CN112148722A - Monitoring data abnormity identification and processing method and system - Google Patents
Monitoring data abnormity identification and processing method and system Download PDFInfo
- Publication number
- CN112148722A CN112148722A CN202011097985.6A CN202011097985A CN112148722A CN 112148722 A CN112148722 A CN 112148722A CN 202011097985 A CN202011097985 A CN 202011097985A CN 112148722 A CN112148722 A CN 112148722A
- Authority
- CN
- China
- Prior art keywords
- data
- value
- updating
- filter
- monitoring
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a method and a system for identifying and processing abnormal monitoring data, which utilize a Kalman filter data prediction principle to predict the monitoring data to be analyzed and judged, preliminarily judge whether the monitoring data is abnormal or not by comparing a predicted value with a monitoring value, carry out secondary correction on the data which is preliminarily judged to be abnormal by manual supervision, and carry out filter updating processing and threshold model updating on a corrected result.
Description
Technical Field
The invention relates to the technical field of abnormal data identification and data cleaning, in particular to a method and a system for identifying and processing monitoring data abnormity.
Background
With the continuous development of sensor technology, more and more data are collected, and powerful support can be provided for related industries by analyzing and deciding the data. However, each piece of data acquired by the device cannot be accurate, and the abnormal data is caused by various reasons, and the abnormal data is used for analysis and decision-making and may make wrong judgment, so that the abnormal judgment and processing of the acquired data items are very important. For processing error data in abnormal data, a commonly used processing method at present is as follows: removing isolated points by means of clustering, regression, binning and the like; noise data which is separated from the distribution can be eliminated through the data characteristic distribution. However, these methods do not consider the time variation of the device, and the recognition accuracy is not high.
Disclosure of Invention
The invention provides a method and a system for identifying and processing monitoring data abnormity, which are used for solving the problems in the prior art.
The technical scheme adopted by the invention is as follows: a monitoring data abnormity identification and processing method comprises the following steps:
step 1: according to the number of data items of the monitoring data, confirming the identification positions corresponding to the distributed data items;
step 2: aiming at each data item, arranging data according to a time ascending sequence, carrying out mean value filtering on a plurality of previous data items, and filtering out obvious abnormal data;
and step 3: aiming at each data item, establishing a respective Kalman filtering model by using the data item processed in the step 2, and feeding back and updating the data item to the current time state in sequence;
and 4, step 4: predicting and outputting the predicted value of each data item at the next moment according to the Kalman filter of the current moment state;
and 5: comparing the predicted value at the next moment with the data item value acquired at the next moment, and judging whether the data item value acquired by the sensor equipment belongs to a suspected abnormal value or not; if the data item is abnormal, setting the identification bit corresponding to the data item to be 1, otherwise, setting the identification bit to be 0;
step 6: reading all suspected abnormal item data;
and 7: displaying the suspected abnormal data and the corresponding historical normal data inquired in the step 6 through a visual interface, and manually checking and correcting the judgment result;
and 8: rewriting the correction result into the record corresponding to the database, updating and replacing the previous judgment result, executing step 9 on the data item judged to be normal, and executing step 11 on the data item with error;
and step 9: regarding the data items which are manually corrected to be normal, taking the part of data as normal monitoring values for updating the monitoring values of the filter model, updating a related parameter matrix of the filter model for predicting the data of the next time, and executing the step 10 after the completion of the updating of the corresponding threshold;
step 10: taking a multiple of a difference value between the current predicted value and the acquired data item value as an updating input, and taking the input data as a monitoring value to update the threshold filter after the threshold filter predicts;
step 11: and performing prediction operation on the threshold filter to match the current time state.
Preferably, step 6 is specifically as follows:
step 6.1, inquiring the record with the identification field not being 0 according to the identification field;
6.2, analyzing the identification field, and obtaining a bit which is not 0 through AND operation;
and 6.3, confirming the corresponding data items according to the abnormal data bits, and inquiring a plurality of data items before the abnormal time.
