CN112949677A - Sorting backflow abnormal relevance analysis method based on proximity algorithm - Google Patents
Sorting backflow abnormal relevance analysis method based on proximity algorithm Download PDFInfo
- Publication number
- CN112949677A CN112949677A CN202110002725.4A CN202110002725A CN112949677A CN 112949677 A CN112949677 A CN 112949677A CN 202110002725 A CN202110002725 A CN 202110002725A CN 112949677 A CN112949677 A CN 112949677A
- Authority
- CN
- China
- Prior art keywords
- algorithm
- sorting
- data
- samples
- time 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.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 13
- 230000002159 abnormal effect Effects 0.000 title claims description 7
- 238000000034 method Methods 0.000 claims abstract description 30
- 238000010219 correlation analysis Methods 0.000 claims abstract description 18
- 238000012795 verification Methods 0.000 claims abstract description 6
- 238000012544 monitoring process Methods 0.000 claims abstract description 5
- 238000013024 troubleshooting Methods 0.000 claims abstract description 5
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000009826 distribution Methods 0.000 claims description 22
- 230000002596 correlated effect Effects 0.000 claims description 5
- 238000005311 autocorrelation function Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 4
- 230000000875 corresponding effect Effects 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims 1
- 238000010992 reflux Methods 0.000 abstract description 7
- 238000003745 diagnosis Methods 0.000 abstract description 6
- 238000011084 recovery Methods 0.000 abstract description 4
- 235000019504 cigarettes Nutrition 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 5
- 230000005856 abnormality Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 2
- 241000208125 Nicotiana Species 0.000 description 1
- 235000002637 Nicotiana tabacum Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24143—Distances to neighbourhood prototypes, e.g. restricted Coulomb energy networks [RCEN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
Abstract
The invention provides a sorting backflow abnormity correlation analysis method based on a proximity algorithm, which comprises a K-neighbor algorithm for performing correlation analysis on historical data of sorting backflow abnormity related events and time sequence data of finished wrapped logistics, and two baseline algorithms for comparing results, wherein the analysis method comprises the following steps: step one, collecting historical data of a bar code PLC of a case sealer, and preprocessing the collected data; step two, verifying the processed data by using a K-nearest neighbor algorithm and two baseline algorithms respectively; and step three, troubleshooting and judging faults of the monitoring system and the automatic control system through the verification result. The method can automatically judge the incidence relation between related events and time sequence data when the sorting reflux abnormity of the roll-up logistics finished products occurs and analyze the incidence relation, thereby providing a diagnosis clue for the reason of the occurrence of the events, reducing the fault recovery time, reducing the occurrence frequency of the sorting reflux abnormity of the finished products and improving the product quality.
Description
Technical Field
The invention relates to the technical field of equipment fault diagnosis of a cigarette stream line in the tobacco industry, in particular to a K-nearest neighbor algorithm-based sorting reflux anomaly correlation analysis method for cigarette stream finished products.
Background
The bar code PLC of the case sealer can automatically acquire data such as logs, performances, events and the like in daily operation, and the data are divided into time sequence data and events. The time series data is a real-valued-time series (usually with a fixed time interval), such as payload bit rate, etc.; and event data is a sequence of events that records the occurrence of a particular event, such as a heartbeat timeout event, etc.
The roll-up logistics is one of the very important production lines of the cigarette production system, the abnormal site of finished product sorting backflow can appear on irregular time of the production line, the problem that cigarettes jump out of the production line is caused, the bar code PLC of the case sealer stops working, and because the monitoring system of the equipment state and the automatic control system are relatively independent, when a fault occurs, a large amount of time is needed for troubleshooting, and the fault depends on personal experience of maintainers, the downtime is long, and great economic loss is generated.
Disclosure of Invention
In order to solve some technical problems in the prior art, the invention provides a sorting backflow abnormity correlation analysis method based on a proximity algorithm, which can automatically judge and analyze the correlation between related events and time sequence data when sorting backflow abnormity of finished products of roll-up logistics occurs, thereby providing diagnosis clues for the reasons of the events, reducing fault recovery time, reducing the occurrence frequency of sorting backflow abnormity of finished products and improving the product quality.
