CN114548176A - Sample library updating method and device, electronic equipment and storage medium - Google Patents

Sample library updating method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN114548176A
CN114548176A CN202210171055.3A CN202210171055A CN114548176A CN 114548176 A CN114548176 A CN 114548176A CN 202210171055 A CN202210171055 A CN 202210171055A CN 114548176 A CN114548176 A CN 114548176A
Authority
CN
China
Prior art keywords
algorithm
fault
order
real
sample library
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.)
Pending
Application number
CN202210171055.3A
Other languages
Chinese (zh)
Inventor
付磊
汪文峰
吴思远
罗建华
袁爱进
闫鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Huaxing Digital Technology Co Ltd
Original Assignee
Shanghai Huaxing Digital Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Huaxing Digital Technology Co Ltd filed Critical Shanghai Huaxing Digital Technology Co Ltd
Priority to CN202210171055.3A priority Critical patent/CN114548176A/en
Publication of CN114548176A publication Critical patent/CN114548176A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/23—Updating
    • G06F16/2379—Updates performed during online database operations; commit processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2474—Sequence data queries, e.g. querying versioned data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/3332—Query translation
    • G06F16/3334—Selection or weighting of terms from queries, including natural language queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/80—Management or planning

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a sample library updating method, a sample library updating device, electronic equipment and a storage medium, and relates to the technical field of fault diagnosis, wherein the method comprises the following steps: acquiring a target associated case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; judging the conversion relation between the algorithm pushing order and the real maintenance order based on a preset matching condition, and determining whether the real maintenance order is the conversion result of the algorithm pushing order; and if so, taking the fault prediction result in the algorithm push order and the operation data of the target equipment as a group of fault cases, and updating the sample library. The method and the device provided by the invention improve the utilization rate of real operation data of the equipment, and automatically realize the updating of the fault diagnosis algorithm sample library, so that the effectiveness and the real-time performance of the sample library are maintained.

Description

Sample library updating method and device, electronic equipment and storage medium
Technical Field
The present invention relates to the field of fault diagnosis technologies, and in particular, to a method and an apparatus for updating a sample library, an electronic device, and a storage medium.
Background
The technology of equipment fault diagnosis has been advanced significantly through long-term development. The conventional diagnosis technology based on signal processing is developed to intelligent fault diagnosis with the technologies of industrial internet, big data, artificial intelligence and the like as the core. With the complexity of equipment and the improvement of the complexity of operating conditions, the traditional fault diagnosis technology can not meet the requirements increasingly, and the intelligent fault diagnosis technology integrates mass data generated by the industrial internet, an artificial intelligence technology and an equipment fault mechanism and becomes a fault diagnosis technology with great prospect.
The intelligent fault diagnosis technology can monitor the health state of the industrial equipment in real time, diagnose equipment faults in advance, ensure the stable operation of the equipment, and simultaneously save the maintenance cost of the equipment and reduce the occurrence of accidents.
The important basis of the intelligent fault diagnosis technology is a sample library, and the quality of the sample library determines the upper limit of the performance of the intelligent fault diagnosis algorithm. In the face of continuously generated mass data and continuously changing equipment states and complex working conditions, how to improve the effectiveness and the real-time performance of the sample library becomes a technical problem to be solved urgently in the industry.
Disclosure of Invention
The invention provides a method and a device for updating a sample library, electronic equipment and a storage medium, which are used for solving the technical problem of how to improve the effectiveness and the real-time performance of the sample library in the prior art.
The invention provides a sample library updating method, which comprises the following steps:
acquiring a target associated case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; the failure prediction result is determined based on the operation data of the target equipment and a sample library;
judging the conversion relation between the algorithm push order and the real maintenance order based on a preset matching condition, and determining whether the real maintenance order is the conversion result of the algorithm push order;
and if so, taking the fault prediction result in the algorithm push order and the operation data of the target equipment as a group of fault cases, and updating the sample library.
According to the sample library updating method provided by the invention, after the step of determining whether the real maintenance order is a conversion result of the algorithm push order, the method further comprises the following steps:
and if not, correcting the fault prediction result in the algorithm push order to be fault-free, and updating the sample library by taking the corrected fault prediction result and the operation data of the target equipment as a group of normal cases.