The invention also discloses a system for identifying and processing the abnormal monitoring data, which comprises
The data exception identification field construction module is used for confirming and distributing identification positions corresponding to all data items according to the number of the data items of the monitoring data;
the data prediction module is used for predicting a predicted value at the next moment;
the suspected abnormal data judgment module is used for reading all suspected abnormal item data;
the auditing and correcting module is used for displaying through a visual interface and manually auditing and correcting the judgment result;
the monitoring filter updating module is used for inputting the data which is manually corrected to be normal into the filter model and updating the monitoring filter;
and the threshold filter updating module updates the data which are corrected to be normal manually after the monitoring filter updating module finishes updating.
Preferably, in the data exception identification field building module, int data type is adopted to identify the exception condition of each data item.
Preferably, in the data anomaly identification field construction module, the generation and analysis method for the identification field is as follows:
and (3) identification field generation: arranging the data items according to the sequence of binary digits from low digits to high digits, determining the position of the binary digit corresponding to each data item, setting the corresponding data digit to be 1 when data of a certain data item is abnormal, and storing int integer result data corresponding to the binary data after all data items are judged and set;
and (3) identification field analysis: the identification field resolution is achieved by an and operation of a binary data bit operation.
Preferably, in the data prediction module, data prediction is realized through kalman filtering, a kalman filter is established for each data item, and a predicted value of each data item at the next time is predicted and output according to the state of the current time of the kalman filter.
Preferably, the data prediction module is further configured, in the data prediction module,
establishing a filter model for the data item of each sensor device, and updating the model through a correct monitoring value for predicting a predicted value at the next moment;
and establishing a filter model for the threshold value of each sensor device, and taking the difference value of the monitoring value and the predicted value as input to predict the change of the threshold value and dynamically adjust the threshold value.
Preferably, in the monitoring filter updating module, the data corrected to be abnormal by manual auditing is not used for filter updating, and for the data corrected to be normal by manual auditing, the part of data is used as a normal monitoring value for monitoring value updating of the filter model, and a related parameter matrix of the filter model is updated for prediction of next data.
Preferably, in the threshold filter updating module, if the current monitoring value is the correct monitoring value, the multiple of the difference value between the current predicted value and the normal monitoring value is used as the updating input; the current monitored value is an erroneous monitored value, and data updating is not performed.
The invention has the beneficial effects that: the invention can realize the recognition and processing of the monitoring data when the error value occurs, and avoid the influence of abnormal data on decision judgment. Meanwhile, the change situation of the equipment in time is considered, the weight of the current normal monitoring value is increased, time correlation is achieved, and the method better conforms to the scenes of change of the monitoring environment, instrument loss and the like.
Drawings
FIG. 1 is a schematic flow chart of a method for recognizing and processing abnormal monitoring data according to the present invention;
fig. 2 is a schematic diagram of the allocation of identification bit values disclosed in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail below with reference to the accompanying drawings, but embodiments of the present invention are not limited thereto.
Example 1:
referring to fig. 1, a method for identifying and processing monitoring data anomalies includes the following steps:
step 1: and confirming the corresponding identification positions of the distributed data items according to the number of the data items of the monitoring data. For example: there are 5 data items, item 1 to item 5. The allocated identification position is shown in fig. 2, and the corresponding int data range is (0-31). In this step, each monitoring instrument data is a data item, and for example, the data collected by the water pressure sensor is a data item.
Step 2: for each data item, the data arranged in ascending order according to time (namely, the data are sorted from morning to evening according to the acquisition time), and the average value filtering is carried out on the previous data (preferably more than 5 data), so as to filter out obvious abnormal data.
And step 3: and (3) aiming at each data item, establishing a respective Kalman filtering model by using the data processed in the step (2), and feeding back and updating the data to the current time state in sequence.
And 4, step 4: and predicting and outputting the predicted value of each data item at the next moment according to the Kalman filter of the current moment state.
And 5: and comparing the predicted value at the next moment with the data item value acquired at the next moment, and judging whether the data item value acquired by the sensor equipment belongs to a suspected abnormal value. If so, the corresponding identification bit of the data item is set to 1, otherwise, the identification bit is set to 0. If the item 4 data and the item 2 data are suspected to be abnormal, the corresponding identification bit is set to be 1, the result identification field value is 00001010, the int value is 12, and the steps are repeated, all monitoring records are traversed and judged, and the identification field values are stored.
