CN109558590B - Method for positioning key fault device based on spacecraft remote measurement parameter word segmentation - Google Patents
Method for positioning key fault device based on spacecraft remote measurement parameter word segmentation Download PDFInfo
- Publication number
- CN109558590B CN109558590B CN201811405302.1A CN201811405302A CN109558590B CN 109558590 B CN109558590 B CN 109558590B CN 201811405302 A CN201811405302 A CN 201811405302A CN 109558590 B CN109558590 B CN 109558590B
- Authority
- CN
- China
- Prior art keywords
- paraname
- parameter
- type
- word segmentation
- typepart
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Abstract
The invention provides a key fault device positioning method based on spacecraft telemetering parameter segmentation, which comprises the steps of dividing each abnormal telemetering parameter into four parts of a device name, a device description, a detection type and a supplementary description, and clustering and counting the device names and the detection type parts of all abnormal telemetering parameters to obtain a frequent item set as a key fault device. According to the method, only a few detection type keywords are defined, so that the analysis efficiency and accuracy of personnel on the spacecraft alarm information are improved, the personnel can clearly describe abnormal phenomena, and the analysis and positioning of key fault devices can be realized.
Description
Technical Field
The invention relates to a method for positioning a key fault device of a spacecraft, belonging to the field of spacecraft fault diagnosis and analysis.
Background
Along with the continuous development of aerospace technology, the function of spacecraft is becoming more complicated, and novel devices are continuously increased, and corresponding monitoring points are required to be set simultaneously, so that the device state is acquired, analyzed and judged, the device abnormality can be timely found, and the on-orbit safe and stable operation of the spacecraft and the smooth development of tasks are effectively ensured. The number of the telemetering parameters of the spacecraft reaches ten thousand levels, the coupling degree between devices is high, and dozens or even hundreds of alarm information of the relevant telemetering parameters can be generated when the attitude or the energy is abnormal.
At present, key fault devices are analyzed and positioned in a large amount of alarm information and manually completed by a system analyst, so that the problems of long time consumption and dependence on experience exist, and the requirements on timeliness and accuracy of spacecraft abnormal positioning treatment cannot be met.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a key fault device positioning method based on spacecraft telemetering parameter word segmentation, which is used for performing word segmentation processing on abnormal telemetering parameters, performing cluster statistics on all telemetering parameter word segmentation results to obtain frequent abnormal devices in alarm information, completing the positioning of key fault devices and improving the accuracy and efficiency of the positioning of fault devices.
The technical scheme adopted by the invention for solving the technical problem comprises the following steps:
step 1, collecting fault alarm information, wherein a single piece of fault alarm information is { paraName: msg }, the paraName is a parameter name of a fault telemetering parameter, the msg is an alarm type, and a value is an upper limit or a lower limit; all alarm information composition set detectionSet = { paraName = 1 :msg 1 ,paraName 2 :msg 2 ,…,paraName n :msg n N pieces of alarm information are counted;
step 2, defining the detection type needing attention as a dictionary sequence, namely, dct = [ type 1 ,type 2 ,…,type m ]M detection types;
and 4, counting and sequencing word segmentation results, and determining frequent fault devices and frequent fault types.
The type of telemetry parameter detection includes, but is not limited to, power, temperature, voltage, current, error, number, and status.
The dictionary sequence ditt = [ type = [ ] 1 ,type 2 ,…,type m ]The detection types are arranged according to the descending order of the length of the character string.