According to the sample library updating method provided by the invention, the preset matching condition comprises the following steps: the equipment number of the target equipment in the algorithm push order is the same as the equipment number in the real maintenance order, the generation time of the algorithm push order is earlier than the generation time of the real maintenance order, the fault prediction result in the algorithm push order is the same as the fault actual measurement result in the real maintenance order, and the fault processing result in the real maintenance order is that the fault is processed.
According to the sample library updating method provided by the invention, before the step of obtaining the target associated case, the method further comprises the following steps:
acquiring multiple groups of algorithm push orders and multiple groups of real maintenance orders;
if the key field in any algorithm push order is matched with the key field in any real maintenance order, taking the algorithm push order and the real maintenance order as a group of associated cases;
wherein the key field includes at least one of a device number, a fault type, a fault location, and a fault time.
The invention provides a sample library updating device, comprising:
the acquisition unit is used for acquiring a target associated case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; the failure prediction result is determined based on the operation data of the target equipment and a sample library;
the judging unit is used for judging the conversion relation between the algorithm push order and the real maintenance order based on a preset matching condition and determining whether the real maintenance order is the conversion result of the algorithm push order;
and the updating unit is used for updating the sample library by taking the fault prediction result in the algorithm push order and the operation data of the target equipment as a group of fault cases when the judgment result is yes.
According to the sample library updating device provided by the invention, the updating unit is further configured to:
and if the judgment result is negative, correcting the fault prediction result in the algorithm push order to be fault-free, and updating the sample library by taking the corrected fault prediction result and the operation data of the target equipment as a group of normal cases.
The sample library updating device provided by the invention further comprises:
the correlation unit is used for acquiring a plurality of groups of algorithm push orders and a plurality of groups of real maintenance orders; and when the key field in any algorithm push order is matched with the key field in any real maintenance order, taking the algorithm push order and the real maintenance order as a group of associated cases;
wherein the key field includes at least one of a device number, a fault type, a fault location, and a fault time.
The invention provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the steps of the sample library updating method when executing the program.
The present invention provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the sample library update method.
The invention provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of the sample library updating method.
The method, the device, the electronic equipment and the storage medium for updating the sample library provided by the invention judge the conversion relation between the algorithm push order and the real maintenance order in the target correlation case by obtaining the target correlation case according to the preset matching condition, if the real maintenance order is the conversion result of the algorithm push order, the fault prediction result in the algorithm push order and the operation data of the target equipment are used as a group of fault cases, update the sample library of the fault diagnosis algorithm in real time, the algorithm push order and the real maintenance order are the real operation data of the equipment and are easy to obtain, generate a new training sample in real time according to the two orders, fully utilize the real fault cases, improve the utilization rate of the real operation data of the equipment, automatically realize the update of the sample library of the fault diagnosis algorithm, and ensure that the sample library keeps effectiveness and real-time, the fault diagnosis algorithm can be iterated and optimized continuously, and the accuracy of the fault diagnosis algorithm is improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a sample library updating method according to the present invention;
FIG. 2 is a detailed diagram of a sample library updating method provided by the present invention;
FIG. 3 is a schematic diagram of an automatic case annotation system according to the present invention;
FIG. 4 is a schematic structural diagram of a sample library updating apparatus according to the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart of a sample library updating method provided by the present invention, as shown in fig. 1, the method includes:
step 110, obtaining a target associated case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; the fault prediction results are determined based on the operational data of the target device and the sample library.
Specifically, the sample library updating method provided by the invention is applied to a server or a cloud platform. The server or cloud platform may be configured as a remote diagnostic system for the target device. In which a fault diagnosis algorithm is run. The target device is a device that needs to be fault diagnosed. The target device may be a hardware device, such as a work machine, a data server, a machine tool, etc.; but also software devices such as meter monitoring systems, operation control systems, etc. The operation machine comprises an excavator, a pile driver, a loader, an overhead working truck, a crane, pumping equipment and the like.
The fault diagnosis algorithm can use a convolutional neural network, a full convolutional neural network, a cyclic neural network and the like as initial models and adopts samples in a sample library for training.