Step 6: reading all suspected abnormal item data, specifically comprising the following steps:
and 6.1, inquiring the record with the identification field not being 0 according to the identification field.
And 6.2, analyzing the identification field, obtaining a bit which is not 0 through AND operation, if the item 5 is abnormal, carrying out AND operation on the identification field value and 00010000(int value is 16), if the result is 16, the item 5 is abnormal, otherwise, the item is normal.
And 6.3, confirming the corresponding data items according to the abnormal data bits, and inquiring a plurality of data items before the abnormal time.
And 7: and 6, displaying the suspected abnormal data and the corresponding historical normal data inquired in the step 6 through a visual interface, and manually checking and correcting the judgment result.
And 8: the corrected result is rewritten into the corresponding record of the database, the judgment result before replacement is updated, step 9 is executed for the data item judged to be normal, and step 11 is executed for the data item with error.
And step 9: and for the data item which is manually corrected to be normal, taking the part of data as a normal monitoring value for updating the monitoring value of the filter model, updating a related parameter matrix of the filter model for predicting the data of the next time, and executing the step 10 after the completion of the updating of the corresponding threshold value.
Step 10: and taking the multiple of the difference value between the current predicted value and the normal monitoring value (the normal monitoring value is the collected data item value) as an updating input, if the input data is normal observation-current predicted value | x 2, and after the threshold filter predicts, taking the input data as the monitoring value to update the threshold filter.
Step 11: and performing prediction operation on the threshold filter to match the current time state.
Example 2
A monitoring data abnormity identification and processing system comprises a data abnormity identification field construction module, a data prediction module, a suspected abnormity data judgment module, an audit correction module, a monitoring filter updating module and a threshold filter updating module.
Data exception identification field construction module
Since there may be more than one item of data collected by a sensor device, the entire data may be made redundant if an identification field is established for each field to identify the data. The invention adopts int data type to mark the abnormal condition of each data item, the concrete method is that each bit of int data type under binary data type marks one collected data item, for example, 8 binary data are available for 8 bit int data: 00000000-11111111, each bit of binary data has 0 or 1 two data values, the invention marks 1 as data abnormal, and 0 as data normal, so that one 8-bit int data type can mark 8 data items at most, and commonly used 32-bit int data can mark 32 data items at most, so that one field can basically meet the mark of each data item.
The generation and analysis method for the identification field is as follows:
and (3) identification field generation: and arranging the data items according to the sequence of the binary digits from the low order to the high order, and determining the position of the binary digit corresponding to each data item. And when data of a certain data item is abnormal, setting the corresponding data bit to be 1, and after all data items are judged and set, storing int integer result data corresponding to binary data.
And (3) identification field analysis: the identification field analysis is realized by the AND operation of binary data bit operation, the operation component of the bit operation can only be integer or character data, the bit operation regards the operation object as bit string information composed of binary bits, the specified operation is completed by bit to obtain the result of the bit string information, and the bit operator comprises:
and (bitwise AND), | (bitwise OR), ^ (bitwise XOR), and ^ (bitwise negation).
Bitwise and operation the corresponding bits of the two operational components are calculated bitwise according to the following rules:
0&0=0,0&1=0,1&0=0,1&1=1。
i.e., bits that are both 1, the result is 1, otherwise the result is 0.
For example, if an identification data binary data is as follows:
00000111
and (3) when judging whether the second data item is abnormal, carrying out AND operation on the second data item and target data with only second bit data being 1(00000010), if the operation result is the same as the target data, judging that the corresponding bit data is abnormal, otherwise, judging that the data is normal.
Data prediction module
The invention realizes data prediction through Kalman filtering, and Kalman filtering (Kalman filtering) is an algorithm which utilizes a linear system state equation, outputs observation data through system input and output and performs optimal estimation on a system state. The optimal estimation can also be seen as a filtering process, since the observed data includes the effects of noise and interference in the system. According to the method, a Kalman filter is established for each data item, and the predicted value of each data item at the next moment is predicted and output according to the Kalman filter in the current moment state.