In order to solve the above-mentioned existing technical problem, the invention adopts the following scheme:
a sorting backflow abnormity correlation analysis method based on a proximity algorithm comprises a K-neighbor algorithm for performing correlation analysis on historical data of sorting backflow abnormity related events and time sequence data of finished roll logistics, and two baseline algorithms for performing result comparison on analysis results of the K-neighbor algorithm, wherein the analysis method comprises the following steps:
step one, collecting historical data of a bar code PLC of a case sealer, wherein the collected historical data comprises connection request non-response times, byte number per second, effective load bit rate, server TCP retransmission packets, client side average ACK time delay, code scanning, bar code transmission, heartbeat IO timeout, operating system restart and a large number of IO read-write, and preprocessing the collected data;
step two, verifying the processed data by using a K-nearest neighbor algorithm and two baseline algorithms respectively;
and step three, troubleshooting and judging faults of the monitoring system and the automatic control system through the verification result.
Further, two of the baseline algorithms include the Pearson correlation algorithm and the J-measure correlation algorithm, with the evaluation criterion being F-score.
Further, the K-nearest neighbor algorithm is calculated in the following manner: firstly, judging and converting the correlation between the finished product sorting backflow abnormity related event and the time sequence data into two sample problems; second, determine if the two samples are from the same distribution.
Further, the manner of determining whether the two samples come from the same distribution is: firstly, selecting corresponding n segments of time sequence sample data with length of k before and after occurrence of an event, and using the selected data Is represented by eiRepresenting a certain event, the sample group theta is a series of sample data with the length of k randomly selected on the time sequence, and then, according to the sample group And judging with the distribution result of the sample group theta, wherein when the distribution is different, the event is related to the time sequence data, and when the distribution is the same, the event is not related to the time sequence data.
Further, the set of samples ΓfrontAnd all samples in the sample group theta are used to calculate the distance between each sample in the two sample groups, one belonging to the sample group, by means of the DTW algorithmSample set ΓfrontOr samples S of a sample group theta, a plurality of nearest neighbor samples m of S are analyzed, and the larger the number of samples belonging to the same sample group as S, the larger the number of samples belonging to the same sample group as S means that the sample group gamma isfrontAnd theta distribution are much different, i.e., the more correlated the event and timing data.
Further, the number m of nearest neighbor samples is a natural logarithm, the first peak of the autocorrelation function curve of the time series length k is a series length, and the overall algorithm is as follows:
Input:Event E=(e1,e2,...,en),and Time Series
S=(s1,s2,...,sm),and the sub-series length k.
OutPut:The direction D,
1 InitializeΓfrontand Γrear;
2Initialize θ;
3Initialize D=NULL,
5Test Γfrontand θ using Nearest Neighbors Method.
The result is denoted as Df.;
6TestΓrearandθusing Nearest Neighbors Method.
The result is denoted as Df.;
7Out put D;
8Algorithm End.
compared with the prior art, the invention has the beneficial effects that:
through an unsupervised and statistical discrimination algorithm K-nearest neighbor algorithm, the incidence relation between related events and time sequence data when the sorting reflux abnormality of the roll-up logistics finished products occurs is automatically judged, and through the analysis of the related time sequence data of the events, a diagnosis clue is provided for the reason of the occurrence of the events, so that the fault recovery time is shortened, the occurrence frequency of the sorting reflux abnormality of the finished products is reduced, and the product quality is improved.
Drawings
FIG. 1 is a sample set analysis diagram based on the K-nearest neighbor algorithm in the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and the detailed description, and it should be noted that any combination of the embodiments or technical features described below can be used to form a new embodiment without conflict.