And the control system of the target equipment sends the operation data of the equipment to the server or the cloud platform, and the operation data is analyzed and processed by a fault diagnosis algorithm operated in the server or the cloud platform to obtain a fault prediction result of the target equipment. An algorithm is generated to push orders based on the fault prediction results.
The algorithm push order is a fault early warning order pushed to a client of the target device by the server or the cloud platform. The order includes information such as a device number of the target device, a health index of the device, a failure prediction result, and the like, wherein the failure prediction result includes a failure type, a failure occurrence probability, a failure occurrence location, failure occurrence time, and the like, and is used for prompting a client of information such as a possible failure of the target device, and the time and location of the failure.
The real maintenance order is an order generated after a maintainer of the target equipment carries out field detection on the target equipment, processes a real fault of the target equipment, and records time, type, position, processing method, processing result and the like of the fault.
The algorithm push order is generated after a fault diagnosis algorithm in the server or the cloud platform performs analysis processing according to the operation data of the target device, and a fault prediction result contained in the algorithm push order may be consistent with a fault actually occurring in the target device or inconsistent with the fault actually occurring in the target device, for example, the target device does not have a fault, and the fault is judged to occur by the fault diagnosis algorithm by mistake only because the operation data changes due to the change of the working condition.
The algorithm push order is generated after a remote diagnosis system in a server or a cloud platform monitors the operation data of the target equipment in real time; the real repair order is generated by a customer sending through a customer relationship management system requesting a service engineer to go to the site of equipment use to repair the equipment. The two types of orders come from different systems. The association case can be generated after the algorithm push order and the real maintenance order are associated. The method of association includes matching through key fields in both orders. The key fields may include equipment number, health indicator, fault type, time and location of fault, etc. The health index is a parameter for judging whether the equipment is in a normal state.
For example, key field matching may be performed in multiple algorithmic push orders and multiple real repair orders, with the equipment number as the key field. If the equipment numbers in any algorithm push order and any real maintenance order are the same, the two orders can be considered to be related to the same equipment, and the fault information may be related. The algorithm push order and the real repair order can be associated as a set of target associated cases.
And step 120, judging a conversion relation between the algorithm push order and the real maintenance order based on a preset matching condition, and determining whether the real maintenance order is a conversion result of the algorithm push order.
Specifically, when the algorithm push order in the target association case is matched with information such as a fault type in the real maintenance order, it may be considered that a conversion relationship exists between the algorithm push order and the real maintenance order, that is, after the user of the target device receives the algorithm push order at the client, it is determined that the target device may have a fault, a maintenance request is sent by the client relationship management system, and the real maintenance order is generated by a service engineer processing the fault on site. That is, the real repair order is the translation result of the algorithm push order.
The preset matching conditions can be set as required, for example, different matching conditions can be determined according to the equipment numbers, the failure times and the failure types in the two orders.
Step 130, if the real maintenance order is the conversion result of the algorithm push order, taking the fault prediction result in the algorithm push order and the operation data of the target device as a group of fault cases, and updating the sample library.
Specifically, if the real maintenance order is a conversion result of the algorithm push order, it may be considered that the fault prediction result obtained by the fault diagnosis algorithm by analyzing the operation data of the target device is accurate. The fault prediction results in the algorithm push order and the operation data of the target device may be used as a set of cases, which may be labeled as fault cases, and the fault cases may be used to update the sample library of the fault diagnosis algorithm.
For example, feature extraction may be performed on the operation data of the target device through feature engineering, the extracted features may reflect change features of parameters or indexes of the target device in a fault state, the extracted features may be used as a training sample, and a fault prediction result may be used as a label of the training sample and added to a sample library of a fault diagnosis algorithm.
The algorithm push order and the real maintenance order can be obtained in real time, the obtained fault sample is generated in real time, the real-time updating of the sample library is realized, and the accuracy of the fault diagnosis algorithm is improved.
According to the sample library updating method provided by the embodiment of the invention, the conversion relation between the algorithm-pushed order and the real maintenance order in the target associated case is judged by acquiring the target associated case and according to the preset matching condition; if the real maintenance order is the conversion result of the algorithm pushing order, taking a fault prediction result in the algorithm pushing order and the operation data of the target equipment as a group of fault cases, and updating a sample library of the fault diagnosis algorithm in real time; the algorithm push order and the real maintenance order are real operation data of the equipment and are easy to obtain, a new training sample is generated in real time according to the two orders, a real fault case is fully utilized, the utilization rate of the real operation data of the equipment is improved, the updating of a fault diagnosis algorithm sample library is automatically realized, the effectiveness and the real-time performance of the sample library are kept, the iteration and the optimization can be continuously carried out on the fault diagnosis algorithm, and the accuracy of the fault diagnosis algorithm is improved.
Based on the above embodiment, step 120 includes:
and if the real maintenance order is not the conversion result of the algorithm push order, correcting the fault prediction result in the algorithm push order to be fault-free, and updating the sample library by taking the corrected fault prediction result and the operation data of the target equipment as a group of normal cases.
Specifically, if the real maintenance order is not the conversion result of the order pushed by the algorithm, it may be considered that the fault prediction result obtained by the fault diagnosis algorithm by analyzing the operation data of the target device is inaccurate. At this time, the target device operates in a normal state.
The failure prediction result can be corrected to be failure-free. The corrected fault prediction result and the operation data of the target equipment are used as a group of cases which can be marked as normal cases, and the normal cases are used for updating a sample library of the fault diagnosis algorithm.
By the method, the sample library of the fault diagnosis algorithm comprises the fault samples and the normal samples, and the fault samples and the normal samples are generated through real operation data of the equipment. When the sample library is adopted to train the fault diagnosis algorithm, the recognition of the algorithm to the fault state and the normal state can be enhanced simultaneously, and the analysis and prediction capability of the algorithm is improved.
Based on any of the above embodiments, the preset matching conditions include that the equipment number in the algorithm push order is the same as the equipment number in the real maintenance order, the generation time of the algorithm push order is earlier than the generation time of the real maintenance order, the fault prediction result in the algorithm push order is the same as the fault actual measurement result in the real maintenance order, and the fault processing result in the real maintenance order is that the fault is processed.
Specifically, the equipment number in the algorithm push order is the same as the equipment number in the real maintenance order, which indicates that the algorithm push order and the real maintenance order belong to the same equipment.
The algorithm push order is generated at a time earlier than the real repair order, indicating that the real repair order was generated after the algorithm push order. According to the sequence of equipment prediction and maintenance, if a real maintenance order is earlier than an algorithm push order, the real maintenance order cannot be converted from the algorithm push order; if the real repair order is later than the algorithmic push order, the real repair order may be translated from the algorithmic push order.
The fault prediction result in the algorithm pushed order is the same as the fault actual measurement result in the real maintenance order, which indicates that the fault prediction result in the algorithm pushed order may be accurate, that is, the actually occurring fault type and fault position of the target device are consistent with the fault prediction result.
The fault processing result in the real maintenance order is that the fault is processed, which shows that the actually occurred fault type of the target equipment is removed after the fault is processed by the maintenance personnel, and further shows that the fault prediction result in the algorithm push order may be accurate.
Through the 4 preset matching conditions, whether the real maintenance order is the conversion result of the algorithm-pushed order can be effectively and accurately judged.
Based on any of the above embodiments, step 110 may be preceded by:
acquiring multiple groups of algorithm push orders and multiple groups of real maintenance orders;
if the key field in any algorithm push order is matched with the key field in any real maintenance order, taking the algorithm push order and the real maintenance order as a group of associated cases;
the key fields comprise equipment numbers, health indexes, fault types, fault time, fault positions and the like.
Specifically, the algorithm push orders may be obtained through a remote diagnostic system, the real repair orders may be obtained through a customer relationship management system, and the two types of orders are from different systems. In the real operation process, the orders generated by the two systems are not directly related or corresponding, and the data volume of the actually generated orders is large.
The algorithm push order and the real maintenance order can be associated through a key field matching method, and the two associated orders are used as a group of associated cases. And generating a basic case library according to the obtained multiple groups of associated cases for further matching and screening to obtain a sample library.
For example, the equipment number and the fault location may be used as key fields, and key field matching may be performed in a plurality of algorithm push orders and a plurality of real maintenance orders. If the equipment numbers and fault positions in any algorithm push order and any real maintenance order are the same, the two orders can be considered to be related to the same fault type of the same equipment, and fault information may be related. Thus, the algorithm push order can be associated with the real service order as a set of associated cases. Based on any of the above embodiments, the algorithm push order is determined based on the following steps:
receiving operation data sent by a control system of target equipment;
determining a fault prediction result of the target equipment based on the operation data and a fault diagnosis algorithm;
and generating an algorithm pushing order containing a fault prediction result.
Specifically, the control system of the target device may communicate with a server or a cloud platform running a remote diagnosis system through a communication module, and send the running data of the target device to the remote diagnosis system.
And after the remote diagnosis system acquires the operation data, inputting the operation data into a fault diagnosis algorithm to predict the operation state of the target equipment to obtain a fault prediction result. And the remote diagnosis system generates an algorithm pushing order according to the fault prediction result and sends the order to a control system of the target equipment or a client of the target equipment through the communication module.
Based on any of the above embodiments, the fault diagnosis algorithm corresponds to the type of the target device.
Specifically, the probability of the same type of device failing or the failure type and the like have similarity. And collecting fault records or operation data of the same type of equipment, and the trained fault diagnosis algorithm has high reference significance for fault early warning and fault diagnosis of all the equipment of the type.
Therefore, the fault diagnosis algorithm can be corresponding to the type of the equipment, and the fault diagnosis algorithm is trained by using the sample data of the same type as the target equipment, so that the accuracy of the fault diagnosis algorithm can be improved.
Based on any one of the embodiments, the invention provides a fault diagnosis sample library updating method based on a big data system and real operation data driving, the method combines two aspects of data of a fault diagnosis algorithm diagnosis result and a real operation condition provided by the big data system, and realizes monitoring of the algorithm application condition and case evaluation through a case evaluation system based on operation tracking, so that a large number of marked real cases are automatically obtained, and the fault diagnosis sample library is further continuously updated through an automatic marking system.
Fig. 2 is a detailed schematic diagram of a sample library updating method provided by the present invention, as shown in fig. 2, the method includes:
step one, a data interface is constructed based on enterprise big data infrastructure, and an algorithm push order generated by a fault diagnosis algorithm can be obtained from a user terminal; acquiring a real fault maintenance record, namely a real maintenance order, from an enterprise customer relationship management system interface; and pushing the association between the order and the real maintenance order by an algorithm based on the matching relation of key fields in the order, and storing the well-associated association case as a basic case into a database as a basic case library after the well-associated association case is coded.
Step two, a mapping relation between the algorithm push order and the real fault is established, a judgment rule is established based on the mapping relation, and the case in the basic case library is judged from the following four aspects: A. whether the equipment numbers are matched; B. the algorithm pushes whether the order generation time is earlier than the generation time of a real maintenance order; C. the fault prediction result in the algorithm pushing order is consistent with the fault actual measurement result in the real maintenance order; D. whether the health index of the equipment processed by the real maintenance order is recovered or whether the fault processing result in the real maintenance order is processed; and deducing whether the algorithm push order is successfully converted into a real maintenance order or not based on the judgment result.
Acquiring equipment numbers from a basic case library, and acquiring time series data [ X ] of all equipment from a large data platform according to the equipment numbers]The device number is no. _ s, the time is t, and the time sequence data of each device is [ X ]]No. s, including n different sensing device monitoring data and control system control signals respectively recorded as x1,x2,x2,x3,……,xnObtaining a health index of each device through a big data fault diagnosis algorithm output ═ m (input): index _ h ═ M (X)no._s,t。
Setting an appropriate threshold for Index _ h
Figure BDA0003517642830000111
When the health index is below the threshold
Figure BDA0003517642830000112
When the health index is equal to or higher than the threshold value, the equipment is considered to be out of order
Figure BDA0003517642830000121
The device is considered to be healthy.
The data of the enterprise customer relationship management system is obtained and recorded as CRM, wherein the CRM comprises information such as a fault equipment number, fault time, a fault position, a fault type and the like, and the CRM records the CRM as [ No. _ s, fault _ time, fault _ location, fault _ type and … ].