According to the method, a filter model is established for the data item of each sensor device, and the model is updated through a correct monitoring value so as to predict the predicted value at the next moment.
According to the invention, a filter model is also established for the threshold value of each sensor device, and the difference value between the monitoring value and the predicted value is used as input for predicting the change of the threshold value and dynamically adjusting the threshold value.
Suspected abnormal data judgment module
And comparing the predicted value of the next moment obtained by the data prediction module with the data item value collected at the next moment, and judging whether the data item value collected by the equipment belongs to a suspected abnormal value. The specific judgment method is that the absolute value of the monitoring value (namely the data item value collected at the next moment) minus the predicted value is less than a certain threshold, the threshold value thread is initially set to be the mean value of the previous m pieces of normal data by 15%, and a threshold filter is established:
l monitoring value-predicted value | < threshold;
the threshold value is (monitor value 1+ monitor value 2+. + monitor value m)/m 15%.
Auditing and correcting module
The auditing and correcting module is used for displaying through a visual interface, representing and storing suspected abnormal data automatically recognized by the system through a data abnormal identification field, inquiring a plurality of pieces of data before the abnormal data moment aiming at the suspected abnormal data when personnel audits and corrects, manually confirming whether the suspected abnormal data is abnormal according to historical data, confirming or modifying an abnormal data identification, and finally storing the identification.
Filter updating module
And for the data which is artificially checked and corrected to be abnormal, the data is not used for updating the filter, and for the data which is artificially corrected to be normal, the data is used as a normal monitoring value for updating the monitoring value of the filter model, and the related parameter matrix of the filter model is updated for predicting the data at the next time.
Threshold filter update module
The threshold filter is updated in two situations, one is that the current monitoring value is the correct monitoring value, and the situation takes the multiple of the difference value between the current predicted value and the normal monitoring value as the updating input, for example, the input data is | normal observation-current predicted value | x 2. Another case is where the current monitored value is a false monitored value, in which case no data update is performed and only filter prediction is performed to match the state at the current time.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (9)
1. A method for recognizing and processing monitoring data abnormity is characterized by comprising the following steps:
step 1: according to the number of data items of the monitoring data, confirming the identification positions corresponding to the distributed data items;
step 2: aiming at each data item, arranging data according to a time ascending sequence, carrying out mean value filtering on a plurality of previous data items, and filtering out obvious abnormal data;
and step 3: aiming at each data item, establishing a respective Kalman filtering model by using the data item processed in the step 2, and feeding back and updating the data item to the current time state in sequence;
and 4, step 4: predicting and outputting the predicted value of each data item at the next moment according to the Kalman filter of the current moment state;
and 5: comparing the predicted value at the next moment with the data item value acquired at the next moment, and judging whether the data item value acquired by the sensor equipment belongs to a suspected abnormal value or not; if the data item is abnormal, setting the identification bit corresponding to the data item to be 1, otherwise, setting the identification bit to be 0;
step 6: reading all suspected abnormal item data;
and 7: displaying the suspected abnormal data and the corresponding historical normal data inquired in the step 6 through a visual interface, and manually checking and correcting the judgment result;
and 8: rewriting the correction result into the record corresponding to the database, updating and replacing the previous judgment result, executing step 9 on the data item judged to be normal, and executing step 11 on the data item with error;
and step 9: regarding the data items which are manually corrected to be normal, taking the part of data as normal monitoring values for updating the monitoring values of the filter model, updating a related parameter matrix of the filter model for predicting the data of the next time, and executing the step 10 after the completion of the updating of the corresponding threshold;
step 10: taking a multiple of a difference value between the current predicted value and the acquired data item value as an updating input, and taking the input data as a monitoring value to update the threshold filter after the threshold filter predicts;
step 11: and performing prediction operation on the threshold filter to match the current time state.