A sorting backflow abnormity correlation analysis method based on a proximity algorithm comprises a K-neighbor algorithm for performing correlation analysis on historical data of sorting backflow abnormity related events and time sequence data of finished roll logistics, and two baseline algorithms for performing result comparison on analysis results of the K-neighbor algorithm, wherein the analysis method comprises the following steps:
step one, collecting historical data of a bar code PLC of a case sealer, wherein the collected historical data comprises connection request non-response times, byte number per second, effective load bit rate, server TCP retransmission packets, client side average ACK time delay, code scanning, bar code transmission, heartbeat IO timeout, operating system restart and a large number of IO read-write, and preprocessing the collected data;
step two, verifying the processed data by using a K-nearest neighbor algorithm and two baseline algorithms respectively;
and step three, troubleshooting and judging faults of the monitoring system and the automatic control system through the verification result.
Further, two of the baseline algorithms include the Pearson correlation algorithm and the J-measure correlation algorithm, with the evaluation criterion being F-score.
Further, the K-nearest neighbor algorithm is calculated in the following manner: firstly, judging and converting the correlation between the finished product sorting backflow abnormity related event and the time sequence data into two sample problems; second, determine if the two samples are from the same distribution.
Further, the manner of determining whether the two samples come from the same distribution is: firstly, selecting corresponding n segments of time sequence sample data with length of k before and after occurrence of an event, and using the selected data Is represented by eiRepresenting a certain event, the sample group theta is a series of sample data with the length of k randomly selected on the time sequence, and then, according to the sample group And judging with the distribution result of the sample group theta, wherein when the distribution is different, the event is related to the time sequence data, and when the distribution is the same, the event is not related to the time sequence data.
Further, the set of samples ΓfrontAnd all samples in the sample group theta are used to calculate the distance between the respective samples in the two sample groups by means of the DTW algorithm, some belonging to the sample group ΓfrontOr samples S of a sample group theta, a plurality of nearest neighbor samples m of S are analyzed, and the larger the number of samples belonging to the same sample group as S, the larger the number of samples belonging to the same sample group as S means that the sample group gamma isfrontAnd theta distribution are much different, i.e., the more correlated the event and timing data.
Further, the number m of nearest neighbor samples is a natural logarithm, the first peak of the autocorrelation function curve of the time series length k is a series length, and the overall algorithm is as follows:
InPut:Event E=(e1,e2,...,en),and Time Series
S=(s1,s2,...,sm),and the sub-series length k.
OutPut:The direction D,
1Initialize ΓfrontandΓrear;
2Initializeθ;
3 Initialize D=NULL.
5Test Γfrontand θ using Nearest Neighbors Method.
The result is denoted as Df.;
6TestΓrearand θ using Nearest Neighbors Method.
The result is denoted as Df.;
7Out put D;
8Algorithm End.
the method automatically judges the incidence relation between related events and time sequence data when the sorting reflux abnormality of the roll-up logistics finished products occurs through an unsupervised and statistical discrimination algorithm K-nearest neighbor algorithm. By analyzing the time sequence data related to the event, a diagnosis clue is provided for the reason of the event occurrence, the fault recovery time is reduced, the occurrence frequency of the sorting reflux abnormality of the finished product is reduced, and the product quality is improved.
The specific embodiment is as follows:
as shown in fig. 1, in the correlation analysis method based on the K-nearest neighbor algorithm, the correlation judgment of the finished product sorting reflow abnormal related event and the time sequence data is first converted into a two-sample problem, and the core of the two-sample hypothesis test is to judge whether the two samples are from the same distribution. Firstly, selecting corresponding n segments of time sequence sample data with length of k before and after occurrence of an event, and using the selected dataIs represented by eiIndicating a certain event. The sample set θ is a series of sample data with length k randomly selected in time series. If the sample setEvents are correlated with timing data, as opposed to sample set θ distribution; if the distributions are the same, the event is independent of the timing data.