In the step 3, the character string of the name of the single telemetering parameter is paraName i ＝[a 1 ,a 2 ,..,a pL ]I is more than or equal to 1 and less than or equal to n, and pL is the length of the character string; search for paraName i Whether or not it contains an item in the type dictionary, if it contains type j J is more than or equal to 1 and less than or equal to m, in paraName i Has a start position of s typej End position is e typej Then the parameter paraName i Detecting type participle part typePart i ＝type j Parameter paraName i Device name and device description participle part elementary part of (1) i ＝paraName i [1:s typej ]Parameter paraName i Is used for describing the participle part detailPart i ＝paraName i [e typej :pL](ii) a typePart if it does not contain any type in the dictionary i ＝””，elementPart i ＝paraName i ，detailPart i ＝””。
In the step 3, the word segmentation of the device name and the device description comprises the following steps: 1) Setting a current segmentation starting position cutStartPos =0, and setting a current segmentation ending position cutEndPos = cutStartPos +1; 2) Define the participle variable 1 as cutWord 1 ＝elementPart i [cutStartPos:cutEndPos–1]The participle variable 2 is cutWord 2 ＝elementPart i [cutStartPos:cutEndPos](ii) a 3) Defining searchResult 1 Looking up cutWord for all parameter names in detectionSet 1 Starting a vector sequence formed by positions, if a certain parameter does not appear in cutWord 1 The value is-1; 4) Defining searchResult 2 Looking up cutWord for all parameter names in detectionSet 2 Starting a vector sequence formed by positions, if a certain parameter does not appear in cutWord 2 The value is-1; 5) If searchResult 1 And searchrresult 2 Is totally produced fromIf the same, updating the value of cutEndPos and adding 1, and returning to 2) repeating the process until the cutEndPos is larger than s typej (ii) a 6) If searchResult 1 Not equal to searchResult 2 Obtaining the parameter paraName i Device name word segmentation result elementNamePart i ＝elementPart i [cutStartPos:cutEndPos-1]Parameter paraName i The device description participle result elementDetailPart of (1) i ＝elementPart i [cutEndPos:s typej ]。
In the step 3, the parameters paraName are telemetered i The word segmentation result is [ elementNamePart ] i ,elementDetailPart i ,typePart i ,detailPart i ]Performing word segmentation on all the telemetry parameters in the detectionSet to obtain a word segmentation result sequence partsResult = [ [ elementNamePart [ ] 1 ,…,elementNamePart n ],[elementDetailPart 1 ,…,elementDetailPart n ],[typePart 1 ,…,typePart n ],[detailPart 1 ,…,detailPart n ]]。
In the step 4, the word segmentation result sequence partsResult [ elementNamePart 1 ,…,elementNamePart n ]Counting all the appeared items, and arranging the items in a descending order according to the counting number to obtain a frequent fault device statistical set partsSet = { elementNamePart = 1 :eCount 1 ,…,elementNamePart p :eCount p }; the part in the partsResult [ typePart 1 ,…,typePart n ]Counting all the appeared items, and sorting the items in descending order according to the counted number to obtain a frequent fault type statistic set typeSet = { typePart = 1 :tCount 1 ,…,typePart q :tCount q }。
The invention has the beneficial effects that: the method overcomes the defects that the prior probability-based Chinese word segmentation method has poor recognition precision of common words depending on the combination of specific parameter sets when segmenting the telemetering parameters, and needs to define a large number of keywords when segmenting the telemetering parameters by the dictionary-based Chinese word segmentation method. Meanwhile, through word segmentation and classification statistical processing of a large number of fault alarm remote measurement parameters, the analysis efficiency and accuracy of the spacecraft alarm information by personnel are improved. The listed frequent fault device set and frequent fault type set can help personnel to effectively locate the key fault device and improve the fault locating capability.
Drawings
FIG. 1 is a flow chart of the present invention;
fig. 2 is a schematic diagram of an application case of the present invention.
Detailed Description
The present invention will be further described with reference to the following drawings and examples, which include, but are not limited to, the following examples.
The invention provides a key fault device positioning method based on spacecraft telemetering parameter segmentation, which comprises the steps of dividing each abnormal telemetering parameter into four parts of a device name, a device description, a detection type and a supplementary description, and clustering and counting the device names and the detection type parts of all abnormal telemetering parameters to obtain a frequent item set as a key fault device. The method is not only beneficial to clearly and clearly describing the abnormal phenomenon by personnel, but also beneficial to realizing the analysis and positioning of the key fault device.