2. The method for identifying and processing the abnormality of the monitored data according to claim 1, wherein the step 6 comprises the following steps:
step 6.1, inquiring the record with the identification field not being 0 according to the identification field;
6.2, analyzing the identification field, and obtaining a bit which is not 0 through AND operation;
and 6.3, confirming the corresponding data items according to the abnormal data bits, and inquiring a plurality of data items before the abnormal time.
3. A system for recognizing and processing the abnormity of monitored data is characterized by comprising
The data exception identification field construction module is used for confirming and distributing identification positions corresponding to all data items according to the number of the data items of the monitoring data;
the data prediction module is used for predicting a predicted value at the next moment;
the suspected abnormal data judgment module is used for reading all suspected abnormal item data;
the auditing and correcting module is used for displaying through a visual interface and manually auditing and correcting the judgment result;
the monitoring filter updating module is used for inputting the data which is manually corrected to be normal into the filter model and updating the monitoring filter;
and the threshold filter updating module updates the data which are corrected to be normal manually after the monitoring filter updating module finishes updating.
4. A system for recognizing and processing abnormality of monitored data according to claim 3, wherein said data abnormality identification field constructing module identifies abnormality of each data item by int data type.
5. A system for identifying and processing anomalies in monitored data as claimed in claim 4,
in the data exception identification field construction module, the generation and analysis method for the identification field is as follows:
and (3) identification field generation: arranging the data items according to the sequence of binary digits from low digits to high digits, determining the position of the binary digit corresponding to each data item, setting the corresponding data digit to be 1 when data of a certain data item is abnormal, and storing int integer result data corresponding to the binary data after all data items are judged and set;
and (3) identification field analysis: the identification field resolution is achieved by an and operation of a binary data bit operation.
6. The system according to claim 3, wherein in the data prediction module, data prediction is implemented through Kalman filtering, a Kalman filter is established for each data item, and a prediction value of each data item at the next time is predicted and output according to the Kalman filter in the current time state.
7. A system for identifying and processing anomalies in monitored data as claimed in claim 6 wherein, in the data prediction module,
establishing a filter model for the data item of each sensor device, and updating the model through a correct monitoring value for predicting a predicted value at the next moment;
and establishing a filter model for the threshold value of each sensor device, and taking the difference value of the monitoring value and the predicted value as input to predict the change of the threshold value and dynamically adjust the threshold value.
8. A system for identifying and processing abnormal monitoring data according to claim 3, wherein in the monitoring filter updating module, the data corrected to be abnormal by manual auditing is not used for updating the filter, and for the data corrected to be normal by manual auditing, the part of data is used as normal monitoring values for updating the monitoring values of the filter model, and the relevant parameter matrix of the filter model is updated for predicting the next data.
9. The system of claim 8, wherein in the threshold filter updating module, if the current monitored value is the correct monitored value, the multiple of the difference between the current predicted value and the normal monitored value is used as the updating input; the current monitored value is an erroneous monitored value, and data updating is not performed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011097985.6A CN112148722B (en) | 2020-10-14 | 2020-10-14 | Monitoring data abnormity identification and processing method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011097985.6A CN112148722B (en) | 2020-10-14 | 2020-10-14 | Monitoring data abnormity identification and processing method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112148722A true CN112148722A (en) | 2020-12-29 |
CN112148722B CN112148722B (en) | 2022-06-03 |
Family
ID=73953084
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011097985.6A Active CN112148722B (en) | 2020-10-14 | 2020-10-14 | Monitoring data abnormity identification and processing method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112148722B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112395468A (en) * | 2020-11-27 | 2021-02-23 | 浙江诺诺网络科技有限公司 | Number management method and device, electronic equipment and storage medium |
CN114491383A (en) * | 2022-04-15 | 2022-05-13 | 江西飞尚科技有限公司 | Abnormal data processing method and system for bridge monitoring |
CN114993353A (en) * | 2022-05-18 | 2022-09-02 | 成都市美幻科技有限公司 | Self-adaptive filtering method, device, equipment and system for multiple disaster monitoring equipment |
CN116189023A (en) * | 2023-04-28 | 2023-05-30 | 成都市环境应急指挥保障中心 | Method and system for realizing environment emergency monitoring based on unmanned aerial vehicle |
CN116502156A (en) * | 2023-06-30 | 2023-07-28 | 中国电力科学研究院有限公司 | Intelligent identification method and system for light CT abnormal state of converter station |
WO2024001105A1 (en) * | 2022-06-27 | 2024-01-04 | Oppo广东移动通信有限公司 | Communication detection method and apparatus, device, system, and storage medium |
CN117375147A (en) * | 2023-09-25 | 2024-01-09 | 南方电网调峰调频(广东)储能科技有限公司 | Safety monitoring early warning and operation management method and system for energy storage power station |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100004949A1 (en) * | 2008-07-03 | 2010-01-07 | Impulse Monitoring, Inc. | Method, system, and computer program product for receiving, extracting, and translating intraoperative neurophysiologic monitoring (ionm) data from multiple device types |
CN103793765A (en) * | 2014-02-12 | 2014-05-14 | 航天东方红卫星有限公司 | Satellite telemetering data predicting method based on Kalman smoothing |
CN109542740A (en) * | 2017-09-22 | 2019-03-29 | 阿里巴巴集团控股有限公司 | Method for detecting abnormality and device |
CN109947747A (en) * | 2017-12-01 | 2019-06-28 | 广州明领基因科技有限公司 | Big data exceptional value method for cleaning based on Outlier Detection |
CN110602723A (en) * | 2019-08-27 | 2019-12-20 | 华侨大学 | Two-stage bidirectional prediction data acquisition method based on underwater edge equipment |
CN110880984A (en) * | 2019-10-18 | 2020-03-13 | 平安科技(深圳)有限公司 | Model-based flow anomaly monitoring method, device, equipment and storage medium |
CN111080484A (en) * | 2019-12-21 | 2020-04-28 | 国网山东省电力公司泰安供电公司 | Method and device for monitoring abnormal data of power distribution network |
CN111193495A (en) * | 2019-12-12 | 2020-05-22 | 浙江工业大学 | Filtering processing method for workpiece alignment data |
-
2020
- 2020-10-14 CN CN202011097985.6A patent/CN112148722B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100004949A1 (en) * | 2008-07-03 | 2010-01-07 | Impulse Monitoring, Inc. | Method, system, and computer program product for receiving, extracting, and translating intraoperative neurophysiologic monitoring (ionm) data from multiple device types |
CN103793765A (en) * | 2014-02-12 | 2014-05-14 | 航天东方红卫星有限公司 | Satellite telemetering data predicting method based on Kalman smoothing |
CN109542740A (en) * | 2017-09-22 | 2019-03-29 | 阿里巴巴集团控股有限公司 | Method for detecting abnormality and device |
CN109947747A (en) * | 2017-12-01 | 2019-06-28 | 广州明领基因科技有限公司 | Big data exceptional value method for cleaning based on Outlier Detection |
CN110602723A (en) * | 2019-08-27 | 2019-12-20 | 华侨大学 | Two-stage bidirectional prediction data acquisition method based on underwater edge equipment |
CN110880984A (en) * | 2019-10-18 | 2020-03-13 | 平安科技(深圳)有限公司 | Model-based flow anomaly monitoring method, device, equipment and storage medium |
CN111193495A (en) * | 2019-12-12 | 2020-05-22 | 浙江工业大学 | Filtering processing method for workpiece alignment data |
CN111080484A (en) * | 2019-12-21 | 2020-04-28 | 国网山东省电力公司泰安供电公司 | Method and device for monitoring abnormal data of power distribution network |
Non-Patent Citations (2)
Title |
---|
LIQING DI 等: "On-line Monitoring of Batch Processes Using Kalman Filter and Multivariate Statistical Methods", 《2006 6TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION》 * |