Assume that samples 0-4 in FIG. 1 above are from sample set ΓfrontAnd a-e belong to a sample group theta, and the distance between two samples is calculated by using a DTW algorithm (the DTW algorithm can be well adapted to the expansion and displacement of sequence data). Some of which belong to a group of samples tfrontOr samples S of the sample group theta, the larger the number of m nearest neighbor samples of S belonging to the same sample group as S, the more the sample group gammafrontAnd theta distribution are much different, i.e., the more correlated the event and timing data.
For example, taking the number of neighbors m to 2, the two nearest neighbors of sample c are 3 and a from two different sample sets, respectively, but the two nearest neighbors of sample a are c and d from the same sample set θ. Confidence coefficients are used to determine the confidence level of the "hypothesis test H1" (the two distributions are not the same, i.e., the event and time series data are related), the greater the confidence coefficient, the more reliable H1. Two key parameters of the algorithm: the number m of nearest neighbors and the length k of a time sequence are obtained, the number of neighbors is the natural logarithm of the number of samples, and the first peak value of an autocorrelation function curve of time sequence data is the length of the sequence.
The overall algorithm is as follows:
Input:Event E=(e1,e2,...,,en),and Time Series
S=(s1,s2,...,sm),and the sub-series length k.
OutPut:The direction D,
1Initialize Γfrontand Γrear;
2Initializeθ;
3Initialize D=NULL,
5Test Γfrontandθusing Nearest Neighbors Method.
The result is denoted as Df.;
6Test Γrearandθusing Nearest Neighbors Method.
The result is denoted as Df.;
7Out put D;
8Algorithm End.
the method has the advantages of easy implementation, simplicity and quickness. However, special attention needs to be paid to modeling, and the K-nearest neighbor algorithm needs to store all training sample data at each prediction time, because all m points adjacent to the training sample need to be found when predicting. Secondly, the K-nearest neighbor algorithm is sensitive to dirty data of a sample, and when a model is established, the data needs to be preprocessed to reduce the influence of abnormal values on a result.
And performing correlation analysis on historical data of the sorting backflow abnormity related events and the time sequence data of the finished wrapped logistics products by using a correlation analysis method based on a K-nearest neighbor algorithm, and performing result comparison by using two baseline algorithms including a Pearson correlation method and a J-measure correlation method, wherein the evaluation standard is F-score.
Firstly, historical data of a box sealing machine bar code PLC is collected, wherein the historical data comprises 5S (connection request non-response times, byte number per second, effective load bit rate, server TCP retransmission packet and client side average ACK time delay) and 5E (code scanning, bar code transmission, heartbeat IO timeout, operating system restart and a large number of IO read-write). The collected data is preprocessed.
After the data processing is finished, the K-nearest neighbor algorithm, the Pearson correlation algorithm and the J-measure correlation algorithm are used for verification respectively. The Pearson correlation algorithm is a correlation mining method widely used in time series, and the J-measure correlation algorithm is a method widely used for correlating event data.
Data correlation results:
y-correlation; n-independent
F-score results for three algorithms:
serial number | Algorithm | F-score |
1 | K-nearest neighbor algorithm | 0.8631 |
2 | Pearson correlation | 0.6974 |
3 | J-measure correlation | 0.6148 |
As can be seen from the verification results, the K-nearest neighbor algorithm performs better than the Pearson correlation algorithm and the J-measure correlation algorithm.
The method focuses on the relation between the events and the time sequence data, provides good inspiration for sorting and backflow abnormity diagnosis of the roll-up logistics finished products, and provides good clue correlation in the aspects of helping to perform root analysis and the like. Based on the characteristics of the cigarette production line, the method can also be applied to abnormal correlation analysis of other PLCs.
The above embodiments are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby, and any insubstantial changes and substitutions made by those skilled in the art based on the present invention are within the protection scope of the present invention.
