CN117236934B

CN117236934B - Industrial Internet remote monitoring operation and maintenance management system

Info

Publication number: CN117236934B
Application number: CN202311437702.1A
Authority: CN
Inventors: 李洪升; 陈志浩; 崔媛媛; 张莹; 赵登军
Original assignee: Shandong Jingwei Information Group Co ltd
Current assignee: Shandong Jingwei Information Group Co ltd
Priority date: 2023-11-01
Filing date: 2023-11-01
Publication date: 2024-05-07
Anticipated expiration: 2043-11-01
Also published as: CN117236934A

Abstract

The invention discloses an industrial Internet remote monitoring operation and maintenance management system, which comprises an operation and maintenance knowledge base module, a management module and a management module, wherein the operation and maintenance knowledge base module is used for inputting and storing professional knowledge and experience knowledge of daily operation and maintenance; the parameter acquisition module is used for acquiring data of the equipment and preprocessing the data; the inspection management module is used for performing intelligent inspection management on the operation and maintenance process; the fault matching module is used for performing fault matching; the AI diagnostic module is used for giving a repairing scheme according to the fault matching result; the operation and maintenance dispatch module is used for issuing operation and maintenance tasks according to the fault matching result; the equipment management module is used for automatically matching the operation and maintenance articles according to the operation and maintenance tasks and the repairing scheme, and registering articles in and out of the warehouse and managing the addition; and the intelligent auxiliary operation module is used for assisting maintenance operation of a maintainer. The invention realizes the fault treatment assistance of the operation and maintenance of the industrial equipment and the intelligent management of personnel, equipment and maintenance processes, and is beneficial to improving the operation and maintenance efficiency and the operation and maintenance quality.

Description

Industrial Internet remote monitoring operation and maintenance management system

Technical Field

The invention relates to the technical field of industrial Internet, in particular to an industrial Internet remote monitoring operation and maintenance management system.

Background

Smart manufacturing (INTELLIGENT MANUFACTURING, IM) is a human-machine integrated intelligent system consisting of intelligent machines and human experts that can conduct intelligent activities such as analysis, reasoning, judgment, conception and decision-making during manufacturing. Through the cooperation of the human and the intelligent machine, the brain work of human expert in the manufacturing process is enlarged, extended and partially replaced. It extends the concept of manufacturing automation to flexibility, intelligence and high integration.

Due to the high-speed development of advanced manufacturing technology, information technology and intelligent technology, the variety of industrial equipment is increased, the structure is increasingly complex, and the operation and maintenance of intelligent equipment become an important link of intelligent manufacturing. The traditional operation and maintenance mode is mainly based on the working experience of maintenance personnel to carry out fault judgment and give a solution, is not high enough in efficiency, mainly depends on the technical level and operation and maintenance experience of the operation and maintenance personnel, cannot be upgraded and shared, and cannot well provide decisions for managers. Therefore, the intelligent operation and maintenance level determines whether the production and operation efficiency of enterprises can be improved, and becomes an influence factor of industrial competitiveness.

Disclosure of Invention

The invention aims to provide an industrial Internet remote monitoring operation and maintenance management system which aims to solve the defects in the prior art.

The invention is implemented by the following technical scheme: the utility model provides an industry internet remote monitoring fortune dimension management system, includes fortune dimension knowledge base module, parameter acquisition module, patrol and examine management module, trouble match module, AI diagnostic module, fortune dimension dispatch module, equipment management module, intelligent auxiliary operation module, wherein:

The operation and maintenance knowledge base module is used for inputting and storing professional knowledge and experience knowledge of daily operation and maintenance and providing functions of knowledge topic management, accurate search, intelligent question and answer, intelligent recommendation and knowledge subscription;

The parameter acquisition module is used for acquiring state data, operation data and fault data of the equipment and preprocessing the data to provide a data base for fault matching and an AI diagnosis module;

the inspection management module is used for making an inspection plan and carrying out inspection work prompt, execution, recording, verification and operation and maintenance evaluation according to the inspection plan;

The fault matching module is used for performing fault matching according to the fault data acquired by the operation and maintenance knowledge base module and the parameter acquisition module;

the AI diagnostic module is used for providing a repairing scheme for the faults according to the fault matching result and combining the trained fault diagnosis model;

The operation and maintenance dispatch module is used for issuing operation and maintenance tasks and repairing schemes to corresponding maintainers according to fault matching results and carrying out operation and maintenance process management;

the equipment management module is used for automatically matching the operation and maintenance articles according to the operation and maintenance tasks and the repairing scheme, and registering articles in and out of the warehouse and managing the addition;

and the intelligent auxiliary operation module is used for assisting maintenance operations of maintainers by adopting intelligent industrial auxiliary equipment which is not limited by a mechanical arm, a 5G robot, a conveyor belt and a stacker crane.

Further, the operation and maintenance knowledge base module comprises a knowledge management unit, a topic classification unit, a precise search unit, an intelligent question-answering unit, an intelligent recommendation unit and a knowledge subscription unit, wherein:

The knowledge management unit is used for inputting and storing professional knowledge and experience knowledge of daily operation and maintenance, including professional technical knowledge, equipment parameter knowledge, equipment function knowledge, equipment operation knowledge, equipment maintenance knowledge, equipment inspection knowledge, historical fault record, historical maintenance record and fault mapping knowledge;

The topic classification unit is used for classifying the knowledge topic of the knowledge stored in the knowledge management unit and establishing classification numbers, wherein the classification numbers are subject, chapter and knowledge point in sequence according to the sequence from big to small;

the accurate searching unit is used for searching keywords according to the knowledge topics and the classification numbers thereof;

the intelligent question-answering unit is used for providing voice and image question-answering information input for the intelligent equipment terminal;

the intelligent recommendation unit is used for matching the classification number of the knowledge topic according to the keyword search and the voice and image question-answer information of the accurate search unit;

and the knowledge subscription unit is used for subscribing related knowledge according to the new requirement of daily operation and maintenance.

Further, the parameter acquisition module comprises a data acquisition unit, a data transmission unit and a data preprocessing unit, wherein:

The data acquisition unit acquires state data, operation data and fault data of the equipment;

the data preprocessing unit performs data cleaning, de-duplication, feature extraction and labeling;

The data transmission unit is used for adopting wireless transmission or wired transmission to the preprocessed data and providing a data base for fault matching and AI diagnosis modes.

Further, the inspection management module comprises a planning unit, an inspection prompting unit, a work execution unit, an inspection recording unit, an inspection checking unit, a maintenance evaluation unit, a maintenance estimation unit and a maintenance alarm unit, wherein:

a plan making unit for making contents including, but not limited to, inspection contents, inspection time, maintenance standards;

The inspection prompt unit is used for issuing inspection prompt information according to inspection content and inspection time;

the work execution unit is used for carrying out inspection and maintenance of the equipment module and the unit according to the inspection prompt information;

The inspection recording unit is used for registering and recording the inspection and maintenance positions, parameters and conditions;

the inspection verification unit is used for verifying the maintenance standard of the registration record and judging whether the work is completed according to the standard;

The maintenance evaluation unit is used for evaluating and scoring the operation and maintenance quality according to the check result of the registration record and the maintenance standard;

the maintenance estimation unit is used for carrying out operation state estimation on the equipment according to parameters and conditions of the inspection record under the condition that the operation quality is qualified;

and the maintenance alarm unit is used for reminding the manager client by sound and text messages according to the operation health degree of the judged equipment or reminding the mailbox or the short message according to the information configured by the user.

Further, the fault matching module comprises a classification extraction unit, a matching unit, a data screening unit, a matching degree calculation unit and a data sorting and positioning unit, wherein:

the classification extraction unit is used for carrying out fault classification according to the fault data acquired by the data acquisition unit and establishing a fault classification number, and the fault classification number corresponds to the knowledge topic classification number of the operation and maintenance knowledge base module;

The matching unit is used for comparing the fault classification number with the knowledge theme classification number to generate classification number similarity;

The data screening unit is used for setting a similarity threshold according to the similarity information of the classification numbers and screening out the knowledge topic classification numbers lower than the similarity threshold;

The matching degree calculation unit sequentially contracts and compares the classification numbers of the screened left knowledge topics according to the sequence of subjects, chapters and knowledge points to obtain branch matching degrees of each subject, chapter and knowledge point, and then carries out weighted calculation on the branch matching degrees to obtain the matching degrees, wherein the weights of the subjects, chapter and knowledge point are 10%, 20%, 30% and 40%;

And the data sequencing and positioning unit is used for sequencing the knowledge topic classification numbers corresponding to the fault classification numbers according to the matching degree, and taking the largest matching degree of the fault classification numbers and the knowledge topic classification numbers as the final matching degree and taking the largest matching degree as the basis of fault diagnosis.

Further, the AI diagnosis module includes a model training unit, a model verification unit, a fault diagnosis unit, wherein:

The model training unit is used for training the historical state data, the operation data, the fault data and the knowledge data of daily operation and maintenance which are preprocessed by the data preprocessing unit by using a machine learning and deep learning algorithm to construct a fault diagnosis model;

The model verification unit is used for evaluating the accuracy and the robustness of the trained model on an unknown fault sample and selecting an optimal model for deployment;

And the fault diagnosis unit is used for giving a fault repairing scheme through inputting the fault locating result located by the data ordering and locating unit.

Further, the operation and maintenance dispatch module comprises a right management unit, a maintainer matching unit, a fault display unit, an operation and maintenance scene recording unit, a progress management unit and a result feedback unit, wherein:

The authority management unit is used for setting and managing the dispatch authority, and the dispatch authority is matched with different administrators according to the equipment fault level;

The maintainer matching unit is used for automatically issuing the operation and maintenance tasks and the repairing schemes provided by the AI diagnostic modules to corresponding maintainers according to different types of faults;

The fault display unit is used for displaying faults in forms of tables, characters, pictures and audios and videos;

the operation and maintenance scene recording unit is used for recording the operation and maintenance process in forms of tables, characters, pictures and audios and videos;

The progress management unit is used for managing and displaying operation and maintenance progress, and the progress comprises states of preparation, inspection, debugging, maintenance and finishing;

And the result feedback unit is used for feeding back the operation and maintenance result to the client.

Further, the device management module comprises an operation and maintenance article management unit, an article automatic scheduling unit, an in-out warehouse management unit and an adding management unit, wherein:

The operation and maintenance article management unit is used for managing operation and maintenance article information which is not limited to instruments, tools and spare parts and is required by operation and maintenance, wherein the operation and maintenance article information comprises positions, models and parameters;

The automatic article scheduling unit is used for automatically matching the position, model and parameters of the corresponding operation and maintenance articles according to the operation and maintenance tasks and the repairing scheme, and automatically popping up the articles through the article storage equipment;

the warehouse-in and warehouse-out management unit is used for recording information of use and return of the operation and maintenance articles and prompting articles in the return period;

And the adding management unit is used for adding prompts for the instruments, tools and spare parts which are missing or exceed the service life.

Further, the intelligent auxiliary operation module comprises intelligent industrial auxiliary equipment which is matched with maintainers on the operation and maintenance site and is not limited to a manipulator, a 5G robot, a conveyor belt and a stacker crane.

The invention has the advantages that:

1. According to the invention, through constructing an operation and maintenance knowledge base, inputting and storing professional knowledge and experience knowledge of daily operation and maintenance, the functions of knowledge topic management, accurate search, intelligent question-answering, intelligent recommendation and knowledge subscription are provided, knowledge data reference is provided for industrial equipment operation and maintenance, knowledge skill learning of operation and maintenance personnel is assisted through a big data technology, and equipment faults are efficiently handled through an artificial intelligent technology.

2. According to the invention, planning, inspection prompting, inspection recording, inspection quality verification and evaluation of equipment operation and maintenance are realized through the inspection management module, and high-quality closed-loop management is realized; meanwhile, maintenance estimation and maintenance alarm are carried out based on the data of inspection, so that fault pre-maintenance in the operation and maintenance process is realized, and the fault pre-maintenance is prevented.

3. According to the invention, fault matching is carried out through the fault matching module according to the fault data acquired by the operation and maintenance knowledge base module and the parameter acquisition module, then a repairing scheme is provided for the fault through the AI diagnosis module according to the fault matching result and combining with the trained fault diagnosis model, so that reference is further provided for operation and maintenance personnel in fault processing, and the rapid processing of equipment faults is facilitated.

4. The invention manages the authority allocation and reasonable matching of the operation and maintenance personnel through the operation and maintenance dispatch module, thereby realizing the orderly management of the operation and maintenance personnel and avoiding the problem of confusion; in the maintenance process, fault display, operation and maintenance scene recording, progress management and result feedback are carried out, so that the transparency of maintenance and the traceability of the process are realized.

5. According to the invention, the equipment management module is used for carrying out operation and maintenance article management, automatic scheduling, warehouse-in and warehouse-out management and addition management based on the RFID technology, so that data guarantee is provided for operation and maintenance article management, automatic warehouse-in and warehouse-out management of articles in an operation and maintenance process is realized, and material resource guarantee is provided for efficient operation and maintenance.

6. According to the invention, by arranging the intelligent auxiliary operation module, intelligent industrial auxiliary equipment which is not limited to the mechanical arm, the 5G robot, the conveyor belt and the stacker crane is adopted to assist maintenance operation of maintainers, and when the maintainers perform maintenance work, the intelligent equipment assists in carrying, dismounting, detecting and other works, so that external force is provided for maintenance work.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of an industrial Internet remote monitoring operation management system according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of an operation and maintenance knowledge base module of an industrial Internet remote monitoring operation and maintenance management system according to an embodiment of the invention;

FIG. 3 is a schematic diagram of a parameter acquisition module of an industrial Internet remote monitoring operation management system according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a patrol management module of an industrial Internet remote monitoring operation and maintenance management system according to an embodiment of the invention;

FIG. 5 is a schematic diagram of a fault matching module of an industrial Internet remote monitoring operation management system according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of an AI diagnostic module of an industrial Internet remote monitoring operation and maintenance management system according to an embodiment of the invention;

FIG. 7 is a schematic diagram of an operation and maintenance dispatch module of an industrial Internet remote monitoring operation and maintenance management system according to an embodiment of the present invention;

FIG. 8 is a schematic diagram of a device management module of an industrial Internet remote monitoring operation and maintenance management system according to an embodiment of the present invention;

fig. 9 is a hardware topology diagram of an industrial internet remote monitoring operation management system according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

As shown in fig. 1, an industrial internet remote monitoring operation and maintenance management system includes a patrol management module 101, an operation and maintenance knowledge base module 102, a fault matching module 103, an AI diagnosis module 104, an operation and maintenance dispatch module 105, a device management module 106, a parameter acquisition module 107, and an intelligent auxiliary operation module 108, wherein:

The operation and maintenance knowledge base module 102 is used for inputting and storing professional knowledge and experience knowledge of daily operation and maintenance, wherein the professional knowledge is mainly theoretical knowledge of subjects related to equipment, and the experience knowledge is history maintenance practice experience knowledge. Meanwhile, the functions of knowledge topic management, accurate search, intelligent question and answer, intelligent recommendation and knowledge subscription are provided; knowledge topic management includes knowledge fields and classifications, for example, the corresponding module of a device relates to one or more knowledge fields of machinery, electrician, electric, communication, circuits, processors, power supplies, materials, optics, semiconductors, computer networks, computer software and the like, and classifications are specific branch knowledge of the above fields, for example, a circuit relates to a diode circuit, a triode circuit, an operational amplifier circuit, a resistor-capacitor circuit, a power amplification circuit, a matching circuit and the like. The precise search is based on computer search algorithms, such as crawler algorithm, enumeration algorithm, depth-first search, breadth-first search, a-th algorithm, backtracking algorithm, monte carlo tree search, hash function, etc., and is used for automatically identifying when the knowledge of the relevant classification needs to be searched. The intelligent question and answer is that the user puts forward the questions in the forms of words, voice and the like, the knowledge base carries out question searching and matching based on the knowledge spectrum to obtain answers to be presented, the knowledge spectrum question and answer based method and the answer sorting based method are adopted in the embodiment, the knowledge spectrum question and answer based method is that firstly, the input questions are converted into structured queries (namely semantic representation) which can be understood and executed by the knowledge spectrum, and then the structured queries are directly executed on the knowledge spectrum to obtain answers corresponding to the questions. The answer sorting method is that a small number of answer candidates are quickly found from a knowledge graph based on an input question, then different answer candidates are scored by adopting a sorting model, and an answer candidate set with the highest score is selected as an answer corresponding to the question. The intelligent recommendation comprises five links of data collection, feature extraction, feature calculation, result ordering and front-end call, wherein the feature extraction (user portrait and commodity portrait), recall calculation and result ordering are core parts of a recommendation system, and comprise: a: and generating a device feature vector and a knowledge theme feature matrix. B: and converting an initial recommended knowledge topic list according to the equipment feature vector and the feature object correlation matrix. C: and filtering and ranking the initial recommendation list to generate a final recommendation result. Knowledge subscription, namely knowledge content in related or similar fields is actively recommended through a knowledge base, and whether subscription is carried out or not is automatically selected by a user.

A parameter obtaining module 107, configured to collect status data, operation data, and fault data of the device and perform preprocessing on the data, so as to provide a data basis for fault matching and AI diagnosis; the state data is parameters such as normal operation, alarm, protection, fault and the like of the equipment, the operation data is parameters such as temperature, speed, acceleration, vibration, air pressure, frequency, voltage, current, power and the like of the equipment when the equipment is in operation, and the fault data is a specific parameter which leads to the equipment fault, such as power failure, mechanical fault, circuit fault, executive component fault and the like. The faults are classified in the knowledge base corresponding to the corresponding knowledge, and serve as the basis for fault knowledge matching and diagnosis.

The inspection management module 101 is used for making an inspection plan and carrying out inspection work prompt, execution, recording, verification and operation and maintenance evaluation according to the inspection plan; the inspection plan is actually carried out by combining field devices, the inspection can be carried out on the project according to the project requirement and the frequency of time, day, week, month and season, and the prompt function can be set in the inspection deadline period to avoid omission. In the inspection process, recording about inspection personnel, inspection time, inspection items, inspection results, treatment modes and the like, and setting a verification link of the inspection process to monitor, and finally providing operation and maintenance evaluation functions based on the verification results.

The fault matching module 103 is used for performing fault matching according to the fault data acquired by the operation and maintenance knowledge base module and the parameter acquisition module; according to the above, the operation and maintenance knowledge base module has established knowledge topics, specifically to various branches. And the equipment state data, the operation data and the fault data acquired by the parameter acquisition module also correspond to corresponding knowledge topics. The purpose of the fault matching module 103 is to match the knowledge topics corresponding to the fault data into an operation and maintenance knowledge base, and provide a fault repairing scheme for the follow-up operation and maintenance knowledge base. The fault matching module 103 specifically adopts a keyword matching algorithm (for example, keyword matching is performed by a TextRank algorithm), which is specifically:

a. Word segmentation

Fault data: the current is overlarge, and the fuse is protected;

fault knowledge topic: current, fusing and protecting;

Operation and maintenance knowledge base theme 1: the current, fusing and protection, namely the overlarge current between the power supply and the load, and the controller detects the current fault to perform equipment protection;

Operation and maintenance knowledge base theme 2: current, fusing and protecting, namely, overlarge current of the power amplifier, and fusing the fuse to protect the module;

Operation and maintenance knowledge base theme 3: current, fusing and protection, namely unstable power supply voltage, and fusing and protection of a power fuse;

Operation and maintenance knowledge base theme 4: current, fusing and protection, and under-voltage protection of equipment.

B. De-duplication intersection

According to the fault knowledge theme: the current, fusing and protecting operation and maintenance knowledge base theme 1 has the intersection of the current and protecting keywords, the current, fusing and protecting operation and maintenance knowledge base theme 2 has the three keywords, the fusing and protecting operation and maintenance knowledge base theme 3 has the keywords, and the operation and maintenance knowledge base theme 4 has only one keyword.

C. Similarity degree

According to the matching result, the operation and maintenance knowledge base theme 2 has three keywords matched, and if the matching degree is highest, the knowledge branch under the operation and maintenance knowledge base theme 2 is selected to provide a fault giving repair scheme.

The AI diagnosis module 104 is configured to provide a repair scheme for the fault according to the fault matching result and in combination with the trained fault diagnosis model; according to the above example, when the failure data occurs: the current is overlarge, the fuse is fused and protected, the power amplifier of the operation and maintenance knowledge base theme 2 is adopted, the fuse is fused and protected, and details about the operation and maintenance knowledge base theme 2 in the knowledge base are about the details of how to replace the fuse, how to detect whether a module is normal or not after replacement, how to debug and recover after replacement, a standby scheme of equipment faults still exists after debugging and recovery, and the like.

The operation and maintenance dispatch module 105 is used for issuing operation and maintenance tasks and repairing schemes to corresponding maintainers according to fault matching results and performing operation and maintenance process management; the fault matching result corresponds to the corresponding position and the knowledge field of the equipment, and according to the fault position and the knowledge field, the fault matching result can be associated to equipment maintenance personnel responsible for the corresponding position. The system establishes the equipment number, the maintainer number, the equipment module number and the maintenance field number in advance, can automatically associate the maintainer according to the fault result, forms the operation and maintenance task corresponding to the fault problem, simultaneously takes the repairing scheme as a reference for the maintainer, improves the maintenance efficiency, and carries out progress management in the maintenance process.

The device management module 106 is configured to automatically match the operation and maintenance items according to the operation and maintenance task and the repair scheme, and perform the registration of the items in and out of the warehouse and the addition management; the device management module 106 establishes operation and maintenance items including instruments, tools, elements, etc. required for corresponding fault maintenance according to the operation and maintenance knowledge base, such as a multimeter, an oscilloscope, a test pencil, a wire stripper, etc. required for circuit maintenance, a wrench, a screwdriver, a pliers, a hammer, etc. required for mechanical maintenance, and performs access management on the taken maintenance items during maintenance. The device management module 106 uses an intelligent cabinet to place a device or a tool, an RFID tag is attached to the device or the tool, a unique ID number is provided, when the object is taken out from the corresponding area in a classified manner, the object taking state can be sensed through an RFID reader, and the taking data are used as a database basis for object access management. Meanwhile, the intelligent cabinet can be provided with a camera for face recognition, acquires the information of the maintenance personnel, and is traceable to the management of the personnel.

The intelligent auxiliary operation module 108 is used for assisting maintenance operations of maintenance staff by adopting intelligent industrial auxiliary equipment not limited to manipulators, 5G robots, conveyor belts and stacker crane. When maintainers carry out maintenance work, the intelligent equipment assists in carrying, dismounting, detecting and other works.

As shown in fig. 2, the operation and maintenance knowledge base module 102 includes a knowledge management unit 1021, a topic classification unit 1022, a precision search unit 1023, an intelligent question-answering unit 1026, an intelligent recommendation unit 1024, and a knowledge subscription unit 1025, wherein:

The knowledge management unit 1021 is configured to record and store expertise and experience knowledge of daily operation and maintenance, including expertise, device parameter knowledge, device function knowledge, device operation knowledge, device maintenance knowledge, device inspection knowledge, historical fault record, historical maintenance record, and fault mapping knowledge; some of the knowledge comes from manufacturing merchants, and some of the knowledge is registered in daily maintenance, but in summary, the knowledge correspondingly penetrates through the index of the corresponding knowledge when a database is built, so that basis is provided for subsequent retrieval.

The topic classification unit 1022 is configured to classify the knowledge stored in the knowledge management unit and establish classification numbers, where the classification numbers sequentially include subjects, chapters, and knowledge points in order from large to small, for example, if a device is a temperature control device, the subjects include electricity, and the lower layer of electricity includes sensors, analog circuits, and subjects of a microprocessor. The sensor subject includes a section of a temperature sensor, and the analog circuit subject lower layer may include a section of a power supply circuit, a temperature detection circuit, a power amplification circuit, or a temperature control circuit, and the microprocessor lower layer includes a section of a single chip microcomputer, a PLC, a DSP, or an FPGA. Knowledge points are the principle, fault analysis, component function knowledge, etc. of the circuits involved in the respective units. Of course, more than one subject is involved in the device, such as temperature control devices typically involve mechanical subject, even chemical subject, biological subject, and the corresponding subject knowledge classification references the above methods.

The accurate searching unit 1023 is used for searching keywords according to the knowledge topics and the classification numbers thereof; the subject classification numbers of the electric naming are D, the subjects of the sensor, the analog circuit and the microprocessor at the lower layer of the electric naming are respectively named D1, D2 and D3, the chapters of the temperature sensor of the sensor subject are named D11, the chapters of the power supply circuit, the temperature detection circuit, the power amplifying circuit or the temperature control circuit at the lower layer of the analog circuit are respectively named D21, D22, D23 and D24, the chapters of the singlechip, the PLC, the DSP or the FPGA at the lower layer of the microprocessor are respectively named D31, D32, D33 and D34, when a certain module of the equipment fails, keyword searching is carried out according to a circuit unit contained in the equipment, and corresponding knowledge details are automatically popped up through inputting the classification numbers.

The intelligent question-answering unit 1026 is used for providing voice and image question-answering information input for the intelligent equipment terminal; the system designs voice recognition and image recognition functions, and automatically matches knowledge required to be obtained through voice and image recognition.

The intelligent recommendation unit 1024 is configured to match the classification number of the knowledge topic according to the keyword search and the voice and image question-answer information of the accurate search unit; this is achieved primarily by existing mainstream machine learning algorithms.

The knowledge subscription unit 1025 is configured to subscribe to related knowledge according to the daily operation, for example, some devices may be related to some conventional knowledge, but may be selected to subscribe due to the added knowledge of other fields related to the new function.

As shown in fig. 3, the parameter acquisition module 107 includes a data acquisition unit 1071, a data preprocessing unit 1072, and a data transmission unit 1073, wherein:

The data acquisition unit 1071 acquires status data, operation data, and fault data of the device; the state data is parameters such as normal operation, alarm, protection, fault and the like of the equipment, the operation data is parameters such as temperature, speed, acceleration, vibration, air pressure, frequency, voltage, current, power and the like of the equipment when the equipment is in operation, and the fault data is a specific parameter which leads to the equipment fault, such as power failure, mechanical fault, circuit fault, executive component fault and the like.

The data preprocessing unit 1072 performs data cleaning, deduplication, feature extraction and labeling to ensure the quality of the data for subsequent analysis, and these preprocessing steps can ensure the consistency and reliability of the data, specifically:

1. The data cleaning comprises the following steps:

(1) Missing value processing:

and (3) filling the mean value: the missing values are replaced with the mean of the data features.

And (5) filling the median: the missing values are replaced with the median of the data features.

Mode filling: the missing values are replaced by the mode of the data feature (the value with the highest frequency of occurrence).

(2) Outlier processing:

the Z-Score method was used: for detecting and processing outliers, the degree of deviation of the data point from the mean is calculated and if the threshold is exceeded, it is considered an outlier.

(3) Data type conversion:

data is converted from one type to another, for example, date data of a character string type is converted to a date-time type.

2. Data deduplication:

(1) De-duplication based on hash values:

the data records are mapped to unique hash values using a hash function, and then the hash values are compared to identify and delete duplicate records.

(2) Data tag allocation: and (3) assigning labels for the supervised learning tasks, and associating the data samples with the corresponding target values.

The feature extraction specifically comprises the following steps:

(1) Word bag model: the data is represented as vectors of words in a vocabulary, with the frequency of each word as a feature value.

(2) TF-IDF: the importance of words in the text is measured, the weight is reduced for common words, and the weight is increased for rare words.

(3) Word embedding: words are mapped to points in the continuous vector space using a pre-trained Word vector model, such as Word2Vec, gloVe, etc.

(4) N-gram model: relationships between words, such as bigrams (2-grams) and trigrams (3-grams), are considered to obtain more context information.

(5) Theme modeling: methods such as LATENT DIRICHLET Allocation (LDA) are used to identify topics in the text, which are characterized.

4. Part of speech tagging: the part of speech of each word in the text is tagged to help understand the role of the word in the sentence.

The data transmission unit 1073 is configured to provide a data basis for fault matching and AI diagnosis modes by adopting wireless transmission or wired transmission for the preprocessed data. The wireless transmission can adopt modes such as 4G/5G, wifi communication, bluetooth communication and the like, and the wired transmission adopts Ethernet transmission, optical fiber transmission and the like.

As shown in fig. 4, the patrol management module 101 includes a plan making unit 1011, a patrol presenting unit 1012, a work executing unit 1013, a patrol recording unit 1014, a patrol checking unit 1015, a maintenance evaluating unit 1016, a maintenance estimating unit 1017, and a maintenance warning unit 1018, wherein:

A plan making unit 1011 for making contents including, but not limited to, inspection contents, inspection time, maintenance standards; the inspection content divides the equipment into a plurality of modules and a plurality of units, the inspection time is correspondingly distributed to a plurality of time points (such as weekly inspection, monthly inspection, quaternary inspection and annual inspection) of one year, and the maintenance standard is which parts of the equipment modules/units need to be inspected, which parts need to achieve what maintenance effect, what work indexes are realized, and the like.

The inspection prompt unit 1012 is configured to issue inspection prompt information according to inspection content and inspection time; the system is set to remind maintenance personnel by sending prompt short messages or WeChat prompt messages before the corresponding time of the inspection items arrives for 3-5 days.

The work execution unit 1013 is configured to perform inspection and maintenance of the equipment module and the unit according to the inspection prompt information; the maintenance personnel conduct inspection operation according to the inspection prompt, and the maintenance personnel register in the module before inspection.

A patrol recording unit 1014 for registering and recording patrol and maintenance positions, parameters and conditions; such recorded information may be recorded automatically by sound, image or other sensor data.

A inspection checking unit 1015 for checking according to the registration record and the maintenance standard, and judging whether the work is completed according to the standard; the verification can adopt various modes such as equipment working data comparison, image recognition, manual verification and the like.

And a maintenance evaluation unit 1016 for performing evaluation scoring on the operation and maintenance quality according to the registration record and the verification result of the maintenance standard. The scoring is performed by a machine learning algorithm according to the verification result, and manual scoring is adopted for some special standards.

The maintenance estimation unit 1017 is configured to perform operation state estimation on the device according to parameters and conditions recorded by inspection under the condition that the operation quality is qualified, and mainly combine collected state data, such as parameters of normal operation, alarm, protection, fault, and the like of the device, and operation data, such as operation parameters of temperature, speed, acceleration, vibration, air pressure, frequency, voltage, current, power, and the like of the device during operation, to compare with a standard value of the device, so as to obtain a parameter deviation degree, where for example, a normal operation current of a certain device is in a range I: a is less than or equal to I and less than or equal to B, the acquired real-time current range is a is less than or equal to I and less than or equal to B, and the parameter deviation degree is: the deviation degree/>, of other parameters such as speed, acceleration, vibration, air pressure, frequency, voltage and the like, can be obtained by the same method 、/>……/>Then, setting weight values of all the parameters according to the importance degree of the parameters, and then carrying out weighted summation to obtain the total deviation/>Setting a deviation threshold value/> according to daily maintenance experienceAccording to the total deviation/>And degree of deviation threshold/>The operating health of the equipment is classified as good and bad.

And the maintenance alarm unit 1018 is used for carrying out sound and text message reminding on the manager client according to the operation health degree of the judged equipment or carrying out mailbox reminding or short message reminding according to the information configured by the user.

As shown in fig. 5, the fault matching module 103 includes a classification extraction unit 1031, a matching unit 1032, a data screening unit 1033, a matching degree calculation unit 1034, and a data sorting positioning unit 1035, wherein:

The classification extraction unit 1031 is configured to perform fault classification according to the fault data collected by the data collection unit and establish a fault classification number, where the fault classification number corresponds to the knowledge topic classification number of the operation and maintenance knowledge base module; the establishment of the fault classification corresponds to the establishment of the topic classification data, for example, the subject classification of the electrical fault classification names is d, the subject fault classifications of the sensor, the analog circuit and the microprocessor at the lower layer of the electrical system are respectively named d1, d2 and d3, the chapter fault classification of the temperature sensor at the subject of the sensor is named d11, the chapter fault classifications of the power circuit, the temperature detection circuit, the power amplification circuit or the temperature control circuit at the lower layer of the analog circuit are respectively named d21, d22, d23 and d24, and the chapter fault classifications of the singlechip, the PLC, the DSP or the FPGA at the lower layer of the microprocessor are respectively named d31, d32, d33 and d34. It can be seen that the corresponding knowledge topic classification corresponds to the fault classification labels "D" and "D", facilitating the search algorithm design.

The matching unit 1032 is used for comparing the fault classification number with the knowledge topic classification number to generate classification number similarity; since a certain failure phenomenon may involve problems with several links, some of these may be major factors and some of these may be minor factors. What is more, is a frequent factor and what is more, is a sporadic factor, so that when a fault matches a topic classification, there is not a single match, but rather a hierarchical "one-to-many" mapping match.

The data screening unit 1033 sets a similarity threshold according to the similarity information of the classification numbers, and screens out knowledge topic classification numbers lower than the similarity threshold; and setting a similarity threshold according to main and secondary factors and frequent and sporadic factors of the fault and topic classification matching, and reserving matching items meeting the threshold.

Matching degree calculating unit 1034, sequentially shrinking and comparing the classified numbers of the screened left knowledge topics according to the sequence of disciplines, subjects, chapters and knowledge points to obtain branch matching degrees of each discipline, subject, chapter and knowledge point, and performing weighted calculation on the branch matching degrees to obtain matching degrees, wherein the weights of the disciplines, subject, chapter and knowledge point are 10%, 20%, 30% and 40%; for example, the temperature control apparatus cannot normally perform temperature adjustment, may be a controller failure, may be a power supply circuit failure, and may be a driving unit circuit failure in which temperature adjustment is performed, and thus weights are set according to primary and secondary factors and frequent and occasional factors, respectively.

The data sorting and positioning unit 1035 sorts the knowledge topic classification numbers corresponding to the fault classification numbers according to the matching degree, and uses the largest fault classification number and the matching degree of the knowledge topic classification numbers as the final matching degree and as the basis of fault diagnosis.

As shown in fig. 6, the AI diagnosis module 104 includes a model training unit 1041, a model verification unit 1042, and a fault diagnosis unit 1043, in which:

The model training unit 1041 is configured to train the historical state data, the operation data, the fault data and the knowledge data of the daily operation and maintenance after the data preprocessing unit is preprocessed by using a machine learning and deep learning algorithm (such as a convolutional neural network CNN or a cyclic neural network RNN), so as to construct a fault diagnosis model; the fault diagnosis model is from experience knowledge of daily data management and operation and maintenance, and is combined with machine learning and deep learning algorithm to further perform data training, so that a normal fault diagnosis strategy is formed, and when faults occur, an accurate fault treatment scheme is generated with high probability through the support of the model.

The model verification unit 1042 is used for evaluating the accuracy and robustness of the trained model on the unknown fault sample, and selecting an optimal model for deployment; besides the fault diagnosis model is used for analyzing normal faults, a plurality of relatively unusual faults are set, and the model is further optimized and trained, so that the model diagnosis is more coverage.

The fault diagnosis unit 1043 orders the fault location results located by the location unit 1035 according to the input data, and gives a fault repair scheme by combining the trained fault diagnosis model with the operation and maintenance knowledge base module 102.

As shown in fig. 7, the operation and maintenance dispatch module 105 includes a rights management unit 1051, a maintainer matching unit 1052, a fault display unit 1053, an operation and maintenance scene recording unit 1054, a progress management unit 1055, and a result feedback unit 1056, wherein:

The authority management unit 1051 is used for setting and managing dispatch authorities, and the dispatch authorities are matched with different administrators according to the equipment fault level; in order to ensure the reliability and safety of operation and maintenance, the invention also provides dispatch authority management of operation and maintenance, and for faults involving key links and key parts, operation and maintenance personnel with rich experience and strong treatment capability are needed because faults possibly involve reliability and timeliness of fault treatment, common operation and maintenance personnel can be qualified when general faults occur, simple faults can be used as training of new people, and the authorities manage and search dogs according to registration by administrators of different levels.

The maintainer matching unit 1052 is configured to automatically issue the operation and maintenance tasks and the repair schemes provided by the AI diagnostic modules to corresponding maintainers according to different types of faults; when the manager corresponding to the fault is determined, and the fault occurs in the authority of the manager, the corresponding manager dispatches the list to the corresponding maintainer managed by the manager.

A fault display unit 1053, configured to display faults in forms of tables, characters, pictures, and audio/video; according to the fault form, the fault data are sent to the handheld terminal of the maintainer in different modes so as to comprehensively and timely know the fault phenomenon.

An operation and maintenance scene recording unit 1054, configured to record an operation and maintenance process in the form of tables, characters, pictures, and audio and video; corresponding to the fault display, the faults are recorded at the same time, so that the follow-up tracing is convenient.

A progress management unit 1055 for managing and displaying operation and maintenance progress including states of preparation, inspection, debugging, maintenance, and completion; according to the maintenance conditions of maintenance personnel, classifying and managing the conditions of each stage of progress;

and the result feedback unit 1056 is configured to feed back the operation and maintenance result to the client, and feed back the status of each stage of the progress to the administrator client in real time.

As shown in fig. 8, the device management module 106 includes an operation and maintenance article management unit 1061, an automatic article scheduling unit 1062, an in-out management unit 1063, and an add-on management unit 1064, where:

the operation and maintenance article management unit 1061 is used for managing operation and maintenance article information which is not limited to instruments, tools and spare parts and is required for operation and maintenance, wherein the operation and maintenance article information comprises positions, models and parameters, the operation and maintenance article information is stored on an RFID label, and the operation and maintenance article information is automatically identified and managed through an RFID reader;

The automatic article scheduling unit 1062 is configured to automatically match the position, model, and parameter of the corresponding operation and maintenance article according to the operation and maintenance task and the repair scheme, and automatically pop up through the article storage device; when an administrator sends an operation and maintenance task, the operation and maintenance task not only comprises a fault situation and a maintenance personnel situation, but also comprises operation and maintenance article information corresponding to the related maintenance requirement, the operation and maintenance article information is automatically matched and then sent to an article management intelligent cabinet controller, and the controller issues an instruction to open corresponding article corresponding sections of the intelligent cabinet, so that the articles are automatically ejected.

The warehouse-in and warehouse-out management unit 1063 is used for recording information of the use and the return of the operation and maintenance articles and prompting the articles in the return period; after the articles are automatically popped up, the RFID reader notifies the record of the taking information, meanwhile, the article returning time limit is set according to the time limit of operation and maintenance personnel, and the returning information is sent to the maintainer client in one end before the time limit, so that article use management is realized, and omission is avoided.

The adding management unit 1064 is configured to add and prompt an instrument, a tool, and a spare part that are missing or exceed the service life. The intelligent cabinet also records the delivery information of the articles, and reminds replacement when the service life expires.

Fig. 9 shows a system hardware topology diagram of the present invention, in which a system server is a center of the system, and mainly implements processing of collected data, operation and maintenance knowledge base construction, fault matching, AI diagnosis, and operation and maintenance dispatch, and the operation and maintenance dispatch communicates with peripheral devices through a 4G/5G network wireless communication or a wired network communication. The maintainer terminal and the administrator terminal receive operation and maintenance instructions issued by the system server to realize corresponding operation and maintenance work, the intelligent cabinet realizes intelligent management of operation and maintenance objects, the equipment sensor is used for collecting equipment operation data and fault data, the intelligent auxiliary equipment realizes operation and maintenance auxiliary work, the data large screen and the display end realize equipment fault display and display of operation and maintenance processes, and the alarm terminal is used for equipment pre-maintenance alarm.

In summary, the invention achieves the following technical effects:

The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.

Claims

1. The industrial Internet remote monitoring operation and maintenance management system is characterized by comprising an operation and maintenance knowledge base module, a parameter acquisition module, a patrol management module, a fault matching module, an AI diagnosis module, an operation and maintenance dispatch module, an equipment management module and an intelligent auxiliary operation module, wherein:

The operation and maintenance knowledge base module is used for inputting and storing professional knowledge and experience knowledge of daily operation and maintenance and providing functions of knowledge topic management, accurate search, intelligent question and answer, intelligent recommendation and knowledge subscription; the operation and maintenance knowledge base module comprises a knowledge management unit, a topic classification unit, an accurate search unit, an intelligent question-answering unit, an intelligent recommendation unit and a knowledge subscription unit, wherein:

the knowledge subscription unit is used for subscribing related knowledge according to the new need of daily operation and maintenance;

the inspection management module is used for making an inspection plan and carrying out inspection work prompt, execution, recording, verification and operation and maintenance evaluation according to the inspection plan; the inspection management module comprises a planning unit, an inspection prompt unit, a work execution unit, an inspection recording unit, an inspection verification unit, a maintenance evaluation unit, a maintenance estimation unit and a maintenance alarm unit, wherein:

the inspection checking unit is used for checking the maintenance standard of the registration record and judging whether the work is completed according to the standard

the maintenance alarm unit is used for reminding the manager client of sound and text messages according to the operation health degree of the judged equipment or reminding the mailbox or the short message according to the information configured by the user;

the fault matching module is used for performing fault matching according to the fault data acquired by the operation and maintenance knowledge base module and the parameter acquisition module; the fault matching module comprises a classification extraction unit, a matching unit, a data screening unit, a matching degree calculation unit and a data ordering and positioning unit, wherein:

The data sorting and positioning unit sorts the knowledge topic classification numbers corresponding to the fault classification numbers according to the matching degree, and takes the largest matching degree of the fault classification numbers and the knowledge topic classification numbers as the final matching degree and the basis of fault diagnosis;

The operation and maintenance dispatch module is used for issuing operation and maintenance tasks and repairing schemes to corresponding maintainers according to fault matching results and carrying out operation and maintenance process management; the operation and maintenance dispatch module comprises a permission management unit, a maintainer matching unit, a fault display unit, an operation and maintenance scene recording unit, a progress management unit and a result feedback unit, wherein:

the result feedback unit is used for feeding back the operation and maintenance result to the client;

The equipment management module is used for automatically matching the operation and maintenance articles according to the operation and maintenance tasks and the repairing scheme, and registering articles in and out of the warehouse and managing the addition; the equipment management module comprises an operation and maintenance article management unit, an article automatic scheduling unit, an in-out warehouse management unit and an adding management unit, wherein:

The adding management unit is used for adding prompts for instruments, tools and spare parts which are missing or exceed the service life;

2. The industrial internet remote monitoring operation and maintenance management system according to claim 1, wherein the parameter acquisition module comprises a data acquisition unit, a data transmission unit and a data preprocessing unit, wherein:

3. The industrial internet remote monitoring operation and maintenance management system of claim 1, wherein the AI diagnosis module comprises a model training unit, a model verification unit, a fault diagnosis unit, wherein:

4. The industrial internet remote monitoring operation and maintenance management system according to claim 1, wherein the intelligent auxiliary operation module comprises intelligent industrial auxiliary equipment which is matched with maintainers on an operation and maintenance site and is not limited to a manipulator, a 5G robot, a conveyor belt and a stacker.