CN115146084B - Method and device for acquiring equipment fault and maintenance data from unstructured data - Google Patents
Method and device for acquiring equipment fault and maintenance data from unstructured data Download PDFInfo
- Publication number
- CN115146084B CN115146084B CN202210834286.8A CN202210834286A CN115146084B CN 115146084 B CN115146084 B CN 115146084B CN 202210834286 A CN202210834286 A CN 202210834286A CN 115146084 B CN115146084 B CN 115146084B
- Authority
- CN
- China
- Prior art keywords
- data
- target
- unstructured
- monitoring data
- preset
- 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
- 238000012423 maintenance Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012544 monitoring process Methods 0.000 claims abstract description 143
- 230000005540 biological transmission Effects 0.000 claims abstract description 63
- 230000009466 transformation Effects 0.000 claims abstract description 63
- 238000012545 processing Methods 0.000 claims abstract description 23
- 230000002159 abnormal effect Effects 0.000 claims abstract description 14
- 238000001514 detection method Methods 0.000 claims description 103
- 238000006243 chemical reaction Methods 0.000 claims description 80
- 238000000605 extraction Methods 0.000 claims description 75
- 230000005856 abnormality Effects 0.000 claims description 9
- 238000002372 labelling Methods 0.000 claims description 8
- 238000009432 framing Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 239000000463 material Substances 0.000 abstract description 7
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/38—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/383—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/12—Use of codes for handling textual entities
- G06F40/151—Transformation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/12—Use of codes for handling textual entities
- G06F40/151—Transformation
- G06F40/154—Tree transformation for tree-structured or markup documents, e.g. XSLT, XSL-FO or stylesheets
-
- 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
- G06Q10/20—Administration of product repair or maintenance
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/19007—Matching; Proximity measures
- G06V30/19093—Proximity measures, i.e. similarity or distance measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
- G06V30/418—Document matching, e.g. of document images
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Library & Information Science (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for acquiring equipment fault and maintenance data from unstructured data, wherein the method comprises the following steps: acquiring data to be monitored of power transmission and transformation equipment; extracting unstructured type monitoring data from the data to be monitored; converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data; determining equipment fault information and target maintenance information of power transmission and transformation equipment according to the target characteristic data; the method solves the problems that the prior art can not timely find equipment faults and occurrence in power transmission and transformation equipment when abnormal information is detected from unstructured data, and consumes a large amount of manpower and material resources, and has lower accuracy and efficiency.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a method and a device for acquiring equipment failure and maintenance data from unstructured data.
Background
The power transmission and transformation equipment is different in type, different in suppliers and different in use time, so that the existence form, the source and the format of the generated data are different, a lot of important data are unstructured, the data analysis also needs to rely on manual detection, the efficiency of the data analysis is reduced, meanwhile, the equipment faults and the problems in the power transmission and transformation equipment cannot be found in time, a large amount of manpower and material resources are consumed, and the accuracy and the efficiency are low.
Disclosure of Invention
The invention aims to solve the technical problems that: the method and the device for acquiring the equipment fault and the maintenance data from the unstructured data are provided, so that the technical problems that the equipment fault and the occurrence problem in the power transmission and transformation equipment cannot be found in time when abnormal information is detected from the unstructured data in the prior art are solved, a large amount of manpower and material resources are consumed, and the accuracy and the efficiency are low are solved.
The technical scheme of the invention is as follows:
a method of acquiring equipment failure and repair data from unstructured data, the method comprising:
acquiring data to be monitored of power transmission and transformation equipment;
extracting unstructured type monitoring data from the data to be monitored;
converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data;
and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data.
The method for converting the unstructured type of monitoring data according to the data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data comprises the following steps:
determining a monitoring data type according to the unstructured type of monitoring data;
determining a feature extraction strategy according to the type of the monitoring data;
performing data conversion on unstructured type monitoring data according to a feature extraction strategy to obtain initial conversion data;
and carrying out data clustering on the initial conversion data to obtain target characteristic data.
The method for converting the unstructured type of monitoring data into the initial converted data according to the feature extraction strategy comprises the following steps:
when the feature extraction strategy is the first extraction strategy, performing image division on the unstructured type of monitoring data according to the first extraction strategy to obtain a division image of the unstructured type of monitoring data;
carrying out image framing on the divided images to obtain framed images;
and extracting key features of the framing images according to a preset image extraction model to obtain initial conversion data and sample labeling images.
Performing data conversion on the unstructured type of monitoring data according to a feature extraction strategy to obtain initial conversion data, wherein the method comprises the following steps:
when the feature extraction strategy is a second extraction strategy, performing text division on the unstructured type of monitoring data according to the second extraction strategy to obtain a division text;
carrying out prior probability calculation on the divided text to obtain target probability corresponding to the divided text;
calculating the similarity between the divided texts according to the target probability to obtain a similarity result;
and determining initial conversion data according to the similarity result and the divided text.
The method for determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data comprises the following steps:
determining a target detection tag corresponding to the target feature data according to the target feature data;
comparing the target detection label with a preset detection label to obtain a comparison result;
and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the comparison result.
The method for determining the target detection tag corresponding to the target feature data according to the target feature data comprises the following steps:
performing data conversion on the target characteristic data according to a preset conversion format to obtain target detection data;
searching a storage label corresponding to the target detection data in a preset relation storage library;
and determining the target detection label corresponding to the target characteristic information according to the storage label.
The preset conversion format refers to a preset format for converting key information of the target feature data, and the target feature data is converted into XML (extensive markup language) by the preset conversion format; converting the target feature data into semi-structured data through a preset conversion format, so as to obtain target detection data corresponding to the target feature data, and searching a storage label corresponding to the target detection data in a preset relation storage library, wherein the storage label corresponding to the target detection data is the target detection label corresponding to the target feature information; when the target feature data is text data, directly converting the target feature data to obtain a corresponding XML file, wherein the XML file corresponding to the target feature data is target detection data, and when the target feature data is image video data, carrying out semantic annotation based on the target feature data to obtain a corresponding text record, and outputting the text record to obtain the XML file; the semantic annotation based on the target feature data is obtained through a preset annotation model, and the preset annotation model is obtained through deep learning of the sample image and the semantic annotation result corresponding to the sample image.
When the comparison result is that the preset detection label is inconsistent with the target detection label, determining an abnormal detection label according to the target detection label; and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the abnormality detection tag and the target feature data corresponding to the abnormality detection tag.
The feature extraction strategy comprises the following steps:
the first extraction strategy is a strategy for extracting key information data from image video data, when unstructured type monitoring data are the image video data, the unstructured type monitoring data are processed according to the first extraction strategy, when the unstructured type monitoring data are video data, the unstructured type monitoring data are divided to obtain a plurality of videos with the same video duration, and the videos with the same video duration are divided images; after obtaining divided images with the same video duration, carrying out frame division identification on the divided images, so as to obtain corresponding frame division images after frame division in each divided image; after obtaining the frame images, extracting key features of the frame images through a preset image extraction model, so as to obtain initial conversion data after feature extraction; the preset image extraction model is a model which is obtained after training the sample labeling image and can extract key features in the image; the sample labeling image refers to an image corresponding to monitoring data with key features already labeled on the image;
the second extraction strategy is a strategy for extracting key information data of the text data, and when the unstructured type of monitoring data is the text data, the unstructured type of monitoring data is processed according to the second extraction strategy; text division is carried out on the unstructured type monitoring data to obtain words in the unstructured type monitoring data, and the words in the unstructured type monitoring data are divided texts; performing word frequency vector modeling according to unstructured type monitoring data, calculating prior probability of each divided text according to the word frequency vector modeling, and calculating similarity between each divided text according to the prior probability to obtain a similarity result; and after the similarity result and each divided text are obtained, each divided text and similarity result data among each divided text are used as initial conversion data.
An apparatus for obtaining equipment failure and repair data from unstructured data, the apparatus comprising:
the acquisition module is used for acquiring data to be monitored of the power transmission and transformation equipment;
the extraction module is used for extracting unstructured type monitoring data from the data to be monitored;
the conversion module is used for converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data;
and the determining module is used for determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data.
The beneficial effects of the invention are as follows:
the method comprises the steps of obtaining data to be monitored of power transmission and transformation equipment; extracting unstructured type monitoring data from the data to be monitored; converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data; and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data. By means of the method, the target characteristic information in the unstructured data is obtained through data conversion of the unstructured data, and the abnormal detection is carried out after the structured conversion is carried out based on the target characteristic information, so that fault equipment information and target maintenance data in power transmission and transformation equipment can be found in time, the detection efficiency and the detection accuracy are improved, and the labor consumption and the material cost are reduced.
The method solves the problems that the prior art can not timely find equipment faults and occurrence in power transmission and transformation equipment when abnormal information is detected from unstructured data, and consumes a large amount of manpower and material resources, and has lower accuracy and efficiency.
Drawings
FIG. 1 is a schematic diagram of an apparatus for obtaining equipment failure and repair data from unstructured data in a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flow chart of a first embodiment of a method of acquiring equipment failure and repair data from unstructured data according to the present invention;
FIG. 3 is a schematic diagram illustrating a second embodiment of a method for acquiring equipment failure and repair data from unstructured data according to the present invention;
fig. 4 is a block diagram of a first embodiment of an apparatus for obtaining equipment failure and repair data from unstructured data according to the present invention.
Detailed Description
Referring to fig. 1, fig. 1 is a schematic diagram of an apparatus structure of a hardware running environment according to an embodiment of the present invention, where apparatus failure and maintenance data are obtained from unstructured data.
As shown in fig. 1, the apparatus for acquiring apparatus failure and maintenance data from unstructured data may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) Memory or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the structure shown in FIG. 1 does not constitute a limitation of a device for obtaining device failure and repair data from unstructured data, and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a storage medium, may include an operating system, a network communication module, a user interface module, and a program for acquiring device failure and maintenance data from unstructured data.
In the device shown in fig. 1 for obtaining device failure and repair data from unstructured data, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the device for acquiring the device fault and the maintenance data from the unstructured data according to the present invention may be disposed in the device for acquiring the device fault and the maintenance data from the unstructured data, and the device for acquiring the device fault and the maintenance data from the unstructured data invokes the program for acquiring the device fault and the maintenance data from the unstructured data stored in the memory 1005 through the processor 1001, and performs the method for acquiring the device fault and the maintenance data from the unstructured data according to the embodiment of the present invention.
An embodiment of the present invention provides a method for obtaining equipment failure and maintenance data from unstructured data, and referring to fig. 2, fig. 2 is a flowchart of a first embodiment of a method for obtaining equipment failure and maintenance data from unstructured data according to the present invention.
The method for acquiring equipment fault and maintenance data from unstructured data comprises the following steps:
step S10: and acquiring data to be monitored of the power transmission and transformation equipment.
It should be noted that, the execution body of the embodiment is a control terminal in the power transmission and transformation system, a system for acquiring equipment failure and maintenance data from unstructured data is installed on the control device, after the control terminal acquires to-be-monitored data of each power transmission and transformation equipment in the power transmission and transformation system, the to-be-monitored data is transmitted to the system for acquiring equipment failure and maintenance data from unstructured data, unstructured type monitoring data is extracted from to-be-monitored data by the system for acquiring equipment failure and maintenance data from unstructured data, and unstructured type monitoring data is converted according to a data processing mode corresponding to unstructured type monitoring data, so as to obtain target feature data, and equipment failure information and target maintenance information of the power transmission and transformation equipment are determined according to the target feature data.
It can be understood that a plurality of power transmission and transformation devices exist in the power transmission and transformation system, each power transmission and transformation device can have corresponding operation data, and the operation data corresponding to each power transmission and transformation device is the data to be monitored of each power transmission and transformation device.
Step S20: and extracting unstructured type monitoring data from the data to be monitored.
It should be noted that, because the kinds and suppliers of the power transmission and transformation devices are different, the data to be monitored generated by the power transmission and transformation devices have different existence forms, and the data to be monitored has the characteristics of various existence forms and large data volume, including the structured type of monitoring data and the unstructured type of monitoring data. Structured data, also called row data, is data logically expressed and implemented by a two-dimensional table structure, strictly following data format and length specifications, and is stored and managed mainly by relational databases. Opposite to structured data is unstructured data that is not suitable for presentation by a two-dimensional table of a database, including office documents in all formats, XML, HTML, various types of reports, pictures and audio, video information, and the like.
It can be understood that the data to be monitored is filtered, and the monitoring data which does not conform to the preset data format and the preset length specification is extracted, and the monitoring data which does not conform to the preset data format and the preset length specification in the data to be monitored is unstructured type monitoring data.
Step S30: and converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data.
It should be noted that, since the unstructured type of monitoring data includes multiple data expression forms, such as text, image, video, etc., different data expression forms correspond to different data processing modes, the unstructured type of monitoring data is converted according to the data processing mode corresponding to the unstructured type of monitoring data, so as to obtain key information data corresponding to the unstructured type of monitoring data, based on the key information data, the unstructured type of monitoring data can be converted into structured data, and the target feature data refers to the key information data corresponding to the unstructured type of monitoring data.
Step S40: and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data.
After the target feature data is obtained, index inquiry can be performed based on the target feature data to obtain structured data corresponding to the target feature data, and finally searching and identifying are performed in the structured data to obtain problems or fault labels in the structured data, and equipment fault information and target maintenance information in the power transmission and transformation equipment are determined based on the problems or fault labels.
It may be appreciated that, in order to obtain accurate fault related information based on the target feature data, further, the determining, according to the target feature data, equipment fault information and target maintenance information of the power transmission and transformation equipment includes: determining a target detection tag corresponding to the target feature data according to the target feature data; comparing the target detection label with a preset detection label to obtain a comparison result; and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the comparison result.
In a specific implementation, a problem relation model exists in a preset relation storage library, the problem relation model organizes and stores data through a graphical structure, the problem relation model realized by the graphical database can effectively store and inquire many-to-many relations and comprises various semi-structured data and corresponding storage labels thereof, the storage labels comprise normal operation labels, problem labels, fault labels and the like of power transmission and transformation equipment corresponding to the data, the problem relation model can be used for creating and maintaining classification of the problem and the fault labels, the labels can be codes, numbers or other expression forms, and the problem relation model can be used for carrying out clear, standardized and non-overlapping classification on the problem labels and the fault labels.
After target feature data corresponding to unstructured type monitoring data is obtained, searching a target detection tag in a problem relation model based on the target feature data, wherein the preset detection tag refers to a tag of normal operation of power transmission and transformation equipment. And comparing the target detection label with a preset detection label, so as to obtain a comparison result, and determining equipment fault information and target maintenance information of the power transmission and transformation equipment based on the comparison result.
It may be appreciated that, in order to obtain an accurate target detection tag based on the target feature data, further, the determining, according to the target feature data, the target detection tag corresponding to the target feature data includes: performing data conversion on the target characteristic data according to a preset conversion format to obtain target detection data; searching a storage label corresponding to the target detection data in the preset relation storage library; and determining a target detection label corresponding to the target characteristic information according to the storage label.
In a specific implementation, the preset conversion format refers to a preset format for converting key information of the target feature data, and in this embodiment, the preset conversion format adopts XML of the target feature data, or may adopt other manners, which is not limited in this embodiment.
It should be noted that, the target feature data is converted into semi-structured data through a preset conversion format, so as to obtain target detection data corresponding to the target feature data, and a storage tag corresponding to the target detection data is searched in a preset relation storage library, where the storage tag corresponding to the target detection data is the target detection tag corresponding to the target feature information.
It can be understood that when the target feature data is text data, the target feature data can be directly converted to obtain a corresponding XML file, and the XML file corresponding to the target feature data is the target detection data. When the target feature data is image video data, semantic annotation is carried out based on the target feature data, so that corresponding text records are obtained, and the text records are output, so that an XML file is obtained. The semantic annotation based on the target feature data is obtained through a preset annotation model, and the preset annotation model is obtained through deep learning of a large number of sample images and semantic annotation results corresponding to the sample images.
In a specific implementation, in order to obtain accurate equipment fault information and target maintenance information based on a comparison result, further, determining the equipment fault information and the target maintenance information of the power transmission and transformation equipment according to the comparison result includes: when the comparison result is that the preset detection label is inconsistent with the target detection label, determining an abnormal detection label according to the target detection label; and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the abnormality detection tag and target feature data corresponding to the abnormality detection tag.
When the comparison result of the target detection tag and the preset detection tag is that the preset detection tag is inconsistent with the target detection tag, determining an abnormal detection tag according to the target detection tag inconsistent with the preset detection tag, and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to target feature data corresponding to the abnormal detection tag and the abnormal detection tag. For example, when the current target detection tag has a/B/C and the preset detection tag is a/B, if the target detection tag C is inconsistent with the preset detection tag, the target detection tag C is an abnormal detection tag, the target feature data 1 corresponding to the target detection tag C, the corresponding unstructured type monitoring data Z and the corresponding fault power transmission and transformation equipment one are obtained, and the abnormal detection tag C, the target feature data 1, the corresponding unstructured type monitoring data Z and the corresponding fault power transmission and transformation equipment one are equipment fault information and target maintenance information of the power transmission and transformation equipment.
The embodiment obtains the data to be monitored of the power transmission and transformation equipment; extracting unstructured type monitoring data from the data to be monitored; converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data; and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data. By means of the method, the target characteristic information in the unstructured data is obtained through data conversion of the unstructured data, and the abnormal detection is carried out after the structured conversion is carried out based on the target characteristic information, so that fault equipment information and target maintenance data in power transmission and transformation equipment can be found in time, the detection efficiency and the detection accuracy are improved, and the labor consumption and the material cost are reduced.
Referring to fig. 3, fig. 3 is a flowchart illustrating a second embodiment of a method for acquiring equipment failure and maintenance data from unstructured data according to the present invention.
Based on the first embodiment, the step S30 in the method for obtaining the equipment failure and maintenance data from unstructured data according to the present embodiment includes:
step S31: and determining the type of the monitoring data according to the unstructured type of the monitoring data.
It should be noted that, since the unstructured type of monitoring data includes multiple data representations, such as text, image, video, etc., the type of monitoring data of the unstructured type of monitoring data needs to be determined according to the unstructured type of monitoring data. For example, the monitoring data of the first power transmission and transformation device in the data to be monitored is unstructured monitoring data, and the data type of the monitoring data of the first power transmission and transformation device is determined to be text type.
Step S32: and determining a feature extraction strategy according to the type of the monitoring data.
It should be noted that, different data expression forms correspond to different data processing modes, and the corresponding data processing modes are determined according to the monitoring data type of the unstructured monitoring data, and the corresponding data processing modes are feature extraction strategies. The feature extraction strategy is a strategy for extracting key information data corresponding to unstructured type monitoring data, and the key information data is data which can directly reflect main operation information of power transmission and transformation equipment after screening, filtering and extracting the monitoring data of the power transmission and transformation equipment.
Step S33: and carrying out data conversion on the unstructured type monitoring data according to the feature extraction strategy to obtain initial conversion data.
It should be noted that, based on the feature extraction policy, data conversion is performed on unstructured type monitoring data, so as to obtain initial conversion data, where the initial conversion data refers to key information data corresponding to the unstructured type monitoring data.
It can be appreciated that, in order to ensure accurate data conversion of unstructured type monitoring data of an image video type, further, the data conversion of the unstructured type monitoring data according to the feature extraction policy to obtain initial conversion data includes: when the feature extraction strategy is a first extraction strategy, performing image division on the unstructured type of monitoring data according to the first extraction strategy to obtain a division image of the unstructured type of monitoring data; carrying out image framing on the divided images to obtain framed images; and extracting key features of the frame-divided images according to a preset image extraction model to obtain initial conversion data and a sample labeling image.
In a specific implementation, the first extraction policy refers to a policy for extracting key information data from image video class data. And when the unstructured type of monitoring data is image video type data, processing the unstructured type of monitoring data according to a first extraction strategy. When the unstructured type monitoring data are video type data, dividing the unstructured type monitoring data to obtain a plurality of videos with the same video duration, wherein the videos with the same video duration are divided images.
After obtaining the divided images with the same video duration, frame division identification is performed on the divided images, so as to obtain corresponding frame division images after frame division in each divided image. And after the frame division image is obtained, extracting key features of the frame division image through a preset image extraction model, so as to obtain initial conversion data after feature extraction. The preset image extraction model is a model which is obtained by training a large number of sample marked images and can extract key features in the images. The sample labeling image refers to an image corresponding to the monitoring data in which the key features have been labeled on the image.
It can be understood that, when the unstructured type of monitoring data is specifically picture type data, feature extraction is performed on the unstructured type of monitoring data only through a preset image extraction model, so as to obtain initial conversion data from which key features are extracted.
In a specific implementation, in order to ensure that the unstructured type of monitoring data of a text type can be accurately converted, further, the step of performing data conversion on the unstructured type of monitoring data according to the feature extraction policy to obtain initial conversion data includes: when the feature extraction strategy is a second extraction strategy, performing text division on the unstructured type of monitoring data according to the second extraction strategy to obtain a division text; performing prior probability calculation on the divided text to obtain target probability corresponding to the divided text; calculating the similarity between the divided texts according to the target probability to obtain a similarity result; and determining initial conversion data according to the similarity result and the divided text.
It should be noted that the second policy refers to a policy for extracting key information data from text data. And when the unstructured type of monitoring data is text type data, processing the unstructured type of monitoring data according to a second extraction strategy. And carrying out text division on the unstructured type monitoring data to obtain words in the unstructured type monitoring data, wherein the words in the unstructured type monitoring data are the division text. And carrying out word frequency vector modeling according to the unstructured type monitoring data, calculating prior probability of each divided text according to the word frequency vector modeling, and calculating similarity between each divided text according to the prior probability to obtain a similarity result. The present embodiment is not limited to this for the manner in which the similarity between texts is calculated.
It is understood that after the similarity result and each divided text are obtained, each divided text and the similarity result data between each divided text are used as the initial conversion data.
Step S34: and carrying out data clustering on the initial conversion data to obtain target characteristic data.
After the initial conversion data is obtained, the initial conversion data needs to be subjected to data clustering, so that final target feature data is obtained. When the initial conversion data is the image video data or the unstructured monitoring data of the picture data, data clustering is needed according to information appearing in each image of the initial conversion data, such as information of scenes, time, characters and the like, so that the clustered target feature data under each category is obtained. When the initial conversion data is the unstructured type monitoring data of the text data, data clustering is required to be carried out according to the similarity of each divided text and the word meaning of each divided text in the initial conversion data, so that the target characteristic data under each semantic category after clustering is obtained.
The embodiment determines the type of the monitoring data according to the unstructured type of the monitoring data; determining a feature extraction strategy according to the type of the monitoring data; performing data conversion on the unstructured type of monitoring data according to the feature extraction strategy to obtain initial conversion data; and carrying out data clustering on the initial conversion data to obtain target characteristic data. And determining a corresponding feature extraction strategy according to the data type of the unstructured type monitoring data, thereby ensuring the accuracy and pertinence of feature extraction, carrying out data clustering on the initial conversion data to obtain target feature data, and ensuring the high efficiency in the follow-up fault monitoring and identification according to the target feature data.
In addition, referring to fig. 4, an embodiment of the present invention further proposes an apparatus for acquiring equipment failure and maintenance data from unstructured data, where the apparatus for acquiring equipment failure and maintenance data from unstructured data includes:
the acquisition module 10 is configured to acquire data to be monitored of the power transmission and transformation device.
And the extracting module 20 is used for extracting unstructured type monitoring data from the data to be monitored.
And the conversion module 30 is configured to convert the unstructured type of monitoring data according to a data processing manner corresponding to the unstructured type of monitoring data, so as to obtain target feature data.
And the determining module 40 is used for determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data.
The embodiment obtains the data to be monitored of the power transmission and transformation equipment; extracting unstructured type monitoring data from the data to be monitored; converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data; and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data. By means of the method, the target characteristic information in the unstructured data is obtained through data conversion of the unstructured data, and the abnormal detection is carried out after the structured conversion is carried out based on the target characteristic information, so that fault equipment information and target maintenance data in power transmission and transformation equipment can be found in time, the detection efficiency and the detection accuracy are improved, and the labor consumption and the material cost are reduced.
In an embodiment, the conversion module 30 is further configured to determine a monitoring data type according to the unstructured type of monitoring data;
determining a feature extraction strategy according to the type of the monitoring data;
performing data conversion on the unstructured type of monitoring data according to the feature extraction strategy to obtain initial conversion data;
and carrying out data clustering on the initial conversion data to obtain target characteristic data.
In an embodiment, the conversion module 30 is further configured to, when the feature extraction policy is a first extraction policy, perform image division on the unstructured type of monitoring data according to the first extraction policy, so as to obtain a divided image of the unstructured type of monitoring data;
carrying out image framing on the divided images to obtain framed images;
and extracting key features of the frame-divided images according to a preset image extraction model to obtain initial conversion data and a sample labeling image.
In an embodiment, the conversion module 30 is further configured to, when the feature extraction policy is a second extraction policy, perform text division on the unstructured type of monitoring data according to the second extraction policy to obtain a division text;
performing prior probability calculation on the divided text to obtain target probability corresponding to the divided text;
calculating the similarity between the divided texts according to the target probability to obtain a similarity result;
and determining initial conversion data according to the similarity result and the divided text.
In an embodiment, the determining module 40 is further configured to determine, according to the target feature data, a target detection tag corresponding to the target feature data;
comparing the target detection label with a preset detection label to obtain a comparison result;
and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the comparison result.
In an embodiment, the determining module 40 is further configured to perform data conversion on the target feature data according to a preset conversion format to obtain target detection data;
searching a storage label corresponding to the target detection data in the preset relation storage library;
and determining a target detection label corresponding to the target characteristic information according to the storage label.
In an embodiment, the determining module 40 is further configured to, when the comparison result is that the preset detection tag is inconsistent with the target detection tag,
determining an abnormality detection tag according to the target detection tag;
and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the abnormality detection tag and target feature data corresponding to the abnormality detection tag.
Because the device adopts all the technical schemes of all the embodiments, the device at least has all the beneficial effects brought by the technical schemes of the embodiments, and the description is omitted here.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium stores a program for acquiring equipment failure and maintenance data from unstructured data, and the program for acquiring the equipment failure and maintenance data from unstructured data realizes the steps of the method for acquiring the equipment failure and maintenance data from unstructured data when being executed by a processor.
Claims (7)
1. A method of obtaining equipment failure and repair data from unstructured data, characterized by: the method comprises the following steps:
acquiring data to be monitored of power transmission and transformation equipment;
extracting unstructured type monitoring data from the data to be monitored;
converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data;
the method for determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data specifically comprises the following steps:
determining a target detection tag corresponding to the target feature data according to the target feature data;
performing data conversion on the target characteristic data according to a preset conversion format to obtain target detection data;
searching a storage label corresponding to the target detection data in a preset relation storage library;
determining a target detection label corresponding to the target characteristic data according to the storage label;
comparing the target detection label with a preset detection label to obtain a comparison result;
and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the comparison result.
2. A method of obtaining equipment failure and repair data from unstructured data according to claim 1, wherein: the method for converting the unstructured type of monitoring data according to the data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data comprises the following steps:
determining a monitoring data type according to the unstructured type of monitoring data;
determining a feature extraction strategy according to the type of the monitoring data;
performing data conversion on unstructured type monitoring data according to a feature extraction strategy to obtain initial conversion data;
and carrying out data clustering on the initial conversion data to obtain target characteristic data.
3. A method of obtaining equipment failure and repair data from unstructured data according to claim 2, wherein: the method for converting the unstructured type of monitoring data into the initial converted data according to the feature extraction strategy comprises the following steps:
when the feature extraction strategy is the first extraction strategy, performing image division on the unstructured type of monitoring data according to the first extraction strategy to obtain a division image of the unstructured type of monitoring data;
carrying out image framing on the divided images to obtain framed images;
extracting key features of the frame images according to a preset image extraction model to obtain initial conversion data;
the feature extraction strategy comprises the following steps:
the first extraction strategy is a strategy for extracting key information data from image video data, when unstructured type monitoring data are the image video data, the unstructured type monitoring data are processed according to the first extraction strategy, when the unstructured type monitoring data are video data, the unstructured type monitoring data are divided to obtain a plurality of videos with the same video duration, and the videos with the same video duration are divided images; after obtaining divided images with the same video duration, carrying out frame division identification on the divided images, so as to obtain corresponding frame division images after frame division in each divided image; after obtaining the frame images, extracting key features of the frame images through a preset image extraction model, so as to obtain initial conversion data after feature extraction; the preset image extraction model is a model which is obtained after training the sample labeling image and can extract key features in the image; the sample labeling image refers to an image corresponding to monitoring data with key features already labeled on the image;
the second extraction strategy is a strategy for extracting key information data of the text data, and when the unstructured type of monitoring data is the text data, the unstructured type of monitoring data is processed according to the second extraction strategy; text division is carried out on the unstructured type monitoring data to obtain words in the unstructured type monitoring data, and the words in the unstructured type monitoring data are divided texts; performing word frequency vector modeling according to unstructured type monitoring data, calculating prior probability of each divided text according to the word frequency vector modeling, and calculating similarity between each divided text according to the prior probability to obtain a similarity result; and after the similarity result and each divided text are obtained, each divided text and similarity result data among each divided text are used as initial conversion data.
4. A method of obtaining equipment failure and repair data from unstructured data according to claim 2, wherein: performing data conversion on the unstructured type of monitoring data according to a feature extraction strategy to obtain initial conversion data, wherein the method comprises the following steps:
when the feature extraction strategy is a second extraction strategy, performing text division on the unstructured type of monitoring data according to the second extraction strategy to obtain a division text;
carrying out prior probability calculation on the divided text to obtain target probability corresponding to the divided text;
calculating the similarity between the divided texts according to the target probability to obtain a similarity result;
and determining initial conversion data according to the similarity result and the divided text.
5. A method of obtaining equipment failure and repair data from unstructured data according to claim 1, wherein: the preset conversion format refers to a preset format for converting key information of the target feature data, and the target feature data is converted into XML (extensive markup language) by the preset conversion format; converting the target feature data into semi-structured data through a preset conversion format, so as to obtain target detection data corresponding to the target feature data, and searching a storage label corresponding to the target detection data in a preset relation storage library, wherein the storage label corresponding to the target detection data is the target detection label corresponding to the target feature data; when the target feature data is text data, directly converting the target feature data to obtain a corresponding XML file, wherein the XML file corresponding to the target feature data is target detection data, and when the target feature data is image video data, carrying out semantic annotation based on the target feature data to obtain a corresponding text record, and outputting the text record to obtain the XML file; the semantic annotation based on the target feature data is obtained through a preset annotation model, and the preset annotation model is obtained through deep learning of the sample image and the semantic annotation result corresponding to the sample image.
6. A method of obtaining equipment failure and repair data from unstructured data according to claim 1, wherein: when the comparison result is that the preset detection label is inconsistent with the target detection label, determining an abnormal detection label according to the target detection label; and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the abnormality detection tag and the target feature data corresponding to the abnormality detection tag.
7. An apparatus for obtaining equipment failure and repair data from unstructured data, characterized by: the device comprises:
the acquisition module is used for acquiring data to be monitored of the power transmission and transformation equipment;
the extraction module is used for extracting unstructured type monitoring data from the data to be monitored;
the conversion module is used for converting the unstructured type of monitoring data according to a data processing mode corresponding to the unstructured type of monitoring data to obtain target characteristic data;
the determining module is used for determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the target characteristic data; the method specifically comprises the following steps:
determining a target detection tag corresponding to the target feature data according to the target feature data;
performing data conversion on the target characteristic data according to a preset conversion format to obtain target detection data;
searching a storage label corresponding to the target detection data in a preset relation storage library;
determining a target detection label corresponding to the target characteristic data according to the storage label;
comparing the target detection label with a preset detection label to obtain a comparison result;
and determining equipment fault information and target maintenance information of the power transmission and transformation equipment according to the comparison result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210834286.8A CN115146084B (en) | 2022-07-14 | 2022-07-14 | Method and device for acquiring equipment fault and maintenance data from unstructured data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210834286.8A CN115146084B (en) | 2022-07-14 | 2022-07-14 | Method and device for acquiring equipment fault and maintenance data from unstructured data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115146084A CN115146084A (en) | 2022-10-04 |
| CN115146084B true CN115146084B (en) | 2023-11-24 |
Family
ID=83412676
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210834286.8A Active CN115146084B (en) | 2022-07-14 | 2022-07-14 | Method and device for acquiring equipment fault and maintenance data from unstructured data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115146084B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108319695A (en) * | 2018-02-02 | 2018-07-24 | 华自科技股份有限公司 | Power station fault data processing method, device, computer equipment and storage medium |
| CN109344298A (en) * | 2018-10-31 | 2019-02-15 | 南方电网科学研究院有限责任公司 | Method and device for converting unstructured data into structured data |
| CN111651601A (en) * | 2020-06-02 | 2020-09-11 | 全球能源互联网研究院有限公司 | Training method and classification method for fault classification model of power information system |
| CN112418451A (en) * | 2020-11-05 | 2021-02-26 | 武汉大学 | Transformer fault diagnosis positioning system based on digital twinning |
| CN113064791A (en) * | 2021-03-25 | 2021-07-02 | 上海佰粒科技有限公司 | Scattered label feature extraction method based on real-time monitoring of mass log data |
| CN113486179A (en) * | 2021-07-13 | 2021-10-08 | 盛景智能科技(嘉兴)有限公司 | Product data analysis method and system based on maintenance work order |
| CN113672743A (en) * | 2021-08-20 | 2021-11-19 | 北京宝兰德软件股份有限公司 | Fault judging method and device, electronic equipment, storage medium and product |
| CN114254668A (en) * | 2021-11-26 | 2022-03-29 | 南方电网数字电网研究院有限公司 | Fault detection method and device based on discharge signal of insulated switchgear |
-
2022
- 2022-07-14 CN CN202210834286.8A patent/CN115146084B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108319695A (en) * | 2018-02-02 | 2018-07-24 | 华自科技股份有限公司 | Power station fault data processing method, device, computer equipment and storage medium |
| CN109344298A (en) * | 2018-10-31 | 2019-02-15 | 南方电网科学研究院有限责任公司 | Method and device for converting unstructured data into structured data |
| CN111651601A (en) * | 2020-06-02 | 2020-09-11 | 全球能源互联网研究院有限公司 | Training method and classification method for fault classification model of power information system |
| CN112418451A (en) * | 2020-11-05 | 2021-02-26 | 武汉大学 | Transformer fault diagnosis positioning system based on digital twinning |
| CN113064791A (en) * | 2021-03-25 | 2021-07-02 | 上海佰粒科技有限公司 | Scattered label feature extraction method based on real-time monitoring of mass log data |
| CN113486179A (en) * | 2021-07-13 | 2021-10-08 | 盛景智能科技(嘉兴)有限公司 | Product data analysis method and system based on maintenance work order |
| CN113672743A (en) * | 2021-08-20 | 2021-11-19 | 北京宝兰德软件股份有限公司 | Fault judging method and device, electronic equipment, storage medium and product |
| CN114254668A (en) * | 2021-11-26 | 2022-03-29 | 南方电网数字电网研究院有限公司 | Fault detection method and device based on discharge signal of insulated switchgear |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115146084A (en) | 2022-10-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110263009B (en) | Method, device and equipment for generating log classification rule and readable storage medium | |
| CN110580308B (en) | Information auditing method and device, electronic equipment and storage medium | |
| WO2008098956A1 (en) | Method and apparatus for automatically discovering features in free form heterogeneous data | |
| EP3916584A1 (en) | Information processing method and apparatus, electronic device and storage medium | |
| CN113449099A (en) | Text classification method and text classification device | |
| CN112433874A (en) | Fault positioning method, system, electronic equipment and storage medium | |
| CN110457193A (en) | Health portrait display method and system based on operation and maintenance data of power information system | |
| CN116244410B (en) | Index data analysis method and system based on knowledge graph and natural language | |
| CN113704667A (en) | Automatic extraction processing method and device for bidding announcement | |
| CN113094512A (en) | Fault analysis system and method in industrial production and manufacturing | |
| CN114491034B (en) | Text classification method and intelligent device | |
| CN115062117A (en) | Method for automatically generating and classifying documents based on natural language processing technology | |
| CN114996936A (en) | Equipment operation and maintenance method, equipment operation and maintenance device, equipment operation and maintenance equipment and storage medium | |
| CN112925905B (en) | Method, device, electronic equipment and storage medium for extracting video subtitles | |
| CN115146084B (en) | Method and device for acquiring equipment fault and maintenance data from unstructured data | |
| CN113609825B (en) | Intelligent customer attribute tag identification method and device | |
| CN110765325A (en) | Operation and maintenance analysis method and system of CEPH distributed storage system | |
| CN116225848A (en) | Log monitoring method, device, equipment and medium | |
| CN115982272A (en) | Data labeling method and device for urban big data management and computer storage medium | |
| CN113821555A (en) | Unstructured data collection processing method of intelligent supervision black box | |
| CN112559739A (en) | Method for processing insulation state data of power equipment | |
| CN111858978B (en) | Method and system for automatically making network courseware icons | |
| CN110633312B (en) | E-commerce public opinion analysis method and system based on Mongodb and HandP | |
| CN111522900B (en) | Automatic analysis method, system, equipment and storage medium for unstructured data | |
| CN117173511A (en) | Category identification method, apparatus, device, storage medium, and program product |
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 |