Claims (6)
1. A sorting backflow abnormal relevance analysis method based on a proximity algorithm is characterized by comprising the following steps: the method comprises a K-neighbor algorithm for performing correlation analysis on historical data of sorting backflow abnormity related events and time sequence data of the finished roll logistics, and two baseline algorithms for performing result comparison on analysis results of the K-neighbor algorithm, wherein the analysis method comprises the following steps:
step one, collecting historical data of a bar code PLC of a case sealer, wherein the collected historical data comprises connection request non-response times, byte number per second, effective load bit rate, server TCP retransmission packets, client side average ACK time delay, code scanning, bar code transmission, heartbeat IO timeout, operating system restart and a large number of IO read-write, and preprocessing the collected data;
step two, verifying the processed data by using a K-nearest neighbor algorithm and two baseline algorithms respectively;
and step three, troubleshooting and judging faults of the monitoring system and the automatic control system through the verification result.
2. The sorting reflow anomaly correlation analysis method based on the proximity algorithm as claimed in claim 1, wherein: two of the baseline algorithms include the Pearson correlation algorithm and the J-measure correlation algorithm, with the evaluation criterion being F-score.
3. The sorting reflow anomaly correlation analysis method based on the proximity algorithm as claimed in claim 1, wherein: the K-nearest neighbor algorithm has the following calculation mode: firstly, judging and converting the correlation between the finished product sorting backflow abnormity related event and the time sequence data into two sample problems; second, determine if the two samples are from the same distribution.
4. The sorting reflow anomaly correlation analysis method based on the proximity algorithm according to claim 3, wherein: the way of judging whether the two samples come from the same distribution is as follows: firstly, selecting corresponding n segments of time sequence sample data with length of k before and after occurrence of an event, and using the selected data Is represented by eiRepresenting a certain event, the sample group theta is a series of sample data with the length of k randomly selected on the time sequence, and then, according to the sample group And judging with the distribution result of the sample group theta, wherein when the distribution is different, the event is related to the time sequence data, and when the distribution is the same, the event is not related to the time sequence data.
5. The sorting reflow anomaly correlation analysis method based on the proximity algorithm according to claim 4, wherein: sample set ΓfrontAnd all samples in the sample group theta are used to calculate the distance between the respective samples in the two sample groups by means of the DTW algorithm, some belonging to the sample group ΓfrontOr samples S of a sample group theta, a plurality of nearest neighbor samples m of S are analyzed, and the larger the number of samples belonging to the same sample group as S, the larger the number of samples belonging to the same sample group as S means that the sample group gamma isfrontAnd theta is dividedThe more different, i.e., the more correlated the event and timing data.
6. The sorting reflow anomaly correlation analysis method based on the proximity algorithm as claimed in claim 1, wherein: the number m of nearest neighbor samples is a natural logarithm, the first peak value of an autocorrelation function curve of a time sequence length k is a sequence length, and the whole algorithm is as follows:
Input:Event E=(et,e2,...,en),and Time Series
S=(s1,s2,...,sm),and the sub-series length k.
Output:The direction D,
1 Initialize Гfront and Γrear;
2 Initializeθ;
3 Initialize D=NULL,
5 Test Гfront andθusing Nearest Neighbors Method.
The result is denoted as Df.;
6 Test Гrear andθusing Nearest Neighbors Method.
The result is denoted as Df.;
7 Out put D;
8 Algorithm End.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110002725.4A CN112949677A (en) | 2021-01-04 | 2021-01-04 | Sorting backflow abnormal relevance analysis method based on proximity algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110002725.4A CN112949677A (en) | 2021-01-04 | 2021-01-04 | Sorting backflow abnormal relevance analysis method based on proximity algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112949677A true CN112949677A (en) | 2021-06-11 |
Family
ID=76235170
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110002725.4A Pending CN112949677A (en) | 2021-01-04 | 2021-01-04 | Sorting backflow abnormal relevance analysis method based on proximity algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112949677A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113361939A (en) * | 2021-06-15 | 2021-09-07 | 红云红河烟草(集团)有限责任公司 | Dynamic association method and system for quality and equipment management of wrapping machine type |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015176565A1 (en) * | 2014-05-22 | 2015-11-26 | 袁志贤 | Method for predicting faults in electrical equipment based on multi-dimension time series |
CN110555049A (en) * | 2018-03-30 | 2019-12-10 | 浙江华云信息科技有限公司 | fault reason suspected degree analysis method based on measurement abnormality correlation degree model |
CN111522804A (en) * | 2020-04-15 | 2020-08-11 | 国网江苏省电力有限公司 | Cleaning method for abnormal data of transformer equipment state monitoring |
-
2021
- 2021-01-04 CN CN202110002725.4A patent/CN112949677A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2015176565A1 (en) * | 2014-05-22 | 2015-11-26 | 袁志贤 | Method for predicting faults in electrical equipment based on multi-dimension time series |
CN110555049A (en) * | 2018-03-30 | 2019-12-10 | 浙江华云信息科技有限公司 | fault reason suspected degree analysis method based on measurement abnormality correlation degree model |
CN111522804A (en) * | 2020-04-15 | 2020-08-11 | 国网江苏省电力有限公司 | Cleaning method for abnormal data of transformer equipment state monitoring |
Non-Patent Citations (2)
Title |
---|
CHEN LUO, ET AL.: ""Correlating Events with Time Series for Incident Diagnosis"", 《SIGKDD’14》 * |
程莹莹等: "上市公司高管变更与股价异常波动关联性分析", 《北京信息科技大学学报(自然科学版)》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113361939A (en) * | 2021-06-15 | 2021-09-07 | 红云红河烟草(集团)有限责任公司 | Dynamic association method and system for quality and equipment management of wrapping machine type |
CN113361939B (en) * | 2021-06-15 | 2022-05-20 | 红云红河烟草(集团)有限责任公司 | Dynamic association method and system for quality and equipment management of wrapping machine type |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6137938B2 (en) | Method for detecting anomalies in time series data | |
US20070282545A1 (en) | Probabilistic Stress Wave Analysis System and Method | |
CN101753381A (en) | Method for detecting network attack behaviors | |
CN112949677A (en) | Sorting backflow abnormal relevance analysis method based on proximity algorithm | |
CN116300835B (en) | Automatic detection equipment remote diagnosis system based on Internet | |
CN112711850A (en) | Unit online monitoring method based on big data | |
CN117148045A (en) | Fault studying and judging management system for running state of power distribution network | |
CN111404756A (en) | Fault diagnosis system for communication equipment | |
CN116335925A (en) | Data enhancement-based intelligent regulation and control system for underground coal mine emulsification pump station | |
CN112787984A (en) | Vehicle-mounted network anomaly detection method and system based on correlation analysis | |
CN116593829B (en) | Transmission line hidden danger on-line monitoring system based on data analysis | |
CN109145773A (en) | A kind of failure prediction method of multi-source tendency information fusion | |
CN111538755A (en) | Equipment operation state anomaly detection method based on normalized cross correlation and unit root detection | |
JP6915763B1 (en) | Abnormality diagnosis system and abnormality diagnosis method | |
CN116562599A (en) | Efficient intelligent factory scheduling system based on deep learning | |
CN116614395A (en) | Data transmission abnormity supervision system and method based on artificial intelligence | |
CN110562296A (en) | LKJ (LKJ series-control) based running file data continuous fluctuation abnormity identification method | |
CN114564758A (en) | Management method and system of operation and maintenance data | |
Wu et al. | Early anomaly detection in wind turbine bolts breaking problem—Methodology and application | |
CN113688791A (en) | Wind turbine CMS abnormal data packet identification method | |
CN113435497A (en) | Coal-rock interface multi-sensing information fusion preference identification method based on hardness characteristics | |
CN102476132B (en) | Method for monitoring rolling abnormality on tail end of steel belt | |
CN111814325A (en) | Single variable alarm threshold value optimization method for non-steady operation of coal-fired power generation equipment | |
CN111010393B (en) | Anomaly detection and elimination method for big data cleaning | |
CN117439827B (en) | Network flow big data analysis method |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210611 |
|
WD01 | Invention patent application deemed withdrawn after publication |