褚凡武: "水电机组运行状态协同监测及异常检测方法研究", 《中国优秀博硕士学位论文全文数据库(博士)工程科技Ⅱ辑》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112395468A (en) * | 2020-11-27 | 2021-02-23 | 浙江诺诺网络科技有限公司 | Number management method and device, electronic equipment and storage medium |
CN114491383A (en) * | 2022-04-15 | 2022-05-13 | 江西飞尚科技有限公司 | Abnormal data processing method and system for bridge monitoring |
CN114491383B (en) * | 2022-04-15 | 2022-09-16 | 江西飞尚科技有限公司 | Abnormal data processing method and system for bridge monitoring |
CN114993353A (en) * | 2022-05-18 | 2022-09-02 | 成都市美幻科技有限公司 | Self-adaptive filtering method, device, equipment and system for multiple disaster monitoring equipment |
WO2024001105A1 (en) * | 2022-06-27 | 2024-01-04 | Oppo广东移动通信有限公司 | Communication detection method and apparatus, device, system, and storage medium |
CN116189023A (en) * | 2023-04-28 | 2023-05-30 | 成都市环境应急指挥保障中心 | Method and system for realizing environment emergency monitoring based on unmanned aerial vehicle |
CN116189023B (en) * | 2023-04-28 | 2023-07-11 | 成都市环境应急指挥保障中心 | Method and system for realizing environment emergency monitoring based on unmanned aerial vehicle |
CN116502156A (en) * | 2023-06-30 | 2023-07-28 | 中国电力科学研究院有限公司 | Intelligent identification method and system for light CT abnormal state of converter station |
CN116502156B (en) * | 2023-06-30 | 2023-09-08 | 中国电力科学研究院有限公司 | Intelligent identification method and system for light CT abnormal state of converter station |
CN117375147A (en) * | 2023-09-25 | 2024-01-09 | 南方电网调峰调频(广东)储能科技有限公司 | Safety monitoring early warning and operation management method and system for energy storage power station |
CN117375147B (en) * | 2023-09-25 | 2024-03-15 | 南方电网调峰调频(广东)储能科技有限公司 | Safety monitoring early warning and operation management method and system for energy storage power station |
Also Published As
Publication number | Publication date |
---|---|
CN112148722B (en) | 2022-06-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112148722B (en) | Monitoring data abnormity identification and processing method and system | |
US8676553B2 (en) | Apparatus abnormality diagnosis method and system | |
US11403160B2 (en) | Fault predicting system and fault prediction method | |
CN111475804A (en) | Alarm prediction method and system | |
CN109255440B (en) | Method for predictive maintenance of power production equipment based on Recurrent Neural Networks (RNN) | |
EP2442288A1 (en) | Device abnormality monitoring method and system | |
CN109420932B (en) | Abnormality detection device | |
US8621305B2 (en) | Methods systems and apparatus for determining whether built-in-test fault codes are indicative of an actual fault condition or a false alarm | |
CN115828170B (en) | Fault detection method based on electronic control data of tunneling machine | |
CN111105082A (en) | Workpiece quality prediction model construction method and prediction method based on machine learning | |
CN115392812B (en) | Abnormal root cause positioning method, device, equipment and medium | |
WO2021199227A1 (en) | Error cause estimation device and estimation method | |
WO2020166236A1 (en) | Work efficiency evaluating method, work efficiency evaluating device, and program | |
CN115793552A (en) | Electronic gas production monitoring method and system based on data processing | |
EP3916502A1 (en) | Operation evaluation device, operation evaluation method, and program | |
CN116992391A (en) | Hard carbon process environment-friendly monitoring data acquisition and processing method | |
US20080177399A1 (en) | Method of process trend matching for identification of process variable | |
CN110781206A (en) | Method for predicting whether electric energy meter in operation fails or not by learning meter-dismantling and returning failure characteristic rule | |
KR20200023882A (en) | Processing method of process data of smart factory | |
CN117574292A (en) | Data fault detection method and system | |
CN116383606A (en) | Constant-current temperature acquisition method and system for distributed medical equipment | |
JP2020187516A (en) | Abnormality diagnosis device and method | |
CN112100919B (en) | Rolling bearing residual life prediction method based on RE-CF-EKF algorithm | |
CN113836539A (en) | Power engineering control system leak full-flow disposal system and method based on precise test | |
CN112948203B (en) | Elevator intelligent inspection method based on big data |
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 |