As shown in fig. 1, the present invention specifically includes the following steps:
the first step is as follows: the fault alarm information is collected and a detection type dictionary is defined;
the second step is that: segmenting the fault telemetry parameters by word segmentation;
the third step: statistical analysis of critical failure devices.
The first step described above specifically includes the steps of:
step 1.1: and collecting fault alarm information. The single alarm message is { paraName: msg }, wherein paraName is the parameter name of the fault remote-measuring parameter, msg is the alarm type, and the value is super An upper limit or an upper limit. All alarm information composition set detectionSet = { paraName = 1 :msg 1 ,paraName 2 :msg 2 ,…,paraName n :msg n N pieces of alarm information.
Step 1.2: a definition of a type dictionary is detected. The detection types of the telemetry parameters, including "power", "temperature", "voltage", "current", "error", "number", "status", are indicated in the telemetry parameter names, and the detection type to be concerned is defined as dictionary sequence dit = [ type ] 1 ,type 2 ,…,type m ]And m detection types are arranged in descending order of the length of the character string.
The second step as described above comprises the steps of:
and performing word segmentation on each telemetering parameter by utilizing the alarm telemetering parameter information and the detection type dictionary. Each telemetering parameter is segmented according to the 'device name, [ device description ], [ detection type ], [ supplementary description ],' the last three parts are all selectable items, namely each telemetering parameter must contain the device name, and the last three items are selected. The method comprises the following specific steps:
step 2.1: and performing word segmentation on the detection type and the supplementary description part in the single telemetry parameter. The string for a single telemetry parameter name is paraName i ＝[a 1 ,a 2 ,..,a pL ]I is more than or equal to 1 and less than or equal to n, and pL is the length of the character string. Search for paraName i Whether it contains a certain item in the type dictionary, if it contains type j J is more than or equal to 1 and less than or equal to m, in paraName i Has a start position s typej End position is e typej Then the parameter paraName i Part typePart of the detection type participle i ＝type j Parameter paraName i Device name and device description participle part elementary part of (1) i ＝paraName i [1:s typej ]Parameter paraName i Is used for describing the participle part detailPart i ＝paraName i [e typej :pL](ii) a typePart if it does not contain any type in the dictionary i ＝””，elementPart i ＝paraName i ，detailPart i ＝””。
Step 2.2: and segmenting the device name and the device description part in the single telemetry parameter by word segmentation. 1) Setting a current segmentation starting position cutStartPos =0, and setting a current segmentation ending position cutEndPos = cutStartPos +1; 2) The segmentation word variable 1 is cutWord 1 ＝elementPart i [cutStartPos:cutEndPos–1]The segmentation variable 2 is cutWord 2 ＝elementPart i [cutStartPos:cutEndPos]；3)searchResult 1 Looking up cutWord for all parameter names in detectionSet 1 Starting a vector sequence formed at a position, if a certain parameter does not appear cutWord 1 The value is-1; 4) searchResult 2 Looking up cutWord for all parameter names in detectionSet 2 Starting a vector sequence formed at a position, if a certain parameter does not appear cutWord 2 The value is-1; 5) If searchrresult 1 And searchrresult 2 Identical, update cutEndPos = cutEndPos +1, return 2) repeat the above process until cutEndPos is greater than s typej (ii) a 6) If searchrresult 1 Not equal to searchResult 2 Obtaining the parameter paraName i Device name word segmentation result elementNamePart i ＝elementPart i [cutStartPos:cutEndPos-1]Parameter paraName i The device description participle result elementDetailPart of (1) i ＝elementPart i [cutEndPos:s typej ]。
Step 2.3: and collecting the word segmentation result of the telemetry parameters. Telemetric parameter paraName i The word segmentation result is [ elementNamePart ] i ,elementDetailPart i ,typePart i ,detailPart i ]Performing word segmentation on all the telemetry parameters in the detectionSet to obtain a word segmentation result sequence partsResult = [ [ elementNamePart [ ] 1 ,…,elementNamePart n ],[elementDetailPart 1 ,…,elementDetailPart n ],[typePart 1 ,…,typePart n ],[detailPart 1 ,…,detailPart n ]]。
The third step as described above includes the steps of:
and performing statistical analysis on the frequent alarm device by using the word segmentation result calculated in the previous step, wherein the statistical analysis comprises the following specific steps:
step 3.1: and counting frequently-failed devices. The resulting sequence of the word segmentation is placed in partsResult [ elementNamePart ] 1 ,…,elementNamePart n ]Counting all the appeared items, and arranging the items in descending order according to the counting number to obtain a set partsSet = { elementNamePart = 1 :eCount 1 ,…,elementNamePart p :eCount p }。
Step 3.2: and counting frequent fault types. The part in the partsResult [ typePart 1 ,…,typePart n ]Counting all the appeared items, and arranging the items in descending order according to the counted number to obtain a detection type participle set typesSet = { typePart = 1 :tCount 1 ,…,typePart q :tCount q }。
Referring to fig. 2, the application case of the present invention is divided into three main steps:
1) And (4) acquiring fault alarm information and defining a detection type dictionary. In the first step of fig. 2, 8 items of fault alarm information are listed, and in practical applications, the number of alarms is much larger than this, and this example is only illustrated as a flow of the method. The first step also requires the user to define a type dictionary, which in fig. 2 defines typical telemetry parameter types for temperature, voltage, current, etc.
2) And segmenting the fault telemetry parameters by word segmentation. And completing the word segmentation of the detection type and the supplementary description part in the telemetry parameter name through a type dictionary, and then completing the word segmentation of the device name and the device description part. The word segmentation results for the 8 alarm telemetry parameters are shown in the second step of fig. 2, with use of "separation" between each word segmentation component.
3) Statistical analysis of critical failure devices. And respectively counting the device names and the detection type parts in all the fault telemetering parameters to obtain a frequent fault device set and a frequent fault type set, wherein as shown in the third step in fig. 2, the key devices of 8 items of alarm information are a south storage battery pack and a UHF power amplifier, and the main fault type is temperature.
In conclusion, the invention provides a key fault device positioning method based on spacecraft telemetry parameter word segmentation. The problem of current under a large amount of fault alarm information circumstances, the efficiency that personnel manual analysis key trouble device leads to is not high, the accuracy is difficult to the technique of guaranteeing is solved.
Claims (1)
1. A key fault device positioning method based on spacecraft telemetry parameter word segmentation is characterized by comprising the following steps:
step 1, collecting fault alarm information, wherein a single piece of fault alarm information is { paraName: msg }, the paraName is a parameter name of a fault telemetering parameter, the msg is an alarm type, and a value is an upper limit or a lower limit; all alarm information composition set detectionSet = { paraName = 1 :msg 1 ,paraName 2 :msg 2 ,…,paraName n :msg n N pieces of alarm information are counted;
the detection types of the telemetry parameters comprise power, temperature, voltage, current, error, times and states;
step 2, defining the detection type needing attention as a dictionary sequence, namely, fact = [ type = 1 ,type 2 ,…,type m ]M detection types;
the dictionary sequence ditt = [ type = [ ] 1 ,type 2 ,…,type m ]The detection types are arranged according to the descending order of the length of the character string;
step 3, utilizing an alarm telemetering parameter information and a detection type dictionary to perform word segmentation on each telemetering parameter according to the 'device name, device description, detection type and supplementary description', wherein each telemetering parameter at least comprises a device name, firstly distinguishing the detection type and the supplementary description, and then distinguishing the device name and the device description;
in the step 3, the character string of the name of the single telemetering parameter is paraName i ＝[a 1 ,a 2 ,..,a pL ]I is more than or equal to 1 and less than or equal to n, and pL is the length of the character string; search for paraName i Whether it contains a certain item in the type dictionary, if it contains type j J is more than or equal to 1 and less than or equal to m, in paraName i Start of (1)Position s typej End position is e typej Then the parameter paraName i Detecting type participle part typePart i ＝type j Parameter paraName i Part of the device name and device description participle i ＝paraName i [1:s typej ]Parameter paraName i Is used for describing the participle part detailPart i ＝paraName i [e typej :pL](ii) a typePart if it does not contain any type in the dictionary i ＝””，elementPart i ＝paraName i ，detailPart i ＝””；
In the step 3, the word segmentation of the device name and the device description comprises the following steps: 1) Setting a current segmentation starting position cutStartPos =0, and setting a current segmentation ending position cutEndPos = cutStartPos +1; 2) Define the participle variable 1 as cutWord 1 ＝elementPart i [cutStartPos:cutEndPos–1]The participle variable 2 is cutWord 2 ＝elementPart i [cutStartPos:cutEndPos](ii) a 3) Defining searchResult 1 Looking up cutWord for all parameter names in detectionSet 1 Starting a vector sequence formed at a position, if a certain parameter does not appear cutWord 1 The value is-1; 4) Defining searchResult 2 Looking up cutWord for all parameter names in detectionSet 2 Starting a vector sequence formed at a position, if a certain parameter does not appear cutWord 2 The value is-1; 5) If searchrresult 1 And searchResult 2 If the two are completely the same, updating the cutEndPos value, adding 1, returning to 2) repeating the process until the cutEndPos is larger than s typej (ii) a 6) If searchrresult 1 Is not equal to searchResult 2 Obtaining the parameter paraName i Device name word segmentation result elementNamePart of i ＝elementPart i [cutStartPos:cutEndPos-1]Parameter paraName i The device description participle result elementDetailPart of i ＝elementPart i [cutEndPos:s typej ]；
In the step 3, the parameter paraName is telemetered i The word segmentation result is [ elementNamePart ] i ,elementDetailPart i ,typePart i ,detailPart i ]Performing word segmentation on all the telemetry parameters in the detectionSet to obtain a word segmentation result sequence partsResult = [ [ elementNamePart [ ] 1 ,…,elementNamePart n ],[elementDetailPart 1 ,…,elementDetailPart n ],[typePart 1 ,…,typePart n ],[detailPart 1 ,…,detailPart n ]]；
Step 4, counting and sequencing word segmentation results, and determining frequent fault devices and frequent fault types;
in the step 4, the word segmentation result sequence partsResult [ elementNamePart 1 ,…,elementNamePart n ]Counting all the appeared items, and arranging the items in a descending order according to the number of the counted items to obtain a frequent fault device statistical set partsSet = { elementNamePart = [ ({ elementNamePart) } 1 :eCount 1 ,…,elementNamePart p :eCount p }; will make [ typePart ] in partResult 1 ,…,typePart n ]Counting all the appeared items, and arranging the items in a descending order according to the counted number to obtain a frequent fault type statistical set typesSet = { typePart = 1 :tCount 1 ,…,typePart q :tCount q }。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811405302.1A CN109558590B (en) | 2018-11-23 | 2018-11-23 | Method for positioning key fault device based on spacecraft remote measurement parameter word segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811405302.1A CN109558590B (en) | 2018-11-23 | 2018-11-23 | Method for positioning key fault device based on spacecraft remote measurement parameter word segmentation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109558590A CN109558590A (en) | 2019-04-02 |
| CN109558590B true CN109558590B (en) | 2022-11-15 |
Family
ID=65867292
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811405302.1A Active CN109558590B (en) | 2018-11-23 | 2018-11-23 | Method for positioning key fault device based on spacecraft remote measurement parameter word segmentation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109558590B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1601969A (en) * | 2003-09-24 | 2005-03-30 | 华为技术有限公司 | Information analytical method for facility configuration |
| CN103699118A (en) * | 2013-12-18 | 2014-04-02 | 北京航天测控技术有限公司 | Method and device for analyzing abnormal state of spacecraft in operating process |
| CN106647695A (en) * | 2016-12-05 | 2017-05-10 | 航天恒星科技有限公司 | Method and system for monitoring running state of on-orbit spacecraft |
| CN107247613A (en) * | 2017-04-25 | 2017-10-13 | 北京航天飞行控制中心 | Sentence analytic method and sentence resolver |
| CN107525686A (en) * | 2017-08-21 | 2017-12-29 | 中国人民解放军63789部队 | A kind of more telemetry parameter analysis window division methods of spacecraft |
| CN108478926A (en) * | 2018-05-02 | 2018-09-04 | 清华大学深圳研究生院 | The method that the wearable device of body temperature can be automatically adjusted and adjust body temperature |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107220232B (en) * | 2017-04-06 | 2021-06-11 | 北京百度网讯科技有限公司 | Keyword extraction method and device based on artificial intelligence, equipment and readable medium |
-
2018
- 2018-11-23 CN CN201811405302.1A patent/CN109558590B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1601969A (en) * | 2003-09-24 | 2005-03-30 | 华为技术有限公司 | Information analytical method for facility configuration |
| CN103699118A (en) * | 2013-12-18 | 2014-04-02 | 北京航天测控技术有限公司 | Method and device for analyzing abnormal state of spacecraft in operating process |
| CN106647695A (en) * | 2016-12-05 | 2017-05-10 | 航天恒星科技有限公司 | Method and system for monitoring running state of on-orbit spacecraft |
| CN107247613A (en) * | 2017-04-25 | 2017-10-13 | 北京航天飞行控制中心 | Sentence analytic method and sentence resolver |
| CN107525686A (en) * | 2017-08-21 | 2017-12-29 | 中国人民解放军63789部队 | A kind of more telemetry parameter analysis window division methods of spacecraft |
| CN108478926A (en) * | 2018-05-02 | 2018-09-04 | 清华大学深圳研究生院 | The method that the wearable device of body temperature can be automatically adjusted and adjust body temperature |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109558590A (en) | 2019-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107577588B (en) | Intelligent operation and maintenance system for mass log data | |
| CN106779505B (en) | Power transmission line fault early warning method and system based on big data driving | |
| CN107391353B (en) | Method for detecting abnormal behavior of complex software system based on log | |
| Wang et al. | An IoT application for fault diagnosis and prediction | |
| CN109522192B (en) | Prediction method based on knowledge graph and complex network combination | |
| US20190205771A1 (en) | Detective method and system for activity-or-behavior model construction and automatic detection of the abnormal activities or behaviors of a subject system without requiring prior domain knowledge | |
| CN108985380B (en) | Point switch fault identification method based on cluster integration | |
| CN104915434A (en) | Multi-dimensional time sequence classification method based on mahalanobis distance DTW | |
| CN111191710A (en) | Abnormal flight identification method based on big data | |
| CN108038211A (en) | A kind of unsupervised relation data method for detecting abnormality based on context | |
| CN112464439A (en) | Three-layer data mining-based power system fault type analysis method and device | |
| Lv et al. | Just‐in‐time learning–multiple subspace support vector data description used for non‐Gaussian dynamic batch process monitoring | |
| Wan et al. | Study on the satellite telemetry data classification based on self-learning | |
| CN109558590B (en) | Method for positioning key fault device based on spacecraft remote measurement parameter word segmentation | |
| WO2022058116A1 (en) | Explaining machine learning output in industrial applications | |
| CN109189840A (en) | A kind of online log analytic method of streaming | |
| CN103268329B (en) | Plasma panel manufacturing process data digging system | |
| CN109858507B (en) | Rare subsequence mining method of multidimensional time sequence data applied to atmospheric pollution control | |
| TWI639908B (en) | Method for detecting and diagnosing an abnormal process | |
| CN116502714A (en) | End-to-end knowledge graph generation method | |
| EP4105813A1 (en) | Method for analyzing data consisting of a large number of individual messages, computer program product and computer system | |
| Ip et al. | ML-assisted monitoring and characterization of IoT sensor networks | |
| CN112884348A (en) | Method for diagnosing production deviation source of aerospace initiator based on dynamic Bayesian network | |
| Nguyen | Feature Engineering and Health Indicator Construction for Fault Detection and Diagnostic | |
| CN115293379B (en) | Knowledge graph-based on-orbit spacecraft equipment anomaly detection 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |