CN116821403B - Intelligent operation and maintenance method and system for factory equipment - Google Patents
Intelligent operation and maintenance method and system for factory equipment Download PDFInfo
- Publication number
- CN116821403B CN116821403B CN202311108904.1A CN202311108904A CN116821403B CN 116821403 B CN116821403 B CN 116821403B CN 202311108904 A CN202311108904 A CN 202311108904A CN 116821403 B CN116821403 B CN 116821403B
- Authority
- CN
- China
- Prior art keywords
- equipment
- image
- maintenance
- feature
- target
- 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 241
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000006870 function Effects 0.000 claims description 23
- 230000002159 abnormal effect Effects 0.000 claims description 21
- 238000004458 analytical method Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 18
- 238000001454 recorded image Methods 0.000 claims description 14
- 238000013507 mapping Methods 0.000 claims description 13
- 238000007405 data analysis Methods 0.000 claims description 6
- 238000007689 inspection Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 5
- 238000007726 management method Methods 0.000 description 10
- 230000008859 change Effects 0.000 description 8
- 238000010801 machine learning Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000003745 diagnosis Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 230000008439 repair process Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 208000037408 Device failure Diseases 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 206010063659 Aversion Diseases 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010252 digital analysis Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000007 visual effect Effects 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
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
-
- 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
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Economics (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Library & Information Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses an intelligent operation and maintenance method and system for factory equipment, belonging to the field of intelligent operation and maintenance, wherein the method comprises the following steps: firstly, acquiring application time and equipment operation data of target equipment, and acquiring two sections of adjacent equipment operation data; then, identifying a target fault and a corresponding target component by using a fault identification model; then searching a target component corresponding to the target fault in the equipment fault database, and setting the target component as a preset operation and maintenance component; then acquiring an operation and maintenance record image of the target equipment, and analyzing the image by using a digital twin network model to obtain an image distinguishing feature set; and finally judging whether the scheduled operation and maintenance component is operated and maintained according to the image distinguishing feature set, and if the scheduled operation and maintenance component is not operated and maintained, sending operation and maintenance reminding information to remind the user to carry out operation and maintenance management. The application solves the technical problems of inaccurate fault identification and low operation and maintenance efficiency of factory equipment in the prior art, and achieves the technical effects of accurately and rapidly identifying equipment faults and improving operation and maintenance working efficiency and accuracy.
Description
Technical Field
The application relates to the field of intelligent operation and maintenance, in particular to an intelligent operation and maintenance method and system for factory equipment.
Background
With the development of industrialization and automation, the number and complexity of devices in a factory are continuously increased, and the difficulty and workload of operation and maintenance of the devices are rapidly increased. The traditional operation and maintenance modes of manual judgment and regular overhaul cannot meet the high-efficiency and stable operation requirements of equipment.
In the existing factory equipment operation and maintenance method, engineering technicians mainly rely on periodic inspection and manual judgment to find equipment faults and component operation and maintenance requirements. The operation and maintenance mode has low fault recognition accuracy and low operation and maintenance efficiency, and is difficult to ensure that all equipment faults and unprocessed operation and maintenance problems are found in time, and equipment accidents are easy to occur.
Disclosure of Invention
The application provides an intelligent operation and maintenance method and system for factory equipment, and aims to solve the technical problems of inaccurate fault identification and low operation and maintenance efficiency of the factory equipment in the prior art.
In view of the above problems, the present application provides an intelligent operation and maintenance method and system for factory equipment.
The first aspect of the present disclosure provides an intelligent operation and maintenance method for factory equipment, which includes: acquiring a target application log of target equipment in a factory equipment terminal, wherein the target application log comprises a plurality of groups of equipment operation data with application time identifiers; acquiring a first application time, and matching first equipment operation data under the first application time in a plurality of groups of equipment operation data with application time identifiers; acquiring second application time, and matching second equipment operation data under the second application time in a plurality of groups of equipment operation data with application time identifiers, wherein the second application time and the first application time are the time of two adjacent applications; analyzing the first equipment operation data and the second equipment operation data through a fault identification model to obtain a target identification fault; matching a target component corresponding to the target identification fault in the equipment fault database, and taking the target component as a preset operation and maintenance component; acquiring a target operation and maintenance record of target equipment in a target time period, and extracting a first operation and maintenance record image in the target operation and maintenance record; analyzing the first operation and maintenance record image through a digital twin network model to obtain a first image distinguishing characteristic set; and according to the first image distinguishing feature set, if the scheduled operation and maintenance component is not operated and maintained, sending out first reminding information, wherein the first reminding information is used for reminding the scheduled operation and maintenance component to be operated and maintained and managed.
In another aspect of the disclosure, an intelligent operation and maintenance system for factory equipment is provided, the system comprising: the system comprises an application log acquisition module, a control module and a control module, wherein the application log acquisition module is used for acquiring a target application log of target equipment in a factory equipment terminal, and the target application log comprises a plurality of groups of equipment operation data with application time identifiers; the first data matching module is used for acquiring the first application time and matching the first device operation data under the first application time in a plurality of groups of device operation data with the application time identifier; the second data matching module is used for acquiring second application time and matching second equipment operation data under the second application time in a plurality of groups of equipment operation data with application time identifiers, wherein the second application time and the first application time are the time of two adjacent applications; the operation data analysis module is used for analyzing the first equipment operation data and the second equipment operation data through the fault identification model to obtain a target identification fault; the operation and maintenance component matching module is used for matching a target component corresponding to the target identification fault in the equipment fault database and taking the target component as a preset operation and maintenance component; the operation and maintenance record acquisition module is used for acquiring a target operation and maintenance record of target equipment in a target time period and extracting a first operation and maintenance record image in the target operation and maintenance record; the recorded image analysis module is used for analyzing the first operation and maintenance recorded image through the digital twin network model to obtain a first image distinguishing characteristic set; and the operation and maintenance management reminding module is used for distinguishing the feature set according to the first image, and sending out first reminding information if the scheduled operation and maintenance component is not operated and maintained, wherein the first reminding information is used for reminding the user to carry out operation and maintenance management on the scheduled operation and maintenance component.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
because the target application log of the target equipment is acquired and comprises application time and equipment operation data, two sections of adjacent equipment operation data are acquired through time matching; analyzing the operation data of the two sections of equipment by using a fault identification model, and identifying target faults and corresponding target components; searching a target component corresponding to the target fault in the equipment fault database, and setting the target component as a preset operation and maintenance component; acquiring an operation and maintenance record image of target equipment in a target time period, and analyzing the image by using a digital twin network model to obtain an image distinguishing feature set; judging whether the scheduled operation and maintenance component is operated or not according to the image distinguishing feature set, if not, sending operation and maintenance reminding information to remind the user of operation and maintenance management of the component, solving the technical problems of inaccurate identification of plant equipment faults and low operation and maintenance efficiency in the prior art, and achieving the technical effects of accurately and rapidly identifying equipment faults and improving operation and maintenance work efficiency and accuracy.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a schematic diagram of a possible flow chart of an intelligent operation and maintenance method of factory equipment according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of a possible target identification fault obtained in an intelligent operation and maintenance method of factory equipment according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a possible flow chart for building a device failure database in an intelligent operation and maintenance method of factory equipment according to an embodiment of the present application;
fig. 4 is a schematic diagram of a possible architecture of an intelligent operation and maintenance system of factory equipment according to an embodiment of the present application.
Reference numerals illustrate: the system comprises an application log acquisition module 11, a first data matching module 12, a second data matching module 13, an operation data analysis module 14, an operation and maintenance component matching module 15, an operation and maintenance record acquisition module 16, a record image analysis module 17 and an operation and maintenance management reminding module 18.
Detailed Description
The technical scheme provided by the application has the following overall thought:
the embodiment of the application provides an intelligent operation and maintenance method and system for factory equipment. Firstly, a target application log of target equipment is obtained, wherein the target application log comprises application time and equipment operation data. And then, extracting two sections of adjacent equipment operation data by applying time matching, and analyzing the equipment operation data by adopting a fault identification model to identify a target fault and a corresponding target component. Then, a target component corresponding to the target fault is found in the equipment fault database and is set as a preset operation and maintenance component. And then, analyzing the operation and maintenance record image by using the digital twin network model to obtain an image distinguishing characteristic set. And finally, judging whether the preset operation and maintenance component is operated and maintained according to the image distinguishing feature set, and automatically sending operation and maintenance reminding information if the operation and maintenance component is not operated and maintained.
In a word, through carrying out digital analysis and processing on the equipment operation log and the operation and maintenance image record, the automatic identification of equipment faults and the operation and maintenance reminding function of key components are realized. Compared with the prior art, the method has the advantages that the operation and maintenance efficiency is greatly improved, the hidden danger of manual judgment is reduced, the accuracy of fault identification is improved, the operation and maintenance cost is reduced, and technical support is provided for safe and stable operation of equipment.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, an embodiment of the present application provides an intelligent operation and maintenance method for factory equipment, where the method includes:
step S100: obtaining a target application log of target equipment in a factory equipment terminal, wherein the target application log comprises a plurality of groups of equipment operation data with application time identifiers;
specifically, the target application log refers to log data generated by the target device in the process of producing the application, and is used for recording the running state of the target device. The target application log contains a plurality of sets of device operation data, each set of device operation data having an application time identifier for distinguishing the operation states of the devices in different periods.
An application log acquisition module is deployed on the target equipment and is used for acquiring the operation data of the target equipment in real time and identifying an application time identifier on the operation data acquired in real time to form a target application log. The application log acquisition module is realized by adopting a data acquisition card and data acquisition software, the data acquisition card is connected with a sensor and an executing mechanism of target equipment, and acquires data such as technological parameters, state parameters and the like of the equipment in real time, and the data acquisition software analyzes and processes the acquired data and adds application time identifiers to form a target application log.
And then, the target application log is sent to a cloud computing center of the factory equipment terminal in real time through a network interface such as Ethernet, wiFi and the like. The cloud computing center is provided with an application log receiving module which is used for receiving target application logs sent by each target device and storing the target application logs in the cloud database. Then, the application log receiving module analyzes the received target application log, extracts information comprising the application time identifier and the equipment operation data, and organizes the information into a plurality of groups of equipment operation data storage, namely, each group of equipment operation data corresponds to one application and has the application time identifier. In the operation and maintenance process, equipment operation data of the target equipment in a specific application time period is inquired and obtained by accessing a cloud database of a cloud computing center, and basic data support is provided for realizing intelligent operation and maintenance of the equipment.
Step S200: acquiring a first application time, and matching first equipment operation data under the first application time in the plurality of groups of equipment operation data with the application time identifier;
specifically, the first application time refers to an application start time point at which the running state of the target device is determined to be queried and analyzed. And setting an application time query module at a human-computer interaction interface of the intelligent operation and maintenance system, wherein the application time query module is used for receiving a first application time input by a user.
Firstly, manually inputting date and time to be judged in a query interface by a user as a first application time; or selecting a preset time range, and automatically determining the first application time in the time range by the system. And then, the application time inquiry module sends the first application time to the cloud computing center, and the application log receiving module of the cloud computing center searches the stored multiple groups of device operation data for the device operation data with the application time identification matched with the first application time after receiving the first application time, and defines the device operation data as first device operation data. After the operation data of the first equipment are obtained, the cloud computing center sends the operation data to the intelligent operation and maintenance system, and information such as technological parameters, state parameters and the like of the target equipment under the first application time is displayed on a man-machine interaction interface, so that input information is provided for the fault identification model.
By receiving the query time input by the user, the first equipment operation data of the target equipment under the specific application time is accurately acquired from the stored multiple groups of equipment operation data, the operation data of the target equipment under the specific application time is accurately acquired, accurate data are provided for equipment operation and maintenance analysis, and the accuracy and efficiency of fault identification are improved.
Step S300: acquiring second application time, and matching second equipment operation data under the second application time in the plurality of groups of equipment operation data with the identification of the application time, wherein the second application time and the first application time are adjacent twice application time;
specifically, the second application time refers to an adjacent application time point after the first application time of the running state of the target device recorded in the cloud database, and is an application termination time point of the running state of the target device to be analyzed. The second application time is acquired similar to the first application time, either manually entered by the user or automatically acquired by the system. After the second application time is acquired, matching the application time identifier with the data of the second application time in a plurality of groups of equipment operation data stored in a cloud database of the cloud computing center. The matched device operation data is the operation state of the target device in the second application time, namely the application period, and is defined as second device operation data.
By acquiring the second equipment operation data and analyzing the second equipment operation data together, dynamic change information of the operation state of the target equipment can be provided, input information is provided for the fault identification model, and the input information is used for judging the target identification fault of the equipment operation so as to accurately maintain the equipment in the factory.
Step S400: analyzing the first equipment operation data and the second equipment operation data through a fault identification model to obtain target identification faults;
specifically, the failure recognition model refers to a target device failure diagnosis model constructed by a machine learning method. The model is obtained by training historical fault data and corresponding normal operation data of the target equipment, can be used for identifying anomalies in new operation data of the target equipment, and realizes fault prediction and diagnosis.
Firstly, collecting a historical application fault log and a corresponding normal running log of target equipment, constructing a fault sample data set, and training to obtain a fault recognition model by using the fault sample data set through a machine learning method of supervised learning, such as SVM, BP neural network, random forest and the like. And then, inputting the first equipment operation data and the second equipment operation data into a fault recognition model, and judging whether the new sample is abnormal or not by analyzing each parameter index in the new sample by the fault recognition model. If some parameter indexes in the first equipment operation data and the second equipment operation data are beyond the normal range of the model, the fault identification model can judge that faults or performance reduction trends exist. Then, the fault identification model identifies the most likely fault type or component according to all parameter indexes of the new sample, so as to obtain a target identification fault, and provide an important reference basis for equipment fault diagnosis and maintenance, thereby improving the efficiency of fault processing.
Step S500: matching a target component corresponding to the target identification fault in an equipment fault database, and taking the target component as a preset operation and maintenance component;
specifically, the equipment fault database refers to a historical fault knowledge base of the target equipment constructed by the cloud computing center, wherein mapping relations between various fault types of the target equipment and corresponding maintenance components are stored, and the mapping relations are used for quickly determining the components needing to be checked and maintained when new fault types of the target equipment are identified.
Firstly, according to the historical maintenance record of the target equipment, analyzing the fault phenomenon, the fault reason and the processed maintenance parts of each maintenance, establishing the corresponding relation between the fault type and the maintenance parts, and constructing an equipment fault database. Then, the fault information of the target identification fault identified by the fault identification model is input into a fault database of the equipment, and the database searches all possible target components corresponding to the target identification fault in the stored corresponding relation. And then, the equipment fault database synthesizes fault parameter characteristics of the target identification fault, matches with historical fault knowledge, calculates the matching degree of each possible target component, and obtains the target component with the highest matching degree, and defines the target component as a preset operation and maintenance component.
By statistically analyzing the historical maintenance knowledge of the target equipment, a knowledge base of the corresponding relation between the faults and the maintenance components is established, and when the target identifies the fault database of the fault input equipment, the fault database is quickly matched with the most likely preset operation and maintenance components, so that important references are provided for equipment maintenance work, and the intellectualization of fault processing is realized.
Step S600: acquiring a target operation and maintenance record of the target equipment in a target time period, and extracting a first operation and maintenance record image in the target operation and maintenance record;
specifically, the target time period refers to a fault repair time period set by the system according to the historical operation and maintenance record and the preset operation and maintenance component after the target identification fault is judged. After the predetermined operation and maintenance component is obtained, the system searches the historical operation and maintenance record of the predetermined operation and maintenance component in the equipment fault database according to the predetermined operation and maintenance component, obtains average operation and maintenance time, obtains a target time period, predicts that the maintenance of the target identification fault can be completed in the time period, and the starting time point of the time period is the time point for determining the predetermined operation and maintenance component. The target operation and maintenance record refers to an overhaul report, an overhaul record and the like of the target equipment in a target time period, and contains detailed information of an equipment overhaul process, such as overhaul time, overhaul personnel, equipment states, processing methods, replacement parts and the like.
First, after acquiring a target time period, the system sends the time range to the cloud computing center. Then, the cloud computing center acquires all operation and maintenance records of the target device in the target time period from the database as target operation and maintenance records. Then, the last operation and maintenance record containing the equipment image information is selected from all the target operation and maintenance records to be used as the first operation and maintenance record. Then, an image of the operating state of the device is extracted from the first operation-dimension record as a first operation-dimension record image. The first operation and maintenance record image is the operation state of the target equipment at the ending time point of the target time period, and provides image information for judging whether the preset operation and maintenance component is operated and maintained.
By acquiring the target operation and maintenance record of the target equipment in the target time period and selecting the first operation and maintenance record image from the target operation and maintenance record, visual equipment operation state information is provided, compared with numerical data, the operation state and change trend of the equipment can be comprehensively mastered, and a basis is provided for operation and maintenance judgment.
Step S700: analyzing the first operation and maintenance record image through a digital twin network model to obtain a first image distinguishing feature set;
specifically, the digital twin network model is an image recognition model constructed by a machine learning method, is obtained by training historical operation and maintenance image data of target equipment, can automatically analyze newly input operation and maintenance images, and recognizes the change characteristics of the target equipment in the images in different application periods.
First, a historical operation and maintenance record image of the target device is collected, device characteristics in the image, such as colors, shapes, structures and the like, are marked, and an image characteristic data set is constructed. And then, using a ResNet deep learning network, training through an image characteristic data set to obtain a digital twin network model, so that the model can automatically extract the characteristics of a new input image and identify the change between images of different time periods. And then, inputting the acquired first operation and maintenance record image into a digital twin network model, analyzing the first operation and maintenance record image by the model, extracting the characteristics of target equipment in the image, such as the appearance, the color, the surface state and the like of the equipment, and obtaining a first image characteristic set. And then, comparing the first image feature set with the image feature set in the normal running state of the target equipment to obtain a first image distinguishing feature set. The first image distinguishing feature set reflects the image difference of the target device from the first operation and maintenance recording image and the normal operation image feature of the device, and provides a basis for judging whether the device is operated and maintained.
The digital twin network model constructed by the machine learning method automatically analyzes the operation and maintenance recorded image, identifies the equipment characteristics in the image and judges the characteristic differences to obtain an image distinguishing characteristic set, realizes the monitoring and analysis of the running state of the equipment, and provides powerful support for ensuring the safe and stable running of the equipment.
Step S800: and sending out first reminding information according to the first image distinguishing feature set if the preset operation and maintenance component is not operated and maintained, wherein the first reminding information is used for reminding the operation and maintenance management of the preset operation and maintenance component.
Specifically, whether the scheduled operation and maintenance component needs to be overhauled is judged according to the first image distinguishing characteristic set. If abnormal changes of the scheduled operation and maintenance components are found but operation and maintenance processing is not carried out in the target time period, reminding information is sent out so as not to influence normal operation of equipment.
When the object equipment is not in the normal running state in the characteristic distinction in the first image distinguishing characteristic set, the object equipment is indicated that after the object identification fault is detected and the preset operation and maintenance component is obtained, the operation and maintenance are not carried out on the preset operation and maintenance component in the recent period (the object time period), and the abnormal condition of the component is indicated to be not maintained. At the moment, a first reminding message is sent out to prompt related personnel to overhaul the scheduled operation and maintenance parts as soon as possible. The first reminding information is displayed to the user through the human-computer interaction interface, and the content comprises a reason for reminding the user of overhauling the scheduled operation and maintenance part, an overhauling proposal scheme and the like. After receiving the first reminding information, the user checks and maintains the scheduled operation and maintenance part in time, so that the production is prevented from being influenced by the failure of the equipment. Besides displaying the reminding information, the system also directly sends the reminding information to related maintenance personnel in a mode of mobile phone short messages or mails so as to ensure that the reminding information can be received and processed in time.
When the abnormal condition that the equipment needs to be processed but is not maintained in time is identified, the reminding information is sent to prompt, the equipment state aversion caused by the management loophole is avoided, the response time of the abnormal processing is shortened to the maximum extent, the safe and stable operation of key equipment is ensured, and the intelligent equipment operation and maintenance management is realized.
Further, as shown in fig. 2, the embodiment of the present application further includes:
step S410: reading a first operation index of the target equipment;
step S420: matching a first index parameter and a second index parameter of the first operation index in the first equipment operation data and the second equipment operation data respectively;
step S430: calculating a first parameter deviation of the first index parameter and the second index parameter;
step S440: if the first parameter deviation does not accord with a first preset deviation threshold value, adding the second index parameter to a list to be analyzed;
step S450: and analyzing each index parameter in the list to be analyzed through the fault identification model to obtain the target identification fault.
Specifically, first operation indexes of the target device, such as the rotation speed, the temperature, the pressure, etc., which can reflect the operation state of the device, are read. And then, respectively matching a first index parameter and a second index parameter corresponding to the first operation index in the first equipment operation data and the second equipment operation data, wherein the first index parameter represents the value of the first operation index of the target equipment under the first application time, and the second index parameter represents the value of the first operation index of the target equipment under the second application time. Then, a difference value between the first index parameter and the second index parameter, namely a first parameter deviation, is calculated, and if the first parameter deviation exceeds a preset first preset deviation threshold value, the value of the first operation index is indicated to be greatly changed at two application points, and the abnormal state of equipment is indicated. At this time, the second index parameters are added to the to-be-analyzed list, and the to-be-analyzed list stores all index parameters needing deep analysis. And then, comprehensively analyzing each index parameter in the to-be-analyzed list through a fault identification model, and identifying the most likely existing fault type of the target equipment to obtain the target identification fault.
By analyzing the parameter changes of the key operation indexes, abnormal parameters are found, and the abnormal parameters are judged through the fault identification model, so that the accurate diagnosis of equipment faults is realized.
Further, the embodiment of the application further comprises:
step S451: collecting historical application fault logs of similar equipment products of the target equipment, and extracting a first application fault log in the historical application fault logs;
the first application fault log comprises a plurality of abnormal index parameters with first fault identifications;
step S452: and performing supervised learning and inspection by taking the plurality of abnormal index parameters with the first fault identification as a training data set to obtain the fault identification model.
Specifically, firstly, an application log and a fault report log of similar equipment products of target equipment are collected, and a historical application fault log is extracted from the application log and comprises abnormal parameter information of various faults. The historical application fault log comprises a plurality of application fault logs, each log corresponds to one fault, a plurality of parameters are abnormal in each fault, and the same fault identification is carried out on the abnormal parameters in each fault. Then, a fault log is extracted from the historical application fault log to serve as a first application fault log, the first application fault log represents occurrence of a fault, the first application fault log comprises a plurality of index parameters with target equipment when the fault occurs, and abnormal indexes have the same fault identifier, namely a first fault identifier.
Then, selecting a supervised learning algorithm suitable for fault recognition, such as SVM, BP neural network, random forest and the like; dividing the training data set into a training set and a verification set, wherein 80% of the training data set is used for the training set and 20% of the training data set is used for the verification set; the training set is input into a selected machine learning algorithm for training, the algorithm establishes a mapping relation between the abnormal parameters and the fault identification by learning the internal association among the training samples, and finally the fault identification model is obtained through training. Subsequently, the verification set is input into the fault recognition model, and the fault recognition capability of the model is evaluated. If the recognition accuracy reaches a preset threshold, finishing model training; otherwise, the model parameters are adjusted to be retrained.
The fault recognition model is obtained by training and verifying the historical fault data by using a supervised learning algorithm, so that intelligent judgment and fault recognition of the equipment operation parameters are realized, and powerful guarantee is provided for safe and stable operation of the target equipment.
Further, as shown in fig. 3, the embodiment of the present application further includes:
step S510: acquiring a first fault component in the first application fault log;
step S520: establishing a first mapping relation between the first fault component and the first fault identification;
Step S530: and constructing the equipment fault database based on the first mapping relation.
Specifically, to construct the equipment failure database, first, a repair order or a repair report corresponding to the first application failure log is acquired, and the actually replaced failure component is resolved therefrom as the first failure component. The first fault component is a component or a component for repairing a fault corresponding to the first application fault log, and then a mapping relationship is established between the first fault component and a first fault identifier given in the first application fault log, which is defined as a first mapping relationship, and is used for associating a plurality of abnormal parameters with the first fault identifier with the corresponding fault component. And repeatedly constructing mapping relations between fault identification parameters and fault components in all application fault logs of the historical application fault logs, and summarizing all the mapping relations to construct a device fault database.
And (3) extracting the corresponding relation between the faults and the replaced parts by analyzing the historical fault log and the maintenance record, and constructing an equipment fault database. The method realizes the effective application of the historical fault data of the equipment and provides important support for rapid and accurate equipment fault diagnosis and response.
Further, the embodiment of the application further comprises:
step S710: collecting historical operation and maintenance records of similar equipment products of the target equipment, and extracting a first historical operation and maintenance record in the historical operation and maintenance records;
step S720: acquiring an image feature set of a first historical operation and maintenance record image in the first historical operation and maintenance record;
step S730: the digital twin network model is built based on the image feature set, wherein the digital twin network model includes a first sub-model and a second sub-model.
Specifically, in order to construct a digital twin network model, first, information such as operation and maintenance reports, records and images of similar products of target equipment is collected to form a historical operation and maintenance record, and one operation and maintenance record is selected as a first historical operation and maintenance record. The first historical operation and maintenance record comprises operation state images of the equipment before and after maintenance as a first historical operation and maintenance record image. And then, carrying out feature extraction on the first historical operation and maintenance record image to obtain color features, texture features and the like, and forming an image feature set. Then, a convolutional neural network algorithm is selected as a machine learning method, and a digital twin network model is constructed, wherein the digital twin network model comprises a first sub-model and a second sub-model, and the first sub-model is used for extracting features of an image before maintenance in a first historical operation and maintenance record image; the second sub-model is used for extracting features of the images before maintenance in the first historical operation and maintenance record images. Then, the digital twin network model obtains a first image distinguishing feature set by comparing the image features respectively extracted from the first sub-model and the second sub-model.
The digital twin network model is constructed based on the historical operation and maintenance image of the target equipment, so that the model can automatically extract and judge the characteristic change of the newly input operation and maintenance image, the supervision and judgment of the operation state of the equipment are realized, and a powerful support is provided for judging whether the fault component of the target equipment is operated and maintained.
Further, the embodiment of the application further comprises:
step S740: analyzing a first image in the first operation and maintenance record image through the first sub-model to obtain a first image feature set, wherein the first image feature set comprises a first color feature and a first texture feature;
step S750: analyzing a second image in the first operation and maintenance record image through the second sub-model to obtain a second image feature set, wherein the second image feature set comprises a second color feature and a second texture feature;
step S760: inputting the first color feature and the second color feature into a color comparison unit to obtain a first color distinguishing feature, and inputting the first texture feature and the second texture feature into a texture comparison unit to obtain a first texture distinguishing feature;
step S770: the first color distinguishing feature and the first texture distinguishing feature constitute the first image distinguishing feature set.
Specifically, firstly, an acquired first operation and maintenance record image is input into a digital twin network model for analysis, wherein the first operation and maintenance record image comprises a first image and a second image, the first image is an operation image of the target equipment before maintenance in the operation and maintenance record in an abnormal state, and the second image is an operation image of the target equipment after maintenance in the operation and maintenance record in a normal state. And then, the digital twin network model inputs a first image in the input first operation and maintenance recorded image into a first sub-model, the first sub-model analyzes the first image, extracts the color features and the texture features of the first image, obtains the first color features and the first texture features, and forms a first image feature set. Wherein the color features represent the dominant color distribution of the image and the texture features represent the texture information of the image. And similarly, analyzing a second image in the first operation and maintenance recorded image through a second sub-model, extracting color features and texture features, obtaining second color features and second texture features, and forming a second image feature set.
And then, inputting the first color feature in the first image feature set and the second color feature in the second image feature set into a color comparison unit, and detecting the difference between the corresponding color features in the first image feature set and the second image feature set by the color comparison unit by adopting the Euclidean distance method to obtain a first color distinguishing feature representing color change. And similarly, inputting the first texture feature in the first image feature set and the second texture feature in the second image feature set into a texture comparison unit, and detecting the difference between the corresponding texture features in the first image feature set and the second image feature set by the texture comparison unit by adopting the same method to obtain a first texture distinguishing feature representing texture change. Finally, the combination of the first color distinguishing feature and the first texture distinguishing feature forms a first image distinguishing feature set so as to reflect complete change information of the state of the target equipment from before maintenance to after maintenance, and provide basis for judging whether the fault of the target equipment is operated and maintained.
Further, the embodiment of the application further comprises:
step S781: reading a first predetermined loss function, wherein the formula of the first predetermined loss function is as follows:
；
；
；
wherein,-said first predetermined loss function, which refers to said first image distinguishing feature set,>color loss data, which refers to said first color distinguishing feature,/for>Texture penalty data, which refers to said first texture distinguishing feature,>refers to the color feature similarity of the first color feature and the second color feature,refers to a preset color feature similarity of the first color feature and the second color feature,means that the texture feature similarity of said first texture feature and said second texture feature,/-or #>Means a preset texture similarity of said first texture and said second texture,/-or #>Coefficients of the color loss data and the texture loss data, respectively, and +.>；
Step S782: performing loss analysis on the first image distinguishing feature set based on the first preset loss function to obtain first loss data;
step S783: the first image difference feature set is adjusted based on the first loss data.
Specifically, to improve the accuracy of the first image distinguishing feature set, its error is analyzed based on a predefined loss function, and further adjustment optimization is performed. Firstly, a preset first preset loss function is read, and the calculation formula of the loss function is as follows:
；
；
；
Wherein,a loss function value representing a first image distinguishing feature set; />Color loss data representing a first color distinguishing feature; />Texture penalty data representing a first texture distinguishing feature; />Color feature similarity representing the first color feature and the second color feature; />A preset color feature similarity representing the first color feature and the second color feature; />Representing a texture feature similarity of the first texture feature and the second texture feature; />Representing a preset texture similarity of the first texture and the second texture; />And->Weight coefficients of color loss data and texture loss data, respectively, and +.>Is empirically set by the equipment operation and maintenance expert.
Then, substituting the first color distinguishing feature and the first texture distinguishing feature in the first image distinguishing feature set into a first preset loss function, and calculating to obtain first loss data so as to represent the error degree of the first image distinguishing feature set, wherein the error degree is used for judging whether the first image distinguishing feature set is accurate or not. If the first loss data exceeds a preset threshold value, the first image distinguishing feature set has larger error and needs to be adjusted and optimized. Based on the first loss data, the first image distinguishing feature set is modified for the feature causing the larger loss, such as reducing the weight of a certain feature or recalculating the feature difference, etc.
And carrying out error evaluation on the first image distinguishing feature set through a predefined loss function, and adjusting based on an evaluation result, so that the accuracy and reliability of the first image distinguishing feature set are effectively improved. The finally obtained optimized first image distinguishing feature set can accurately train the digital twin network model, and the accuracy of the operation and maintenance of the factory equipment is improved.
In summary, the intelligent operation and maintenance method for the factory equipment provided by the embodiment of the application has the following technical effects:
acquiring a target application log of target equipment in a factory equipment terminal, wherein the target application log comprises a plurality of groups of equipment operation data with application time identifiers, and providing data support for identifying faults of the target equipment; acquiring a first application time, and matching first equipment operation data under the first application time in a plurality of groups of equipment operation data with application time identifiers; acquiring second application time, and matching second equipment operation data under the second application time in a plurality of groups of equipment operation data with application time identifiers, wherein the second application time and the first application time are adjacent twice application time, and providing input data for fault identification model analysis by acquiring the first equipment operation data and the second equipment operation data; analyzing the first equipment operation data and the second equipment operation data through a fault identification model to obtain a target identification fault, realizing automatic identification of equipment faults, and improving the accuracy and the identification efficiency of fault identification; matching a target component corresponding to the target identification fault in the equipment fault database, taking the target component as a preset operation and maintenance component, and providing a basis for judging whether to carry out operation and maintenance reminding or not in the follow-up; acquiring a target operation and maintenance record of target equipment in a target time period, extracting a first operation and maintenance record image in the target operation and maintenance record, and providing input data for digital twin network model analysis; analyzing the first operation and maintenance record image through a digital twin network model to obtain a first image distinguishing characteristic set, and providing characteristic reference for judging whether the operation and maintenance of the preset operation and maintenance component is carried out or not; and sending out first reminding information according to the first image distinguishing feature set if the preset operation and maintenance component is not operated and maintained, wherein the first reminding information is used for reminding operation and maintenance management on the preset operation and maintenance component, so that the operation and maintenance efficiency is improved, and the safe and stable operation of the target equipment is ensured. The method realizes the whole process technical scheme of equipment fault identification, operation and maintenance part judgment and corresponding reminding, and achieves the effects of improving the accuracy of fault identification and operation and maintenance efficiency, reducing operation and maintenance cost and guaranteeing the reliable and stable operation of target equipment.
Example two
Based on the same inventive concept as the intelligent operation and maintenance method of the factory equipment in the foregoing embodiment, as shown in fig. 4, an embodiment of the present application provides an intelligent operation and maintenance system of the factory equipment, where the system includes:
an application log obtaining module 11, configured to obtain a target application log of a target device in a factory device terminal, where the target application log includes a plurality of sets of device operation data with application time identifiers;
the first data matching module 12 is configured to obtain a first application time, and match first device operation data under the first application time in the multiple sets of device operation data with application time identifiers;
a second data matching module 13, configured to obtain a second application time, and match second device operation data under the second application time in the multiple sets of device operation data with application time identifiers, where the second application time and the first application time are adjacent two application times;
an operation data analysis module 14, configured to analyze the first device operation data and the second device operation data through a fault recognition model, so as to obtain a target recognition fault;
The operation and maintenance component matching module 15 is configured to match a target component corresponding to the target identification fault in an equipment fault database, and take the target component as a predetermined operation and maintenance component;
an operation and maintenance record obtaining module 16, configured to obtain a target operation and maintenance record of the target device in a target time period, and extract a first operation and maintenance record image in the target operation and maintenance record;
the recorded image analysis module 17 is configured to analyze the first operation and maintenance recorded image through a digital twin network model, so as to obtain a first image distinguishing feature set;
the operation and maintenance management reminding module 18 is configured to send out first reminding information if the predetermined operation and maintenance component is not operated and maintained according to the first image distinguishing feature set, where the first reminding information is used to remind to perform operation and maintenance management on the predetermined operation and maintenance component.
Further, the operation data analysis module 14 includes the following steps:
reading a first operation index of the target equipment;
matching a first index parameter and a second index parameter of the first operation index in the first equipment operation data and the second equipment operation data respectively;
calculating a first parameter deviation of the first index parameter and the second index parameter;
If the first parameter deviation does not accord with a first preset deviation threshold value, adding the second index parameter to a list to be analyzed;
and analyzing each index parameter in the list to be analyzed through the fault identification model to obtain the target identification fault.
Further, the operation data analysis module 14 further includes the following steps:
collecting historical application fault logs of similar equipment products of the target equipment, and extracting a first application fault log in the historical application fault logs;
the first application fault log comprises a plurality of abnormal index parameters with first fault identifications;
and performing supervised learning and inspection by taking the plurality of abnormal index parameters with the first fault identification as a training data set to obtain the fault identification model.
Further, the operation and maintenance component matching module 15 includes the following steps:
acquiring a first fault component in the first application fault log;
establishing a first mapping relation between the first fault component and the first fault identification;
and constructing the equipment fault database based on the first mapping relation.
Further, the recorded image analysis module 17 includes the following steps:
Collecting historical operation and maintenance records of similar equipment products of the target equipment, and extracting a first historical operation and maintenance record in the historical operation and maintenance records;
acquiring an image feature set of a first historical operation and maintenance record image in the first historical operation and maintenance record;
the digital twin network model is built based on the image feature set, wherein the digital twin network model includes a first sub-model and a second sub-model.
Further, the recorded image analysis module 17 further includes the following steps:
analyzing a first image in the first operation and maintenance record image through the first sub-model to obtain a first image feature set, wherein the first image feature set comprises a first color feature and a first texture feature;
analyzing a second image in the first operation and maintenance record image through the second sub-model to obtain a second image feature set, wherein the second image feature set comprises a second color feature and a second texture feature;
inputting the first color feature and the second color feature into a color comparison unit to obtain a first color distinguishing feature, and inputting the first texture feature and the second texture feature into a texture comparison unit to obtain a first texture distinguishing feature;
The first color distinguishing feature and the first texture distinguishing feature constitute the first image distinguishing feature set.
Further, the recorded image analysis module 17 further includes the following steps:
reading a first predetermined loss function, wherein the formula of the first predetermined loss function is as follows:
；
；
；
wherein,-said first predetermined loss function, which refers to said first image distinguishing feature set,>color loss data, which refers to said first color distinguishing feature,/for>Texture penalty data, which refers to said first texture distinguishing feature,>refers to the color feature similarity of the first color feature and the second color feature,refers to a preset color feature similarity of the first color feature and the second color feature,means that the texture feature similarity of said first texture feature and said second texture feature,/-or #>Means a preset texture similarity of said first texture and said second texture,/-or #>Coefficients of the color loss data and the texture loss data, respectively, and +.>；
Performing loss analysis on the first image distinguishing feature set based on the first preset loss function to obtain first loss data;
The first image difference feature set is adjusted based on the first loss data.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any method for implementing an embodiment of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (5)
1. An intelligent operation and maintenance method for factory equipment is characterized by comprising the following steps:
obtaining a target application log of target equipment in a factory equipment terminal, wherein the target application log comprises a plurality of groups of equipment operation data with application time identifiers;
acquiring a first application time, and matching first equipment operation data under the first application time in the plurality of groups of equipment operation data with the application time identifier;
Acquiring second application time, and matching second equipment operation data under the second application time in the plurality of groups of equipment operation data with the identification of the application time, wherein the second application time and the first application time are adjacent twice application time;
analyzing the first equipment operation data and the second equipment operation data through a fault identification model to obtain target identification faults;
matching a target component corresponding to the target identification fault in an equipment fault database, and taking the target component as a preset operation and maintenance component;
acquiring a target operation and maintenance record of the target equipment in a target time period, and extracting a first operation and maintenance record image in the target operation and maintenance record;
analyzing the first operation and maintenance record image through a digital twin network model to obtain a first image distinguishing feature set;
according to the first image distinguishing feature set, if the preset operation and maintenance component is not operated and maintained, sending first reminding information, wherein the first reminding information is used for reminding the operation and maintenance management of the preset operation and maintenance component;
before said analyzing said first operation and maintenance record image by means of a digital twin network model, comprising:
Collecting historical operation and maintenance records of similar equipment products of the target equipment, and extracting a first historical operation and maintenance record in the historical operation and maintenance records;
acquiring an image feature set of a first historical operation and maintenance record image in the first historical operation and maintenance record;
building the digital twin network model based on the image feature set, wherein the digital twin network model comprises a first sub-model and a second sub-model;
the analyzing the first operation and maintenance record image through the digital twin network model to obtain a first image distinguishing feature set comprises the following steps:
analyzing a first image in the first operation and maintenance record image through the first sub-model to obtain a first image feature set, wherein the first image feature set comprises a first color feature and a first texture feature;
analyzing a second image in the first operation and maintenance record image through the second sub-model to obtain a second image feature set, wherein the second image feature set comprises a second color feature and a second texture feature;
inputting the first color feature and the second color feature into a color comparison unit to obtain a first color distinguishing feature, and inputting the first texture feature and the second texture feature into a texture comparison unit to obtain a first texture distinguishing feature;
The first color distinguishing feature and the first texture distinguishing feature constitute the first image distinguishing feature set;
after the first color distinguishing feature and the first texture distinguishing feature constitute the first image distinguishing feature set, further comprising:
reading a first predetermined loss function, wherein the formula of the first predetermined loss function is as follows:
；
；
；
wherein,-said first predetermined loss function, which refers to said first image distinguishing feature set,>for the first color feature, +.>For the second color feature, +.>Color loss data, which refers to said first color distinguishing feature,/for>Texture penalty data, which refers to said first texture distinguishing feature,>means that the color feature similarity of said first color feature and said second color feature,/->Means a preset color feature similarity of said first color feature and said second color feature,/->Means that the texture feature similarity of said first texture feature and said second texture feature,/-or #>Means a preset texture similarity of said first texture and said second texture,/-or #>The color loss data and the texture, respectivelyCoefficients of the data are lost, and ∈ >；
Performing loss analysis on the first image distinguishing feature set based on the first preset loss function to obtain first loss data;
the first image difference feature set is adjusted based on the first loss data.
2. The method of claim 1, wherein analyzing the first device operational data and the second device operational data via a fault identification model to obtain a target identification fault comprises:
reading a first operation index of the target equipment;
matching a first index parameter and a second index parameter of the first operation index in the first equipment operation data and the second equipment operation data respectively;
calculating a first parameter deviation of the first index parameter and the second index parameter;
if the first parameter deviation does not accord with a first preset deviation threshold value, adding the second index parameter to a list to be analyzed;
and analyzing each index parameter in the list to be analyzed through the fault identification model to obtain the target identification fault.
3. The method according to claim 2, wherein before said analyzing each index parameter in said list to be analyzed by said fault identification model to obtain said target identification fault, the method comprises:
Collecting historical application fault logs of similar equipment products of the target equipment, and extracting a first application fault log in the historical application fault logs;
the first application fault log comprises a plurality of abnormal index parameters with first fault identifications;
and performing supervised learning and inspection by taking the plurality of abnormal index parameters with the first fault identification as a training data set to obtain the fault identification model.
4. A method according to claim 3, wherein before matching the target component corresponding to the target identification fault in the equipment fault database, the method comprises:
acquiring a first fault component in the first application fault log;
establishing a first mapping relation between the first fault component and the first fault identification;
and constructing the equipment fault database based on the first mapping relation.
5. An intelligent operation and maintenance system for a factory floor device, for implementing an intelligent operation and maintenance method for a factory floor device according to any of claims 1-4, the system comprising:
the system comprises an application log acquisition module, a control module and a control module, wherein the application log acquisition module is used for acquiring a target application log of target equipment in a factory equipment terminal, and the target application log comprises a plurality of groups of equipment operation data with application time identifiers;
The first data matching module is used for acquiring first application time and matching first equipment operation data under the first application time in the plurality of groups of equipment operation data with application time identifiers;
the second data matching module is used for acquiring second application time and matching second equipment operation data under the second application time in the plurality of groups of equipment operation data with the identification of the application time, wherein the second application time and the first application time are the time of two adjacent applications;
the operation data analysis module is used for analyzing the operation data of the first equipment and the operation data of the second equipment through a fault identification model to obtain a target identification fault;
the operation and maintenance component matching module is used for matching the target component corresponding to the target identification fault in the equipment fault database and taking the target component as a preset operation and maintenance component;
the operation and maintenance record acquisition module is used for acquiring a target operation and maintenance record of the target equipment in a target time period and extracting a first operation and maintenance record image in the target operation and maintenance record;
The recorded image analysis module is used for analyzing the first operation and maintenance recorded image through a digital twin network model to obtain a first image distinguishing characteristic set;
the operation and maintenance management reminding module is used for distinguishing the feature set according to the first image, and sending first reminding information if the preset operation and maintenance component is not operated and maintained, wherein the first reminding information is used for reminding operation and maintenance management on the preset operation and maintenance component;
the recorded image analysis module comprises the following execution steps:
collecting historical operation and maintenance records of similar equipment products of the target equipment, and extracting a first historical operation and maintenance record in the historical operation and maintenance records;
acquiring an image feature set of a first historical operation and maintenance record image in the first historical operation and maintenance record;
building the digital twin network model based on the image feature set, wherein the digital twin network model comprises a first sub-model and a second sub-model;
the recorded image analysis module further comprises the following execution steps:
analyzing a first image in the first operation and maintenance record image through the first sub-model to obtain a first image feature set, wherein the first image feature set comprises a first color feature and a first texture feature;
Analyzing a second image in the first operation and maintenance record image through the second sub-model to obtain a second image feature set, wherein the second image feature set comprises a second color feature and a second texture feature;
inputting the first color feature and the second color feature into a color comparison unit to obtain a first color distinguishing feature, and inputting the first texture feature and the second texture feature into a texture comparison unit to obtain a first texture distinguishing feature;
the first color distinguishing feature and the first texture distinguishing feature constitute the first image distinguishing feature set;
reading a first predetermined loss function, wherein the formula of the first predetermined loss function is as follows:
；
；
；
wherein,-said first predetermined loss function, which refers to said first image distinguishing feature set,>for the first color feature, +.>For the second color feature, +.>Color loss data, which refers to said first color distinguishing feature,/for>Refers to whatTexture penalty data for said first texture distinguishing feature, < >>Means that the color feature similarity of said first color feature and said second color feature,/->Means a preset color feature similarity of said first color feature and said second color feature,/- >Means that the texture feature similarity of said first texture feature and said second texture feature,/-or #>Means a preset texture similarity of said first texture and said second texture,/-or #>Coefficients of the color loss data and the texture loss data, respectively, and +.>；
Performing loss analysis on the first image distinguishing feature set based on the first preset loss function to obtain first loss data;
the first image difference feature set is adjusted based on the first loss data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311108904.1A CN116821403B (en) | 2023-08-31 | 2023-08-31 | Intelligent operation and maintenance method and system for factory equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311108904.1A CN116821403B (en) | 2023-08-31 | 2023-08-31 | Intelligent operation and maintenance method and system for factory equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116821403A CN116821403A (en) | 2023-09-29 |
| CN116821403B true CN116821403B (en) | 2023-12-05 |
Family
ID=88127851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311108904.1A Active CN116821403B (en) | 2023-08-31 | 2023-08-31 | Intelligent operation and maintenance method and system for factory equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116821403B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106779421A (en) * | 2016-12-21 | 2017-05-31 | 国网上海市电力公司 | A kind of electric energy meter collection operation management system based on GPS location |
| CN106846815A (en) * | 2017-03-29 | 2017-06-13 | 深圳万智联合科技有限公司 | A kind of efficient intelligent traffic administration system big data analysis system |
| CN112187547A (en) * | 2020-10-09 | 2021-01-05 | 南京邮电大学 | Network model based on digital twins |
| WO2022095616A1 (en) * | 2020-11-03 | 2022-05-12 | 国网智能科技股份有限公司 | On-line intelligent inspection system and method for transformer substation |
| CN115375172A (en) * | 2022-09-07 | 2022-11-22 | 重庆千变速维科技有限公司 | BIM-based intelligent operation and maintenance method and system |
| CN115412947A (en) * | 2022-08-26 | 2022-11-29 | 武汉烽火技术服务有限公司 | Fault simulation method and system based on digital twin and AI algorithm |
| CN115484173A (en) * | 2021-05-28 | 2022-12-16 | 中国移动通信有限公司研究院 | Flow simulation method of digital twin network and digital twin network |
| CN115936676A (en) * | 2022-12-16 | 2023-04-07 | 江苏龙兴航宇智能科技有限公司 | Equipment health management method and system based on digital twins |
| CN115981984A (en) * | 2023-01-06 | 2023-04-18 | 郑州云海信息技术有限公司 | Equipment fault detection method, device, equipment and storage medium |
| CN116091901A (en) * | 2023-01-31 | 2023-05-09 | 中煤科工集团重庆智慧城市科技研究院有限公司 | Operation and maintenance management platform based on fault diagnosis and analysis of monitoring equipment |
| CN116389316A (en) * | 2023-04-23 | 2023-07-04 | 广东电网有限责任公司 | Fault sensing method, device, equipment and storage medium for power optical communication network |
| CN116390138A (en) * | 2023-04-25 | 2023-07-04 | 中南大学 | Fault diagnosis method based on digital twin network and related equipment |
| CN116415851A (en) * | 2023-04-12 | 2023-07-11 | 湖南海龙国际智能科技股份有限公司 | Intelligent identification and evaluation method for facility operation and maintenance performance indexes based on deep learning |
| CN116562848A (en) * | 2023-05-05 | 2023-08-08 | 江西意孚欧科技有限公司 | Operation and maintenance management platform |
| CN116614344A (en) * | 2023-05-29 | 2023-08-18 | 北京城建智控科技股份有限公司 | Fault diagnosis method and system based on digital twinning |
-
2023
- 2023-08-31 CN CN202311108904.1A patent/CN116821403B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106779421A (en) * | 2016-12-21 | 2017-05-31 | 国网上海市电力公司 | A kind of electric energy meter collection operation management system based on GPS location |
| CN106846815A (en) * | 2017-03-29 | 2017-06-13 | 深圳万智联合科技有限公司 | A kind of efficient intelligent traffic administration system big data analysis system |
| CN112187547A (en) * | 2020-10-09 | 2021-01-05 | 南京邮电大学 | Network model based on digital twins |
| WO2022095616A1 (en) * | 2020-11-03 | 2022-05-12 | 国网智能科技股份有限公司 | On-line intelligent inspection system and method for transformer substation |
| CN115484173A (en) * | 2021-05-28 | 2022-12-16 | 中国移动通信有限公司研究院 | Flow simulation method of digital twin network and digital twin network |
| CN115412947A (en) * | 2022-08-26 | 2022-11-29 | 武汉烽火技术服务有限公司 | Fault simulation method and system based on digital twin and AI algorithm |
| CN115375172A (en) * | 2022-09-07 | 2022-11-22 | 重庆千变速维科技有限公司 | BIM-based intelligent operation and maintenance method and system |
| CN115936676A (en) * | 2022-12-16 | 2023-04-07 | 江苏龙兴航宇智能科技有限公司 | Equipment health management method and system based on digital twins |
| CN115981984A (en) * | 2023-01-06 | 2023-04-18 | 郑州云海信息技术有限公司 | Equipment fault detection method, device, equipment and storage medium |
| CN116091901A (en) * | 2023-01-31 | 2023-05-09 | 中煤科工集团重庆智慧城市科技研究院有限公司 | Operation and maintenance management platform based on fault diagnosis and analysis of monitoring equipment |
| CN116415851A (en) * | 2023-04-12 | 2023-07-11 | 湖南海龙国际智能科技股份有限公司 | Intelligent identification and evaluation method for facility operation and maintenance performance indexes based on deep learning |
| CN116389316A (en) * | 2023-04-23 | 2023-07-04 | 广东电网有限责任公司 | Fault sensing method, device, equipment and storage medium for power optical communication network |
| CN116390138A (en) * | 2023-04-25 | 2023-07-04 | 中南大学 | Fault diagnosis method based on digital twin network and related equipment |
| CN116562848A (en) * | 2023-05-05 | 2023-08-08 | 江西意孚欧科技有限公司 | Operation and maintenance management platform |
| CN116614344A (en) * | 2023-05-29 | 2023-08-18 | 北京城建智控科技股份有限公司 | Fault diagnosis method and system based on digital twinning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116821403A (en) | 2023-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111027615B (en) | Middleware fault early warning method and system based on machine learning | |
| CN114723285B (en) | Power grid equipment safety evaluation prediction method | |
| CN115630839B (en) | Intelligent feedback production regulation and control system based on data mining | |
| CN115358155A (en) | Power big data abnormity early warning method, device, equipment and readable storage medium | |
| CN113570200A (en) | Power grid operation state monitoring method and system based on multidimensional information | |
| CN116028887B (en) | Analysis method of continuous industrial production data | |
| CN111159487A (en) | Predictive maintenance intelligent system for automobile engine spindle | |
| CN113253037A (en) | Current ripple-based edge cloud cooperative equipment state monitoring method and system and medium | |
| CN117056688A (en) | New material production data management system and method based on data analysis | |
| CN117390529A (en) | Multi-factor traceable data center information management method | |
| CN116483015A (en) | Workshop equipment monitoring method, device, equipment and storage medium | |
| CN116821403B (en) | Intelligent operation and maintenance method and system for factory equipment | |
| CN117312879A (en) | Injection molding machine production data supervision and early warning method, system and medium | |
| CN115186935B (en) | Electromechanical device nonlinear fault prediction method and system | |
| CN116704729A (en) | Industrial kiln early warning system and method based on big data analysis | |
| CN115392663A (en) | Data acquisition and processing method based on big data | |
| CN114764867A (en) | Fan fault diagnosis system and method based on image main feature extraction and application | |
| CN110569277A (en) | Method and system for automatically identifying and classifying configuration data information | |
| CN111539642A (en) | Power plant data acquisition and processing system and method based on object-oriented | |
| CN115018332B (en) | Visual management system of thermal power plant based on virtual reality | |
| CN117826771B (en) | Cold rolling mill control system abnormality detection method and system based on AI analysis | |
| CN116760033B (en) | Real-time power demand prediction system based on artificial intelligence | |
| CN117692610B (en) | AR workshop inspection system | |
| CN117371985A (en) | Unmanned aerial vehicle power inspection report production method | |
| CN117609923A (en) | Electronic information processing system and method for Internet of things |
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 |