And constructing a mapping relation between the health Index and the Fault Type, and constructing a mapping function based on the mapping relation, wherein Fault _ Type is Type [ Index _ h ].
And (3) according to the standard whether the algorithm push order is successfully converted into a real maintenance order, setting a judgment condition:
and judging a condition A, wherein the equipment number in the algorithm push order is the same as the equipment number in the real maintenance order.
no._s[Index_h]＝no._s[CRM]
And judging a condition B: the algorithm pushes the generation time of the order, i.e., the health index is less than the threshold time, before the real repair order generation time.
Figure BDA0003517642830000122
And C, judging a condition C, wherein the fault type of the algorithm push order forecast is consistent with the fault type indicated by the real maintenance order.
Type(Index_h)＝[fault_type][crm]
And judging the condition D that the equipment recovers the health state after the real maintenance order is sent out, namely the equipment health index is equal to or higher than the threshold value.
Figure BDA0003517642830000123
And thirdly, constructing an automatic case annotation system based on operation data tracking, and dividing the cases into fault cases which are successfully converted and normal cases without real faults according to the set judgment conditions to realize automatic annotation of the cases. Fig. 3 is a working schematic diagram of the automatic case labeling system provided by the present invention, and as shown in fig. 3, whether the algorithm prediction order is successfully converted into the real maintenance order is determined according to the 4 determination conditions, and the associated case is automatically labeled.
A judgment function J is constructed, and the final judgment result can be expressed as R ═ J (a, B, C, D). If the R value is calculated to be 1, the judgment result is true, namely the case is a fault case of successful conversion; otherwise, the result is false, that is, the case is a normal case without failure.
And step four, constructing an automatic sample acquisition system, extracting the characteristics of the marked cases, warehousing and examining the extracted samples according to warehousing rules, and automatically adding the examined samples into a sample library to realize real-time updating of the sample library.
By constructing a complete system, all functional modules are organically integrated, the process is continuously and automatically realized, and the continuous updating and iteration of the sample library are realized.
Based on any of the above embodiments, fig. 4 is a schematic structural diagram of a sample library updating apparatus provided by the present invention, as shown in fig. 4, the apparatus includes:
an obtaining unit 410, configured to obtain a target association case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; the failure prediction result is determined based on the operation data of the target equipment and the sample library;
a determining unit 420, configured to determine, based on a preset matching condition, a conversion relationship between the algorithm pushed order and the real maintenance order, and determine whether the real maintenance order is a conversion result of the algorithm pushed order;
and the updating unit 430 is configured to update the sample library by using the fault prediction result in the algorithm push order and the operation data of the target device as a group of fault cases when the determination result is yes.
The sample base updating device provided by the embodiment of the invention judges the conversion relation between the algorithm push order and the real maintenance order in the target associated case by obtaining the target associated case and according to the preset matching condition, if the real maintenance order is the conversion result of the algorithm push order, the fault prediction result in the algorithm push order and the operation data of the target equipment are taken as a group of fault cases, the sample base of the fault diagnosis algorithm is updated in real time, the algorithm push order and the real maintenance order are the real operation data of the equipment and are easy to obtain, a new training sample is generated in real time according to the two orders, the real fault cases are fully utilized, the utilization rate of the real operation data of the equipment is improved, the updating of the fault diagnosis algorithm sample base is automatically realized, and the sample base keeps effectiveness and real-time performance, the fault diagnosis algorithm can be iterated and optimized continuously, and the accuracy of the fault diagnosis algorithm is improved.
Based on any embodiment above, the update unit is further configured to:
and if the judgment result is negative, correcting the fault prediction result in the algorithm push order to be fault-free, and updating the sample library by taking the corrected fault prediction result and the operation data of the target equipment as a group of normal cases.
Based on any of the above embodiments, the preset matching conditions include that the equipment number in the algorithm push order is the same as the equipment number in the real maintenance order, the generation time of the algorithm push order is earlier than the generation time of the real maintenance order, the fault prediction result in the algorithm push order is the same as the fault actual measurement result in the real maintenance order, and the fault processing result in the real maintenance order is that the fault is processed.
Based on any embodiment above, the apparatus further comprises:
the correlation unit is used for acquiring a plurality of groups of algorithm push orders and a plurality of groups of real maintenance orders; and when the key field in any algorithm push order is matched with the key field in any real maintenance order, taking the algorithm push order and the real maintenance order as a group of associated cases;
wherein the key field includes at least one of a device number, a fault type, a fault location, and a fault time.
Based on any of the above embodiments, the algorithm push order is determined based on the following steps:
receiving operation data sent by a control system of target equipment;
determining a fault prediction result of the target equipment based on the operation data and a fault diagnosis algorithm;
and generating an algorithm pushing order containing a fault prediction result.
Based on any of the above embodiments, the fault diagnosis algorithm corresponds to the type of the target device.
Based on any of the above embodiments, fig. 5 is a schematic structural diagram of an electronic device provided by the present invention, and as shown in fig. 5, the electronic device may include: a Processor (Processor)510, a communication Interface (Communications Interface)520, a Memory (Memory)530, and a communication Bus (Communications Bus)540, wherein the Processor 510, the communication Interface 520, and the Memory 530 communicate with each other via the communication Bus 540. Processor 510 may call logical commands in memory 530 to perform the following method:
acquiring a target associated case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; the failure prediction result is determined based on the operation data of the target equipment and the sample library; judging the conversion relation between the algorithm pushing order and the real maintenance order based on a preset matching condition, and determining whether the real maintenance order is the conversion result of the algorithm pushing order; and if so, taking the fault prediction result in the algorithm push order and the operation data of the target equipment as a group of fault cases, and updating the sample library.
In addition, the logic commands in the memory 530 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic commands are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of commands for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The processor in the electronic device provided in the embodiment of the present invention may call a logic instruction in the memory to implement the method, and the specific implementation manner of the method is consistent with the implementation manner of the method, and the same beneficial effects may be achieved, which is not described herein again.
Embodiments of the present invention further provide a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented to perform the method provided in the foregoing embodiments when executed by a processor, and the method includes:
acquiring a target associated case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; the failure prediction result is determined based on the operation data of the target equipment and the sample library; judging the conversion relation between the algorithm pushing order and the real maintenance order based on a preset matching condition, and determining whether the real maintenance order is the conversion result of the algorithm pushing order; and if so, taking the fault prediction result in the algorithm push order and the operation data of the target equipment as a group of fault cases, and updating the sample library.
When the computer program stored on the non-transitory computer readable storage medium provided in the embodiments of the present invention is executed, the method is implemented, and the specific implementation manner of the method is consistent with the implementation manner of the method, and the same beneficial effects can be achieved, which is not described herein again.
Embodiments of the present invention provide a computer program product, which includes a computer program, and when being executed by a processor, the computer program implements the steps of the method.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A method for updating a sample library, comprising:
acquiring a target associated case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; the failure prediction result is determined based on the operation data of the target equipment and a sample library;
judging the conversion relation between the algorithm push order and the real maintenance order based on a preset matching condition, and determining whether the real maintenance order is the conversion result of the algorithm push order;
and if so, taking the fault prediction result in the algorithm push order and the operation data of the target equipment as a group of fault cases, and updating the sample library.
2. The method of claim 1, wherein the step of determining whether the real repair order is a conversion result of the algorithm push order further comprises:
and if not, correcting the fault prediction result in the algorithm push order to be fault-free, and updating the sample library by taking the corrected fault prediction result and the operation data of the target equipment as a group of normal cases.
3. The method according to claim 1 or 2, wherein the preset matching condition comprises: the equipment number of the target equipment in the algorithm push order is the same as the equipment number in the real maintenance order, the generation time of the algorithm push order is earlier than the generation time of the real maintenance order, the fault prediction result in the algorithm push order is the same as the fault actual measurement result in the real maintenance order, and the fault processing result in the real maintenance order is that the fault is processed.
4. The method for updating the sample library according to claim 1 or 2, wherein before the step of obtaining the target associated case, the method further comprises:
acquiring multiple groups of algorithm push orders and multiple groups of real maintenance orders;
if the key field in any algorithm push order is matched with the key field in any real maintenance order, taking the algorithm push order and the real maintenance order as a group of associated cases;
wherein the key field includes at least one of a device number, a fault type, a fault location, and a fault time.
5. A sample library update apparatus, comprising:
the acquisition unit is used for acquiring a target associated case; the target associated case comprises an algorithm pushing order and a real maintenance order; the algorithm pushes the order to contain the fault prediction result of the target equipment; the failure prediction result is determined based on the operation data of the target equipment and a sample library;
the judging unit is used for judging the conversion relation between the algorithm push order and the real maintenance order based on a preset matching condition and determining whether the real maintenance order is the conversion result of the algorithm push order;
and the updating unit is used for updating the sample library by taking the fault prediction result in the algorithm push order and the operation data of the target equipment as a group of fault cases when the judgment result is yes.
6. The sample library updating apparatus according to claim 5, wherein the updating unit is further configured to:
and if the judgment result is negative, correcting the fault prediction result in the algorithm push order to be fault-free, and updating the sample library by taking the corrected fault prediction result and the operation data of the target equipment as a group of normal cases.
7. The apparatus according to claim 5 or 6, further comprising:
the correlation unit is used for acquiring a plurality of groups of algorithm push orders and a plurality of groups of real maintenance orders; and when the key field in any algorithm push order is matched with the key field in any real maintenance order, taking the algorithm push order and the real maintenance order as a group of associated cases;
wherein the key field includes at least one of a device number, a fault type, a fault location, and a fault time.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the sample library update method according to any one of claims 1 to 4 are implemented when the program is executed by the processor.
9. A non-transitory computer readable storage medium, having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the steps of the sample library update method of any of claims 1 to 4.
10. A computer program product comprising a computer program, wherein the computer program when executed by a processor implements the steps of the sample library update method of any one of claims 1 to 4.
CN202210171055.3A 2022-02-23 2022-02-23 Sample library updating method and device, electronic equipment and storage medium Pending CN114548176A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210171055.3A CN114548176A (en) 2022-02-23 2022-02-23 Sample library updating method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210171055.3A CN114548176A (en) 2022-02-23 2022-02-23 Sample library updating method and device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN114548176A true CN114548176A (en) 2022-05-27

Family

ID=81677948

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210171055.3A Pending CN114548176A (en) 2022-02-23 2022-02-23 Sample library updating method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN114548176A (en)

Similar Documents

Publication Publication Date Title
CN111459700B (en) Equipment fault diagnosis method, diagnosis device, diagnosis equipment and storage medium
CN109800127A (en) A kind of system fault diagnosis intelligence O&M method and system based on machine learning
CN111259947A (en) Power system fault early warning method and system based on multi-mode learning
CN117196066A (en) Intelligent operation and maintenance information analysis model
CN109374631B (en) Tunnel state evaluation method
CN111915026A (en) Fault processing method and device, electronic equipment and storage medium
CN110597235A (en) Universal intelligent fault diagnosis method
CN116416884B (en) Testing device and testing method for display module
CN117234785B (en) Centralized control platform error analysis system based on artificial intelligence self-query
CN115269314A (en) Transaction abnormity detection method based on log
CN114968727B (en) Database through infrastructure fault positioning method based on artificial intelligence operation and maintenance
CN115225536A (en) Unsupervised learning-based virtual machine anomaly detection method and system
CN117194919A (en) Production data analysis system
CN116150191A (en) Data operation acceleration method and system for cloud data architecture
CN116932523B (en) Platform for integrating and supervising third party environment detection mechanism
CN117114420B (en) Image recognition-based industrial and trade safety accident risk management and control system and method
CN114548176A (en) Sample library updating method and device, electronic equipment and storage medium
CN115600695A (en) Fault diagnosis method of metering equipment
CN113656287B (en) Method and device for predicting software instance faults, electronic equipment and storage medium
CN115765182A (en) Digital twinning method and device based on time sequence database
CN113076217B (en) Disk fault prediction method based on domestic platform
CN114398347A (en) Fuel cell data analysis system and electronic device based on data correlation
CN114416417A (en) System abnormity monitoring method, device, equipment and storage medium
CN116738347A (en) Fault analysis positioning method and system based on machine learning
CN116089396A (en) Simulation test data management method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination