obtained; for the firstThe pressure data is obtained, the absolute value of the difference value of the pressure data and each pressure data in the pressure data sequence is obtained, and the pressure data with the minimum absolute value of the difference value is recorded as the pressure dataIs defined as the reference pressure data of (a); meanwhile, for a plurality of pieces of data before the time sequence of the pressure data, calculating the absolute value of the difference between the collection time of the last piece of pressure data in each piece of data and the collection time of the pressure data, and recording the absolute value as the time difference between each piece of data and the pressure data; for a plurality of pieces of data after the time sequence of the pressure data, calculating the absolute value of the difference between the acquisition time of the first piece of data in each piece of data and the acquisition time of the pressure data, recording the absolute value as the time difference between each piece of data and the pressure data, and taking the piece of data with the smallest time difference as the reference piece of data of the pressure data; then->Abnormal probability of individual pressure data->The calculation method of (1) is as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,represent the firstVibration disturbance factors of the individual pressure data,represent the firstThe time difference between the individual pressure data and the reference pressure data,represent the firstThe number of vibration possible data in the reference segment data of the individual pressure data,represent the firstReference segment data of the pressure dataVibration disturbance factors of the individual vibration possible data,representing an exponential function based on a natural constant, the present embodiment employsThe model presents the inverse proportional relationship and the normalization process,for inputting the model, an implementer can set an inverse proportion function and a normalization function according to actual conditions;
firstly, the larger the vibration interference factor is, the larger the outlier distance is, and the larger the outlier distance of the abnormal data is, so that the larger the vibration interference factor is, the larger the abnormal probability is; meanwhile, noise data is randomly and independently distributed, so that the time difference is larger, the abnormal probability is smaller, the vibration interference factor is larger, the time difference is smaller, the possible data of continuous vibration are possible, and the abnormal probability is larger; then, according to the reference segment data, making abnormal judgment, ifGreater than 0, thenThe value range of (1) is from 1 to positive infinity, so that the abnormal probability is increased; if->Less than 0, thenThe value range of (1) is 0 to 1, so as to realize the reduction of the abnormal probability value, if +.>Equal to 0>1, keeping the abnormal probability value unchanged; comparing the continuous distribution with the reference segment data, wherein the larger the vibration interference factor is, the larger the abnormal probability is, and the distribution of the abnormal data for a period of time is more consistent; the anomaly probability of each pressure data is obtained according to the above method.
Further, the correction method for the variance model according to the anomaly probability and the vibration interference factor comprises the following steps:
wherein, the liquid crystal display device comprises a liquid crystal display device,representing the result of the pressure data sequence output by means of the modified variance model,/for>Representing the number of pressure data in a pressure data sequence, wherein +.>；/>Indicate->Weight of individual pressure data, +.>Indicate->Data value of individual pressure data, +.>Representing the average of all pressure data in the sequence of pressure data; />Indicate->Vibration disturbance factor of individual pressure data, +.>Indicate->Abnormal probability of individual pressure data, +.>Representing an exponential function based on natural constants, the present embodiment employs +.>Model to present inverse proportional relationship and normalization process, < ->For inputting the model, an implementer can set an inverse proportion function and a normalization function according to actual conditions; on the basis of the traditional variance calculation, adding a weight to each item in the calculation process, and comparing the vibration interference factor with the anomaly probability to be used as the weight for adding, so that the data with larger anomaly probability is more sensitive in the variance calculation; the data with larger vibration interference factors and smaller abnormal probability reduces sensitivity in variance data, so that influence on classification results is avoided; if the abnormal probability is larger, the abnormal probability is directly used as the weight, and if the vibration interference factor is larger, the sensitivity is required to be reduced as the vibration interference factor is larger, and the weight is smaller through the inverse proportion relation, so that the sensitivity of noise data is reduced.
Further, after the result output by the variance model after the correction of the pressure data sequence is obtained, comparing the result with an abnormal pressure judgment threshold, and if the result is smaller than or equal to the abnormal pressure judgment threshold, clustering all the pressure data into the existing category of the pressure standard data sequence through an isodata algorithm; if the pressure difference is larger than the abnormal pressure judgment threshold value, classifying the existing pressure standard data sequence and all pressure data in the pressure data sequence through an isodata algorithm, obtaining a classification result after classification, and carrying out subsequent abnormal judgment according to the classification result.
So far, the pressure data are clustered, and a classification result is obtained.
And S004, judging whether the pressure data is abnormal according to the classification result, and acquiring the pressure data in real time to perform abnormality monitoring.
After clustering a pressure data sequence formed by 100 pieces of pressure data through an isodata algorithm, a classification result can be obtained, when the number of classifications in the classification result is more than or equal to 2, the abnormal condition of the pressure transmitter in the process of acquiring 100 pieces of pressure data is indicated, an early warning is sent out, the pressure transmitter is regulated, and the abnormal condition monitoring of the pressure data is realized; and when the classification number of the classification results is smaller than 2, namely the classification number is 1, the pressure transmitter is indicated to be abnormal in the process.
Furthermore, after the abnormality monitoring of 100 pieces of pressure data is completed, 50 pieces of pressure data are collected each time to form a pressure data sequence, the sampling time interval is still 1 second, the pressure data sequences obtained each time are clustered according to the method, classification results are obtained, and the judgment of whether the pressure transmitter is abnormal or not in the pressure data collection process is completed according to the classification number in the classification results, so that the abnormality monitoring of the pressure transmitter on the pressure data is realized.
Thus, the anomaly monitoring of the pressure data by the pressure transmitter based on big data is completed.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. The pressure transmitter abnormality monitoring method based on big data is characterized by comprising the following steps:
acquiring a pressure standard data sequence, and acquiring a plurality of pressure data of a pressure transmitter to form a pressure data sequence;
clustering the pressure standard data sequences to obtain an abnormal pressure judgment threshold value, and obtaining vibration interference factors of each pressure data according to a standard pressure Gaussian model of the pressure standard data;
acquiring abnormal probability of each pressure data according to vibration interference factors and distribution of the pressure data; correcting the variance model according to the abnormal probability and the vibration interference factor, and obtaining a classification result according to the variance result and an abnormal pressure judgment threshold;
judging whether the pressure data is abnormal according to the classification result, and acquiring the pressure data in real time to perform abnormality monitoring.
2. The abnormal pressure monitoring method based on big data of claim 1, wherein the clustering of the pressure standard data sequence to obtain the abnormal pressure judgment threshold comprises the following specific steps:
clustering the pressure standard data sequence through an isodata algorithm, wherein the distance measurement between elements in the sequence adopts the absolute value of the difference value of the two elements, and the clustering result is a category; and calculating variances for all elements in the pressure standard data sequence, and taking the obtained variances as abnormal pressure judgment thresholds of single categories.
3. The anomaly monitoring method for pressure transmitters based on big data according to claim 2, wherein the vibration disturbance factor of each pressure data is obtained by the following specific method:
obtaining a standard pressure Gaussian model according to a clustering result of the pressure standard data sequence; multiplying each data in the standard pressure Gaussian model by the inverse of the peak value in the model, and marking the obtained result as the Gaussian value of each data in the standard pressure Gaussian model;
taking any one pressure data in the pressure data sequence as target pressure data, inputting the target pressure data into a standard Gaussian distribution model to obtain a Gaussian value of the target pressure data, and marking a difference obtained by subtracting the Gaussian value from 1 as a vibration interference factor of the target pressure data;
a vibration disturbance factor is obtained for each pressure data in the sequence of pressure data.
4. The pressure transmitter anomaly monitoring method based on big data of claim 3, wherein the standard pressure gaussian model is obtained by the specific method of:
acquiring a Gaussian function distribution model according to the pressure standard data sequence, and marking the Gaussian function distribution model as a standard pressure Gaussian model; the peak in the standard pressure gaussian model corresponds to the center point of the category in the clustered result.
5. The anomaly monitoring method for pressure transmitters based on big data according to claim 1, wherein the anomaly probability of each pressure data is specifically obtained by:
the pressure data with the vibration interference factor larger than the vibration interference threshold value in the pressure data sequence is recorded as possible vibration data; obtaining reference pressure data and reference segment data of each pressure data according to the data value and vibration interference factor of each pressure data; first, theAbnormal probability of individual pressure data->The calculation method of (1) is as follows:
wherein (1)>Indicate->Vibration disturbance factor of individual pressure data, +.>Indicate->Time difference of the individual pressure data and the reference pressure data, < >>Indicate->The number of possible vibration data in the reference section data of the individual pressure data, < >>Indicate->Reference segment data of the pressure data +.>Vibration disturbance factor of the individual vibration possible data, +.>An exponential function that is based on a natural constant;
the anomaly probability of each pressure data is acquired.
6. The method for monitoring anomalies of pressure transmitters based on big data according to claim 5, wherein the obtaining of the reference pressure data and the reference section data of each pressure data comprises the following specific steps:
acquiring a plurality of pieces of data in the pressure data sequence according to the vibration possible data; for the firstPressure data is acquiredFirst->Absolute value of difference between each pressure data and each pressure data in the pressure data sequence, and marking the pressure data with the smallest absolute value of difference as +.>Reference pressure data for the individual pressure data;
for the firstA plurality of pieces of data before the time sequence of the pressure data, calculating the acquisition time and the +.>Absolute value of difference between the acquisition time of each pressure data is recorded as each segment of data and +.>Time differences of the individual pressure data; for->A plurality of pieces of data after the time sequence of the pressure data, calculating the acquisition time and the +.>Absolute value of difference between the acquisition time of each pressure data is recorded as each segment of data and +.>Time difference of the pressure data, the data with the smallest time difference is regarded as +.>Reference segment data for the individual pressure data;
and acquiring reference pressure data and reference segment data of each pressure data.
7. The method for monitoring the abnormality of the pressure transmitter based on big data according to claim 6, wherein the step of acquiring the pieces of data in the pressure data sequence based on the vibration possible data comprises the following specific steps:
and forming a section of data by adjacent vibration possible data in the pressure data sequence, wherein the vibration possible data of which the left and right adjacent pressure data are not the vibration possible data are independently used as a section of data, so as to obtain a plurality of sections of data in the pressure data sequence.
8. The method for monitoring the anomaly of the pressure transmitter based on big data according to claim 1, wherein the correcting the variance model according to the anomaly probability and the vibration-interference factor comprises the following specific steps:
wherein (1)>Representing the result of the pressure data sequence output by means of the modified variance model,/for>Representing the number of pressure data in the sequence of pressure data, respectively>Indicate->Weight of individual pressure data, +.>Indicate->Data value of individual pressure data, +.>Representing the average of all pressure data in the sequence of pressure data; />Indicate->Vibration disturbance factor of individual pressure data, +.>Indicate->Abnormal probability of individual pressure data, +.>An exponential function based on a natural constant is represented.
9. The method for monitoring the anomaly of the pressure transmitter based on big data according to claim 1, wherein the classifying result is obtained according to the variance result and the anomaly pressure judgment threshold value, comprises the following specific steps:
comparing the result output by the variance model after the correction of the pressure data sequence with an abnormal pressure judgment threshold value, and clustering all the pressure data into the existing categories of the pressure standard data sequence through an isodata algorithm if the result is smaller than or equal to the abnormal pressure judgment threshold value; if the pressure data is larger than the abnormal pressure judgment threshold, classifying the existing pressure standard data sequence and all pressure data in the pressure data sequence through an isodata algorithm, and obtaining a classification result after classification.
10. The method for monitoring the abnormality of the pressure transmitter based on big data according to claim 1, wherein the step of judging whether the abnormality occurs in the pressure data according to the classification result comprises the following specific steps:
if the classification number in the classification result is more than or equal to 2, in the pressure data acquisition process in the pressure data sequence, the pressure transmitter is abnormal; if the classification number in the classification result is 1, the pressure transmitter is not abnormal, and the pressure data are normal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311150132.8A CN116881745B (en) | 2023-09-07 | 2023-09-07 | Pressure transmitter abnormality monitoring method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311150132.8A CN116881745B (en) | 2023-09-07 | 2023-09-07 | Pressure transmitter abnormality monitoring method based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116881745A true CN116881745A (en) | 2023-10-13 |
| CN116881745B CN116881745B (en) | 2023-12-19 |
Family
ID=88266691
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311150132.8A Active CN116881745B (en) | 2023-09-07 | 2023-09-07 | Pressure transmitter abnormality monitoring method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116881745B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117171604A (en) * | 2023-11-03 | 2023-12-05 | 城资泰诺(山东)新材料科技有限公司 | Sensor-based insulation board production line abnormality monitoring system |
| CN117216484A (en) * | 2023-11-08 | 2023-12-12 | 山东汇蓝环保科技有限公司 | Environmental data monitoring method based on multidimensional data analysis |
| CN117235557A (en) * | 2023-11-14 | 2023-12-15 | 山东贺铭电气有限公司 | Electrical equipment fault rapid diagnosis method based on big data analysis |
| CN117433591A (en) * | 2023-12-21 | 2024-01-23 | 河北省农林科学院农业信息与经济研究所 | Automatic equipment monitoring system based on multi-sensor data |
| CN117828516A (en) * | 2024-03-06 | 2024-04-05 | 湖南大学 | On-site data optimization acquisition system for digital pressure tank production workshop |
| CN117828516B (en) * | 2024-03-06 | 2024-05-24 | 湖南大学 | On-site data optimization acquisition system for digital pressure tank production workshop |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040249583A1 (en) * | 1996-03-28 | 2004-12-09 | Evren Eryurek | Pressure transmitter with diagnostics |
| US20180095004A1 (en) * | 2016-10-03 | 2018-04-05 | International Business Machines Corporation | Diagnostic fault detection using multivariate statistical pattern library |
| CN109557560A (en) * | 2018-12-17 | 2019-04-02 | 北斗航天卫星应用科技集团有限公司 | A kind of Beidou Navigation System fault detection method and detection system based on ISODATA clustering algorithm |
| CN110941648A (en) * | 2019-11-19 | 2020-03-31 | 广州市自来水有限公司 | Abnormal data identification method, system and storage medium based on cluster analysis |
| CN112800686A (en) * | 2021-03-29 | 2021-05-14 | 国网江西省电力有限公司电力科学研究院 | Transformer DGA online monitoring data abnormal mode judgment method |
| WO2022110557A1 (en) * | 2020-11-25 | 2022-06-02 | 国网湖南省电力有限公司 | Method and device for diagnosing user-transformer relationship anomaly in transformer area |
| CN115860579A (en) * | 2023-02-27 | 2023-03-28 | 山东金利康面粉有限公司 | Production quality monitoring system for flour processing |
| CN115862301A (en) * | 2023-02-23 | 2023-03-28 | 深圳市特安电子有限公司 | Remote intelligent communication system for pressure transmitter |
-
2023
- 2023-09-07 CN CN202311150132.8A patent/CN116881745B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040249583A1 (en) * | 1996-03-28 | 2004-12-09 | Evren Eryurek | Pressure transmitter with diagnostics |
| US20180095004A1 (en) * | 2016-10-03 | 2018-04-05 | International Business Machines Corporation | Diagnostic fault detection using multivariate statistical pattern library |
| CN109557560A (en) * | 2018-12-17 | 2019-04-02 | 北斗航天卫星应用科技集团有限公司 | A kind of Beidou Navigation System fault detection method and detection system based on ISODATA clustering algorithm |
| CN110941648A (en) * | 2019-11-19 | 2020-03-31 | 广州市自来水有限公司 | Abnormal data identification method, system and storage medium based on cluster analysis |
| WO2022110557A1 (en) * | 2020-11-25 | 2022-06-02 | 国网湖南省电力有限公司 | Method and device for diagnosing user-transformer relationship anomaly in transformer area |
| CN112800686A (en) * | 2021-03-29 | 2021-05-14 | 国网江西省电力有限公司电力科学研究院 | Transformer DGA online monitoring data abnormal mode judgment method |
| CN115862301A (en) * | 2023-02-23 | 2023-03-28 | 深圳市特安电子有限公司 | Remote intelligent communication system for pressure transmitter |
| CN115860579A (en) * | 2023-02-27 | 2023-03-28 | 山东金利康面粉有限公司 | Production quality monitoring system for flour processing |
Non-Patent Citations (1)
| Title |
|---|
| 王燕妮 等: "模糊ISODATA聚类结合直方图熵值算法的异常行为检测", 《现代电子技术》, vol. 40, no. 12, pages 120 - 123 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117171604A (en) * | 2023-11-03 | 2023-12-05 | 城资泰诺(山东)新材料科技有限公司 | Sensor-based insulation board production line abnormality monitoring system |
| CN117171604B (en) * | 2023-11-03 | 2024-01-19 | 城资泰诺(山东)新材料科技有限公司 | Sensor-based insulation board production line abnormality monitoring system |
| CN117216484A (en) * | 2023-11-08 | 2023-12-12 | 山东汇蓝环保科技有限公司 | Environmental data monitoring method based on multidimensional data analysis |
| CN117216484B (en) * | 2023-11-08 | 2024-01-26 | 山东汇蓝环保科技有限公司 | Environmental data monitoring method based on multidimensional data analysis |
| CN117235557A (en) * | 2023-11-14 | 2023-12-15 | 山东贺铭电气有限公司 | Electrical equipment fault rapid diagnosis method based on big data analysis |
| CN117235557B (en) * | 2023-11-14 | 2024-02-02 | 山东贺铭电气有限公司 | Electrical equipment fault rapid diagnosis method based on big data analysis |
| CN117433591A (en) * | 2023-12-21 | 2024-01-23 | 河北省农林科学院农业信息与经济研究所 | Automatic equipment monitoring system based on multi-sensor data |
| CN117433591B (en) * | 2023-12-21 | 2024-02-27 | 河北省农林科学院农业信息与经济研究所 | Automatic equipment monitoring system based on multi-sensor data |
| CN117828516A (en) * | 2024-03-06 | 2024-04-05 | 湖南大学 | On-site data optimization acquisition system for digital pressure tank production workshop |
| CN117828516B (en) * | 2024-03-06 | 2024-05-24 | 湖南大学 | On-site data optimization acquisition system for digital pressure tank production workshop |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116881745B (en) | 2023-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116881745B (en) | Pressure transmitter abnormality monitoring method based on big data | |
| CN117093879B (en) | Intelligent operation management method and system for data center | |
| CN112284440B (en) | Sensor data deviation self-adaptive correction method | |
| CN104462794A (en) | Algorithm for finding abnormal data of environmental monitoring based on comparative statistic analysis | |
| CN115409131B (en) | Production line abnormity detection method based on SPC process control system | |
| CN116070163B (en) | Indoor harmful gas concentration anomaly monitoring data processing method | |
| CN115829422A (en) | Industrial equipment operation abnormal state identification method based on big data | |
| CN115876258B (en) | Livestock and poultry breeding environment abnormity monitoring and alarming system based on multi-source data | |
| CN111191855B (en) | Water quality abnormal event identification and early warning method based on pipe network multi-element water quality time sequence data | |
| CN116066343A (en) | Intelligent early warning method and system for fault model of oil delivery pump unit | |
| CN110852906B (en) | Method and system for identifying electricity stealing suspicion based on high-dimensional random matrix | |
| CN109255201B (en) | SOM-MQE-based ball screw pair health assessment method | |
| CN117390591B (en) | Operation monitoring method and system for coal conveying belt sampling machine based on electric parameter analysis | |
| CN117556366A (en) | Data abnormality detection system and method based on data screening | |
| CN116662111B (en) | Intelligent network management control system for train | |
| CN112949735A (en) | Liquid hazardous chemical substance volatile concentration abnormity discovery method based on outlier data mining | |
| CN116503025A (en) | Business work order flow processing method based on workflow engine | |
| CN114611633B (en) | Health monitoring method of electromagnetic valve | |
| CN113836813B (en) | Blast furnace tuyere water leakage detection method based on data analysis | |
| CN115163195A (en) | Coal mine coal face gas sensor numerical value distortion identification method and system | |
| CN116956089A (en) | Training method and detection method for temperature anomaly detection model of electrical equipment | |
| CN117406161B (en) | Metering device magnitude deviation early warning method, system, equipment and medium | |
| CN117421616B (en) | Mine shaft detection system and method | |
| CN117665224B (en) | Intelligent laboratory management method for food detection | |
| CN117668684B (en) | Power grid electric energy data anomaly detection method based on big 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 |