Device failure knowledge recommendation method and system

Abstract

The present disclosure discloses a device fault knowledge recommendation method and system, and relates to the field of fault maintenance. The method comprises: establishing a product network comprising multiple levels; analyzing the operation data of each device based on a digital twin algorithm and an artificial intelligence algorithm to obtain the fault knowledge of each device; integrating and fusing the fault knowledge of each device according to the product network to obtain fault knowledge of different levels, and storing the fault knowledge of different levels in the database of the corresponding level; and identifying the fault state corresponding to the real-time operation parameter of the device to be processed based on the digital twin model, and recommending the fault knowledge of different levels of the device to be processed obtained by querying in the database. The present disclosure can improve the accuracy of fault identification and knowledge recommendation, thereby increasing the real-time maintenance capability of the device.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of fault maintenance, and particularly relates to a device fault knowledge recommendation method and system. BACKGROUND

[0002] Complex high-end equipment has the characteristics of high system complexity and complex maintenance operation. In the actual use and operation process of the equipment, the existing fault knowledge can realize the real-time fault diagnosis of the equipment, and then the maintenance records in the database can be referred to to provide guidance for the maintenance of the complex high-end equipment.

[0003] In the related art, by searching and scoring the historical maintenance records, the maintenance personnel are recommended with the maintenance schemes with high similarity, and then the familiar faults are quickly maintained. Or, on the basis of creating knowledge, by analyzing the fault related information and searching the keywords in the system, the fault related knowledge recommendation service is realized. The fault identification and maintenance scheme provided by the two ways have low accuracy, thereby the maintenance efficiency and quality of the equipment are low. SUMMARY

[0004] One of the technical problems to be solved by the present disclosure is to provide a device fault knowledge recommendation method and system, which can improve the accuracy of fault identification and knowledge recommendation, and thereby increase the real-time maintenance capability of the equipment.

[0005] According to an aspect of the present disclosure, a device fault knowledge recommendation method is provided, comprising: establishing a product network comprising a plurality of levels; based on a digital twin algorithm and an artificial intelligence algorithm, analyzing the operation data of each device to obtain the fault knowledge of each device; based on the product network, integrating and fusing the fault knowledge of each device to obtain fault knowledge of different levels, and storing the fault knowledge of different levels in the corresponding level database; and based on the digital twin model, identifying the fault state corresponding to the real-time operation parameters of the to-be-processed device, and recommending the fault knowledge of different levels of the to-be-processed device obtained by querying in the database.

[0006] In some embodiments, establishing the product network comprising a plurality of levels comprises: establishing a product node for each device and obtaining operation data of each device; establishing a product-level database capable of storing data and failure knowledge of each product node; performing cluster analysis on a plurality of devices to obtain a plurality of product families; establishing a product family node for each product family and establishing a connection relationship between each product node and the product family node to which the product node belongs; establishing a product family-level database capable of storing data and product-level failure knowledge of each product family node according to the connection relationship between each product node and the product family node to which the product node belongs; establishing a product system node and establishing a connection relationship between the product system node and each product family node; and establishing a product system-level database capable of storing product system-level failure knowledge according to the connection relationship between the product system node and each product family node.

[0007] In some embodiments, analyzing the operation data of each device based on the digital twin algorithm and the artificial intelligence algorithm to obtain failure knowledge of each device comprises: analyzing the operation data of each device based on the artificial intelligence algorithm to obtain failure knowledge of each device; simulating the operation state of each device based on the digital twin model and the operation data of each device to obtain a failure state of each device; and verifying the failure knowledge of each device based on the failure state of each device.

[0008] In some embodiments, according to the operation data of each device, a first key performance indicator affecting the operation of the device is obtained; the preprocessed operation data of each device is input into the digital twin model to simulate and emulate the operation process of each device, and a simulated second key performance indicator is obtained; and the digital twin model is trained and optimized according to the comparison result of the first key performance indicator and the second key performance indicator.

[0009] In some embodiments, based on the digital twin model, identifying a failure corresponding to a real-time operation parameter of a to-be-processed device and recommending different levels of failure knowledge of the to-be-processed device obtained by querying the database comprises: determining that the to-be-processed device is in a parameter abnormal state when the real-time operation parameter of the to-be-processed device exceeds a parameter threshold range; inputting the key operation parameter of the to-be-processed device into the digital twin model to simulate the operation state of the to-be-processed device; if it is determined that the to-be-processed device fails according to the simulation result, generating a failure knowledge query request according to the failure state information and the parameter abnormal state information of the to-be-processed device, so as to query different levels of failure knowledge of the to-be-processed device according to the failure knowledge query request.

[0010] In some embodiments, querying the different levels of fault knowledge of the to-be-processed device comprises: querying whether the product-level database contains the fault knowledge corresponding to the to-be-processed device; if the product-level database contains the fault knowledge corresponding to the to-be-processed device, pushing the fault knowledge corresponding to the to-be-processed device, otherwise, querying whether the product family-level database contains the product family-level fault knowledge corresponding to the to-be-processed device; if the product family-level database contains the product family-level fault knowledge corresponding to the to-be-processed device, pushing the product family-level fault knowledge corresponding to the to-be-processed device, otherwise, querying whether the product system-level database contains the system-level fault knowledge corresponding to the to-be-processed device; if the product system-level database contains the system-level fault knowledge corresponding to the to-be-processed device, pushing the system-level fault knowledge corresponding to the to-be-processed device, otherwise, sending the fault state information and the parameter abnormal state information of the to-be-processed device to the maintenance personnel.

[0011] In some embodiments, the different levels of fault knowledge of the device are pushed in the form of augmented reality (AR) or virtual reality (VR).

[0012] In some embodiments, the fault knowledge comprises fault information and corresponding fault maintenance information.

[0013] According to another aspect of the present disclosure, a device fault knowledge recommendation system is also provided, comprising: a product network establishing unit configured to establish a product network comprising a plurality of levels; a fault knowledge obtaining unit configured to analyze the operation data of each device based on a digital twin algorithm and an artificial intelligence algorithm, obtain the fault knowledge of each device, integrate and fuse the fault knowledge of each device according to the product network, obtain the fault knowledge of different levels, and store the fault knowledge of different levels in the corresponding database; and a fault knowledge recommendation unit configured to identify the fault state corresponding to the real-time operation parameter of the to-be-processed device based on a digital twin model, and recommend the different levels of fault knowledge of the to-be-processed device obtained by querying in the database.

[0014] According to another aspect of the present disclosure, a device fault knowledge recommendation system is also provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute the device fault knowledge recommendation method as described above based on the instructions stored in the memory.

[0015] According to another aspect of the present disclosure, a computer readable storage medium is also provided, which stores computer program instructions, the instructions being executed by a processor to implement the device fault knowledge recommendation method as described above.

[0016] In the embodiments of the present disclosure, a multi-level product network is established, real-time sharing of device data and information can be realized, different levels of device fault knowledge can be mined based on digital twin technology and intelligent algorithms, real-time identification of faults and pushing of related fault solutions can be realized based on the digital twin model, the accuracy of fault identification and knowledge recommendation can be improved, and the real-time maintenance capability of the device is increased.

[0017] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings, which form a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0019] The present disclosure can be understood more readily by reference to the following detailed description of exemplary embodiments of the present disclosure and the attached drawings, of which:

[0020] Figure 1 Flowchart of some embodiments of the device fault knowledge recommendation method of the present disclosure.

[0021] Figure 2 Structural diagram of some embodiments of the product network of the present disclosure.

[0022] Figure 3 Flowchart of some embodiments of the product network establishment of the present disclosure.

[0023] Figure 4 Flowchart of some embodiments of the acquisition process of the various levels of fault knowledge of the present disclosure.

[0024] Figure 5 Flowchart of some embodiments of the recommended fault knowledge of the present disclosure.

[0025] Figure 6 Flowchart of some other embodiments of the recommended fault knowledge of the present disclosure.

[0026] Figure 7 Structural diagram of some embodiments of the device fault knowledge recommendation system of the present disclosure.

[0027] Figure 8 Structural diagram of some other embodiments of the device fault knowledge recommendation system of the present disclosure.

[0028] Figure 9 Structural diagram of some other embodiments of the device fault knowledge recommendation system of the present disclosure. DETAILED DESCRIPTION

[0029] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. Note that the relative arrangement, numerical expressions, and numerical values of components and steps set forth in these embodiments are not limiting to the scope of the present disclosure unless specifically stated otherwise.

[0030] It should be understood, of course, that the dimensions of the various parts illustrated in the various drawings are shown for simplicity and the actual dimensions can depend on the specific application.

[0031] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the scope of the disclosure, its application, or uses.

[0032] Techniques, methods, and devices known to those of ordinary skill in the relevant art can not be discussed in detail herein, but should be considered as part of the description of the present disclosure where appropriate.

[0033] In all examples shown and discussed herein, any specific values should be interpreted as merely illustrative and not as a limitation. Thus, other examples of the exemplary embodiments can have different values.

[0034] It should be noted that like reference numerals and letters refer to like items in the following drawings, and thus, once an item is defined in one drawing, it need not be discussed further in subsequent drawings.

[0035] To make the purposes, technical solutions, and advantages of the present disclosure clearer, the present disclosure is further described in detail below with reference to specific embodiments and in conjunction with the accompanying drawings.

[0036] Figure 1 Flowchart of some embodiments of the device fault knowledge recommendation method of the present disclosure.

[0037] In step 110, a product network including multiple levels is established.

[0038] In some embodiments, as shown in Figure 2 a three-level product hierarchical network including product level, product family level, and product system level is established, each level having a corresponding database to realize real-time sharing of information, data, and knowledge.

[0039] In step 120, the operation data of each device is analyzed based on a digital twin algorithm and an artificial intelligence algorithm to obtain the fault knowledge of each device.

[0040] In some embodiments, the operation data of each device is analyzed based on an artificial intelligence algorithm to obtain fault knowledge of each device, the operation state of each device is simulated based on the operation data of each device by using a digital twin model to obtain a fault state of each device, and the fault knowledge of each device is verified based on the fault state of each device.

[0041] In some embodiments, the device is, for example, complex high-end equipment. The fault knowledge includes fault information and corresponding fault maintenance information.

[0042] In step 130, the fault knowledge of each device is integrated and fused based on the product network to obtain fault knowledge at different levels, and the fault knowledge at different levels is stored in the database at the corresponding level.

[0043] Through steps 120 and 130, on the basis of multi-source heterogeneous real-time data obtained in the hierarchical enterprise product network, the digital twin technology and the artificial intelligence algorithm are fused to obtain the fault knowledge of the device, and the fault knowledge is integrated and fused, which can improve the quality and effectiveness of the fault knowledge.

[0044] In step 140, based on the digital twin model, a fault state corresponding to a real-time operation parameter of a to-be-processed device is identified, and different levels of fault knowledge of the to-be-processed device obtained by querying the database are recommended.

[0045] In some embodiments, when the real-time operation parameter of the to-be-processed device exceeds the parameter threshold range, it is determined that the to-be-processed device is in a parameter abnormal state; the key operation parameter of the to-be-processed device is input into the digital twin model to simulate the operation state of the to-be-processed device; if it is determined that the to-be-processed device fails according to the simulation result, a fault knowledge query request is generated according to the fault state information and the parameter abnormal state information of the to-be-processed device, so that different levels of fault knowledge of the to-be-processed device are queried according to the fault knowledge query request.

[0046] Since the digital twin is an integrated multi-physical, multi-granularity and probabilistic simulation, it fully utilizes product physical models, sensor updated data, historical records, etc. to map product life cycle activities. In this step, based on the identification of the real-time state of the device, the performance of the device can be simulated by using the digital twin model, and then the simulation result and the actual data are integrated to match and query the fault solution of the product at different levels.

[0047] In the above embodiment, a multi-level product network is established, real-time sharing of device data and information is realized, different levels of device fault knowledge is mined based on digital twin technology and intelligent algorithms, real-time identification of faults and pushing of relevant fault solutions are realized based on the digital twin model, the accuracy of fault identification and knowledge recommendation is improved, and the real-time maintenance capability of the device is increased.

[0048] Figure 3 Flowchart of some embodiments of the present disclosure for establishing a product network.

[0049] In step 310, a product node is established for each device, and running data of each device is obtained.

[0050] In some embodiments, a product node is established for each complex high-end equipment in the product network, the intelligent complex high-end equipment is communicated through the network node, and real-time data of the sensors configured on the complex high-end equipment is obtained.

[0051] In step 320, a product-level database capable of storing data and fault knowledge of each product node is established.

[0052] In step 330, a plurality of product families are obtained by cluster analysis of a plurality of devices, a product family node is established for each product family, and a connection relationship between each product node and the product family node to which it belongs is established.

[0053] In some embodiments, the complex high-end equipment is clustered into different product families according to the functions and tasks that can be completed by the complex high-end equipment, and corresponding product family nodes are established, each product family node is connected with the product nodes under it.

[0054] In step 340, a product family-level database capable of storing data and product-level fault knowledge of each product family node is established according to the connection relationship between each product node and the product family node to which it belongs.

[0055] In some embodiments, the product-level fault knowledge is general product family fault knowledge.

[0056] In step 350, a product system node is established, and a connection relationship between the product system node and each product family node is established.

[0057] In step 360, a product system-level database capable of storing product system-level fault knowledge is established according to the connection relationship between the product system node and each product family node.

[0058] In some embodiments, the product system-level fault knowledge includes system-level general knowledge and experiential knowledge for guiding product operation and maintenance.

[0059] In subsequent operations, based on different product network nodes, through hierarchical, classified query and search of the multi-level database, real-time sharing of information, data and knowledge in the entire product system can be realized. The product-level fault knowledge can be directly applied to the maintenance of the product. After the product family-level fault knowledge is pushed to the maintenance personnel, the maintenance personnel can select a more feasible solution for maintenance in combination with the actual fault characteristics. The product system-level fault knowledge can provide the maintenance personnel with method and strategy guidance for the maintenance of this type of fault.

[0060] In the above embodiments, on the basis of establishing a hierarchical enterprise product network, product databases of different levels are constructed, efficient sharing of different types of fault knowledge in the entire enterprise product system is realized, and the utilization rate of fault knowledge can be improved.

[0061] Figure 4 Flowchart of some embodiments of the acquisition process of each level of fault knowledge of the present disclosure.

[0062] In step 410, the operation data of each device is cleaned, compressed and data-converted.

[0063] In some embodiments, the cleaning operation includes detecting repeated items and missing items in the data, and removing repeated and incomplete data from the data set.

[0064] In some embodiments, the compression operation includes using data cubes, data vectors and other data formats to compress repeated parameters in the same type of data, so as to improve the quality and logicality of data storage.

[0065] In some embodiments, the data conversion operation includes using vector transformation, data extraction, normalization method, etc. to convert the source data into a data format that meets the input requirements of the digital twin model simulation analysis and artificial intelligence algorithm model, so as to improve the analyzability of the data.

[0066] In this step, through the preprocessing of the operation data, the effectiveness of the multi-source heterogeneous real-time data in the data analysis and knowledge mining process can be improved.

[0067] In step 420, the first key performance indicator affecting the operation of each device is obtained according to the operation data of each device.

[0068] In some embodiments, based on the actual operation state and data of the device, the first key performance indicator affecting the normal operation of the device is obtained through the performance calculation formula of the device and the output parameters of the device. The key performance indicator refers to the parameter that can affect the operation state of the device, such as speed and energy consumption. The speed can be directly collected, and the energy consumption parameter needs to be calculated.

[0069] In some embodiments, the formula of the first key performance indicator is: KeyPerfIn={kpi1, kpi2,..., kpi n}, where kpi s represents the s-th key performance indicator, and n represents the number of key performance indicators.

[0070] In step 430, the pre-processed operation data of each device is input into the digital twin model, and the operation process of each device is simulated to obtain the simulated second key performance indicator.

[0071] In some embodiments, the formula of the second key performance indicator is: SimuPerfIn={vpi1, vpi2,..., vpi n}, where vpi s represents the s-th simulated key performance indicator, and n represents the number of simulated key performance indicators.

[0072] In step 440, the digital twin model is trained and optimized according to the comparison result of the first key performance indicator and the second key performance indicator.

[0073] In some embodiments, the first key performance indicator and the second key performance indicator are compared, and when the difference between the two exceeds the acceptable threshold δ, it indicates that the digital twin model deviates greatly from the real operation state of the physical entity, and the digital twin model needs to be retrained and iteratively optimized. The process can be represented as:

[0074] if(|kpi s -vpi s |>δ s )

[0075] then(update virtual models)

[0076] δ={δ1, δ2,..., δ n}

[0077] where δ s represents the threshold size of the s-th key performance indicator allowable deviation.

[0078] Through steps 420-440, the actual operation of the device is compared with the simulation result, and the digital twin model is updated to make the digital twin model be able to accurately map the operation state of the device.

[0079] In step 450, the operating data of each device is analyzed based on artificial intelligence algorithms to obtain fault knowledge for each device. Based on the operating data of each device, the operating status of each device is simulated using a digital twin model to obtain the fault status of each device. Based on the fault status of each device, the fault knowledge of each device is verified.

[0080] In some embodiments, a series of intelligent algorithms, such as deep neural networks, association rule mining, and decision tree models, are employed to analyze real-time operational fault data of the equipment and acquire equipment fault knowledge. This fault knowledge includes fault types, fault causes, and fault solutions. A digital twin model is then used to verify the fault knowledge obtained from real-time operation. After digital analysis and simulation verification, the product fault-related information is comprehensively evaluated and integrated to obtain product-level fault knowledge of the equipment, which is stored in a product-level database. This fault knowledge can be directly used for real-time equipment maintenance.

[0081] In some embodiments, product-level fault knowledge is represented as:

[0082] if(Tpye Pknowledge Fault Pknowledge parameters Pknowledge )

[0083] then(method Pmaintenance )

[0084] Among them, Type Pknowledge Indicates the device model, Fault Pknowledge Indicates the fault that occurred, parameters Pknowledge Indicates key abnormal parameters of equipment operation, method pmaintenance This indicates a maintenance method or plan.

[0085] In step 460, based on the product network, the fault knowledge of each device is integrated and fused to obtain fault knowledge at different levels, and the fault knowledge at different levels is stored in the corresponding level database.

[0086] In some embodiments, data fusion and integration methods such as entity alignment and conflict resolution are employed to analyze and process fault knowledge existing throughout the entire product family, and integrate it into product family-level fault knowledge. This fault knowledge includes product family-wide general fault solutions, providing multiple general product maintenance solutions for real-time equipment maintenance. Maintenance personnel can select similar fault solutions based on fault characteristics, thereby improving maintenance efficiency.

[0087] In some embodiments, product family-level fault knowledge is represented as:

[0088] if (FmailyID Fknowledge , Fault Fknowledge , parameters rknowledge )

[0089] then (method Fmaintenance )

[0090] method Fmaintenance = {(pro1, feature1, solution1), (pro2,

[0091] feature2, solution2),..., (pro d , feature d , solution d )}

[0092] wherein, FmailyID Fknowledge represents the type of product family, Fault Fknowledge represents the occurred fault, parameters Fknowledge represents the key abnormal parameters of product family equipment operation, method pmaintenance represents the maintenance method or scheme set, each maintenance method or scheme contains the probability pro i of its occurrence, the significant external feature feature i , and the specific solution solution i , and then the maintenance personnel can select the appropriate maintenance mode according to experience.

[0093] In some embodiments, on the basis of obtaining product family level fault knowledge, the maintenance guidance information is extracted and generalized to obtain product system level product fault suggestive knowledge.

[0094] In some embodiments, the product system level fault knowledge is represented as:

[0095] if (Fault Sknowledge )

[0096] then (Guideline Fmaintenance )

[0097] wherein, Fault Sknowledge represents the occurred fault, and Guideline Fmaintenance represents the provided maintenance guidance, including possible fault causes, possible fault positions, and maintenance suggestions.

[0098] In the above embodiments, by fusing digital twinning and artificial intelligence algorithms, device operation data is mined, product fault knowledge is obtained, and knowledge is integrated and fused to obtain product fault solutions at different levels, thereby improving the utilization rate of product data and the effectiveness and availability of product fault knowledge.

[0099] Figure 5 Flowchart of some embodiments of the recommended fault knowledge of the present disclosure.

[0100] In step 510, when the real-time operation parameters of the device to be processed exceed the parameter threshold range, it is determined that the device to be processed is in a parameter abnormal state.

[0101] In some embodiments, by configuring embedded and external sensors on the physical entity of the device, real-time operation parameters are obtained, and the operation parameters are compared with the parameter threshold. When the operation parameters exceed the parameter threshold, it indicates that the device has generated an abnormal signal.

[0102] In some embodiments, the judgment of the abnormal signal is represented as:

[0103] para i ≥P ULi or para i ≤P LLi ,(i=1,2,...,b)

[0104] wherein para i represents the i-th operation parameter of the device, P ULi and P LLi represent the upper threshold and lower threshold of the i-th operation parameter of the device, respectively.

[0105] In step 520, the key operation parameters of the device to be processed are input into the digital twinning model, and the operation state of the device to be processed is simulated.

[0106] In some embodiments, the key operation parameters of the device are input into the corresponding digital twinning simulation model to simulate the actual operation process of the device and determine whether the device has a fault and what kind of fault is generated.

[0107] In some embodiments, the result of the digital twinning simulation is represented as:

[0108] SimuRes=DTmodel(para1,para j ,...,para g )

[0109] wherein DTmodel represents the digital twinning model of the device.

[0110] At step 530, if it is determined that the to-be-processed equipment fails according to the simulation result, a fault knowledge query request is generated according to the fault state information and the parameter abnormal state information of the to-be-processed equipment, so as to query the fault knowledge of different levels of the to-be-processed equipment according to the fault knowledge query request.

[0111] In some embodiments, the fault state information and the parameter abnormal state information simulated in combination with the digital twin model are used to obtain query request information for querying the maintenance solution, which is represented as:

[0112] Request={ProductID, SimuRes, abPara}

[0113] abPara={para i |para i ≥P ULi or para i ≤P LLi},(i=1,2,...,b)

[0114] ProductID={TypeID, FmailyID}

[0115] wherein ProductID represents ID information of the product, including product model information TypeID and product family information FmailyID, and abpara represents a set of key operating parameters that are abnormal.

[0116] In some embodiments, the query request information is sent to a data cloud to execute product solutions of different levels.

[0117] In this embodiment, active diagnosis and maintenance services during the delivery process of the equipment can be realized, and the maintenance efficiency and quality of the equipment are improved.

[0118] Figure 6 Flowchart of another embodiment of the recommended fault knowledge of the present disclosure.

[0119] At step 610, it is queried whether the product-level database contains the fault knowledge corresponding to the to-be-processed equipment, if yes, step 620 is executed, otherwise, step 630 is executed.

[0120] In some embodiments, the product individual and the product-level database of the same product model are queried, and the obtained results can be directly used for solution knowledge of product maintenance.

[0121] In some embodiments, the query matching process is represented as:

[0122]

[0123] In step 620, the corresponding fault knowledge of the device to be processed is pushed.

[0124] In step 630, it is queried whether the product family level database contains the product family level fault knowledge corresponding to the device to be processed. If yes, step 640 is executed, otherwise, step 650 is executed.

[0125] In this step, the general knowledge of the product family is found through the product family level database. The fault knowledge includes solutions to multiple related faults or abnormalities. Maintenance personnel select a more feasible solution for maintenance according to the actual fault characteristics, thereby improving product maintenance efficiency.

[0126] In some embodiments, the query matching process is represented as:

[0127]

[0128] In step 640, the product family level fault knowledge corresponding to the device to be processed is pushed.

[0129] In step 650, it is queried whether the product system level database contains the system level fault knowledge corresponding to the device to be processed. If yes, step 660 is executed, otherwise, step 670 is executed.

[0130] In step 660, the system level fault knowledge corresponding to the device to be processed is pushed.

[0131] In this step, the maintenance suggestion for guiding the fault is found through the product system level database. When a new fault or abnormality occurs in the product, intelligent maintenance suggestions are quickly provided to maintenance personnel.

[0132] In some embodiments, the query matching process is represented as:

[0133] (SimuRes=Fault Sknowledge )=true

[0134] In step 670, the fault state information and parameter abnormal state information of the device to be processed are sent to the maintenance personnel.

[0135] In this step, since the solution cannot be found in the database, the fault or abnormality information is sent to the maintenance personnel for manual diagnosis and development of maintenance measures, and the maintenance process is recorded to generate a related fault solution.

[0136] In some embodiments, the fault knowledge is simulated and verified based on a digital twin model to ensure the robustness and effectiveness of the fault knowledge. The fault knowledge that passes the verification is updated to the database.

[0137] In the above embodiment, on the basis of real-time data acquisition and evaluation and digital twin model simulation analysis, a hierarchical product knowledge recommendation system is established, and then real-time and accurate fault identification and different level fault maintenance scheme recommendations are provided for the operation and maintenance process of the equipment, so that the maintenance efficiency and quality of the equipment are improved.

[0138] In some other embodiments of the present disclosure, different levels of fault knowledge of the equipment are pushed in the form of AR (Augmented Reality) or VR (Virtual Reality). The pushed fault knowledge includes product fault solution, product maintenance guidance video related to the fault, virtual maintenance process information based on digital twin, etc.

[0139] In this embodiment, the product fault solution and related guidance information are pushed to the on-site product maintenance personnel in real time through AR and VR means, so as to realize immersive operation and maintenance guidance in virtual and real interactive mode.

[0140] Figure 7 A structural schematic diagram of some embodiments of the equipment fault knowledge recommendation system of the present disclosure. The system includes a product network establishing unit 710, a fault knowledge acquiring unit 720, and a fault knowledge recommending unit 730.

[0141] The product network establishing unit 710 is configured to establish a product network containing multiple levels.

[0142] In some embodiments, a three-layer product hierarchical network containing product level, product family level, and product system level is established, each level has a corresponding database, and real-time sharing of information, data, and knowledge is realized.

[0143] In some embodiments, the product network establishing unit 710 establishes product nodes for each device and acquires operation data of each device; establishes a product level database capable of storing data and fault knowledge of each product node; performs clustering analysis on multiple devices to obtain multiple product families; establishes product family nodes for each product family and establishes connection relationship between each product node and the product family node to which it belongs; according to the connection relationship between each product node and the product family node to which it belongs, a product family level database capable of storing data and product level fault knowledge of each product family node is established; product system nodes are established, and connection relationship between the product system nodes and each product family node is established; according to the connection relationship between the product system nodes and each product family node, a product system level database capable of storing product system level fault knowledge is established.

[0144] In this embodiment, on the basis of establishing a hierarchical enterprise product network, product databases of different levels are constructed, efficient sharing of different types of fault knowledge in the entire enterprise product system is realized, and the utilization rate of fault knowledge can be improved.

[0145] The fault knowledge acquisition unit 720 is configured to analyze the operation data of each device based on a digital twin algorithm and an artificial intelligence algorithm, acquire the fault knowledge of each device, integrate and fuse the fault knowledge of each device according to the product network, obtain fault knowledge of different levels, and store the fault knowledge of different levels in the database of the corresponding level.

[0146] In some embodiments, the fault knowledge acquisition unit 720 is configured to analyze the operation data of each device based on an artificial intelligence algorithm to obtain the fault knowledge of each device; simulate the operation state of each device based on the operation data of each device using a digital twin model to obtain the fault state of each device; and verify the fault knowledge of each device based on the fault state of each device.

[0147] By fusing the digital twin and artificial intelligence algorithms, mining the device operation data, acquiring product fault knowledge, and integrating and fusing the knowledge, product fault solutions of different levels are obtained, and the utilization rate of product data and the effectiveness and availability of product fault knowledge are improved.

[0148] In some embodiments, the fault knowledge acquisition unit 720 is further configured to clean, compress, and data-convert the operation data of each device.

[0149] Through the preprocessing of the operation data, the effectiveness of the multi-source heterogeneous real-time data in the data analysis and knowledge mining process can be improved.

[0150] The fault knowledge recommendation unit 730 is configured to identify the fault state corresponding to the real-time operation parameter of the to-be-processed device based on the digital twin model, and recommend the fault knowledge of different levels of the to-be-processed device obtained by querying in the database.

[0151] In some embodiments, when the real-time operation parameter of the to-be-processed device exceeds the parameter threshold range, it is determined that the to-be-processed device is in a parameter abnormal state; the key operation parameter of the to-be-processed device is input into the digital twin model to simulate the operation state of the to-be-processed device; if it is determined that the to-be-processed device fails according to the simulation result, a fault knowledge query request is generated according to the fault state information and the parameter abnormal state information of the to-be-processed device, so as to query the fault knowledge of different levels of the to-be-processed device according to the fault knowledge query request.

[0152] In some embodiments, it is inquired whether the product-level database contains the fault knowledge corresponding to the to-be-processed device; if the product-level database contains the fault knowledge corresponding to the to-be-processed device, the fault knowledge corresponding to the to-be-processed device is pushed, otherwise, it is inquired whether the product family-level database contains the product family-level fault knowledge corresponding to the to-be-processed device; if the product family-level database contains the product family-level fault knowledge corresponding to the to-be-processed device, the product family-level fault knowledge corresponding to the to-be-processed device is pushed, otherwise, it is inquired whether the product system-level database contains the system-level fault knowledge corresponding to the to-be-processed device; if the product system-level database contains the system-level fault knowledge corresponding to the to-be-processed device, the system-level fault knowledge corresponding to the to-be-processed device is pushed, otherwise, the fault state information and the parameter abnormal state information of the to-be-processed device are sent to the maintenance personnel.

[0153] On the basis of real-time data acquisition and evaluation and digital twin model simulation analysis, a hierarchical product knowledge recommendation system is established, and real-time and accurate fault identification and different-level fault maintenance scheme recommendations are provided for the operation and maintenance process of the equipment, so that the maintenance efficiency and quality of the equipment are improved.

[0154] In the above embodiment, a multi-level product network is established, real-time sharing of equipment data and information is achieved, different-level equipment fault knowledge is mined based on digital twin technology and intelligent algorithms, and real-time fault identification and related fault solution schemes are pushed based on the digital twin model, so that the accuracy of fault identification and knowledge recommendation is improved, and the real-time maintenance capability of the equipment is increased.

[0155] In some embodiments, the fault knowledge recommendation unit 730 is further configured to push the different-level fault knowledge of the equipment in the manner of AR or VR.

[0156] In this embodiment, the product fault solution and related guidance information are pushed to the on-site product maintenance personnel in real time through AR and VR means, so that immersive operation and maintenance guidance in virtual-real interactive mode is realized.

[0157] In some embodiments of the present disclosure, as shown in Figure 8 The system further includes a digital twin model training unit 810 configured to obtain a first key performance indicator affecting the operation of each device according to the operation data of each device; input the preprocessed operation data of each device into a digital twin model, simulate the operation process of each device, and obtain a simulated second key performance indicator; and train and optimize the digital twin model according to the comparison result of the first key performance indicator and the second key performance indicator.

[0158] In this embodiment, the actual running real state of the device is compared with the simulation result, and the digital twin model is constantly updated, so that the digital twin model can accurately map the running state of the device.

[0159] Figure 9 A structural schematic diagram of another embodiment of the device fault knowledge recommendation system of the present disclosure. The system includes a memory 910 and a processor 920. Among them: the memory 910 can be a disk, a flash memory or any other non-volatile storage medium. The memory 910 is used to store Figures 1-9 instructions in the corresponding embodiment. The processor 920 is coupled to the memory 910 and can be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 920 is used to execute the instructions stored in the memory.

[0160] In another embodiment of the present disclosure, the processor 920 is coupled to the memory 910 through the BUS bus 930. The device 900 can also be connected to an external storage device 950 through a storage interface 940 to call external data, and can also be connected to a network or another computer system (not shown) through a network interface 960, which will not be described in detail here.

[0161] In this embodiment, the data instructions are stored in the memory, and the above instructions are processed by the processor, which can improve the accuracy of fault identification and knowledge recommendation.

[0162] In another embodiment, a computer readable storage medium has computer program instructions stored thereon, which are executed by a processor to implement Figures 1-6 the steps of the method in the corresponding embodiment. Those skilled in the art should understand that the embodiments of the present disclosure can be provided as a method, an apparatus, or a computer program product. Therefore, the present disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure can take the form of a computer program product implemented on one or more computer usable non-transitory storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0163] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0164] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0165] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0166] Thus far, the present disclosure has been described in detail. In order to avoid obscuring the concept of the present disclosure, some details well-known in the art are not described. Those skilled in the art can fully understand how to implement the technical solutions disclosed herein according to the above description.

[0167] Although some specific embodiments of the present disclosure have been described in detail by way of examples, those skilled in the art should understand that the above examples are only for illustration, and are not intended to limit the scope of the present disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A device failure knowledge recommendation method, comprising: establishing a product network comprising a plurality of levels, including: establishing a product node for each device, and obtaining operation data of each device, establishing a product-level database capable of storing data and failure knowledge of each product node, performing cluster analysis on a plurality of devices to obtain a plurality of product families, establishing a product family node for each product family, and establishing a connection relationship between each product node and the product family node to which it belongs, establishing a product family-level database capable of storing data and product-level failure knowledge of each product family node according to the connection relationship between each product node and the product family node to which it belongs, establishing a product system node, and establishing a connection relationship between the product system node and each product family node, and establishing a product system-level database capable of storing product system-level failure knowledge according to the connection relationship between the product system node and each product family node; analyzing the operation data of each device based on an artificial intelligence algorithm to obtain failure knowledge of each device; based on the operation data of each device, simulating the operating state of each device using a digital twin model to obtain the failure state of each device; verifying the failure knowledge of each device based on the failure state of each device; integrating and fusing the failure knowledge of each device according to the product network to obtain failure knowledge at different levels, and storing the failure knowledge at different levels in the corresponding level database; and Based on the digital twin model, a fault state corresponding to a real-time running parameter of a to-be-processed device is identified, and different levels of fault knowledge of the to-be-processed device obtained by querying in a database is recommended, wherein a matching process of the querying comprises: , represents product model information, represents the model of the device, represents a digital twin simulation result, represents a fault that occurs, represents a set of key running parameters that appear to be abnormal, represents a key abnormal parameter of a product family device, represents true.

2. The device failure knowledge recommendation method of claim 1, further comprising: obtaining a first key performance indicator affecting device operation according to the operation data of each device; inputting the preprocessed operation data of each device into a digital twin model to simulate the operation process of each device and obtain a simulated second key performance indicator; and training and optimizing the digital twin model according to the comparison result of the first key performance indicator and the second key performance indicator.

3. The device failure knowledge recommendation method according to claim 1, wherein, Based on the digital twin model, identifying the failure of the to-be-processed device corresponding to the real-time operation parameters, and recommending the different levels of failure knowledge of the to-be-processed device obtained by querying the database, comprising: determining that the to-be-processed device is in a parameter abnormal state when the real-time operation parameters of the to-be-processed device exceed the parameter threshold range; inputting the key operation parameters of the to-be-processed device into the digital twin model to simulate the operating state of the to-be-processed device; and if it is determined that the to-be-processed device fails according to the simulation result, generating a failure knowledge query request according to the failure state information and the parameter abnormal state information of the to-be-processed device, so as to query the different levels of failure knowledge of the to-be-processed device according to the failure knowledge query request.

4. The device failure knowledge recommendation method according to claim 3, wherein, Querying the different levels of failure knowledge of the to-be-processed device comprises: querying whether the product-level database contains the failure knowledge corresponding to the to-be-processed device; if the product-level database contains the failure knowledge corresponding to the to-be-processed device, pushing the failure knowledge corresponding to the to-be-processed device, otherwise, querying whether the product family-level database contains the product family-level failure knowledge corresponding to the to-be-processed device; If the product family level database contains the product family level fault knowledge corresponding to the to-be-processed device, the product family level fault knowledge corresponding to the to-be-processed device is pushed, otherwise, it is queried whether the system level fault knowledge corresponding to the to-be-processed device is contained in the product system level database; and If the product system level database contains the system level fault knowledge corresponding to the to-be-processed device, the system level fault knowledge corresponding to the to-be-processed device is pushed, otherwise, the fault state information and parameter abnormal state information of the to-be-processed device are sent to a maintenance personnel.

5. The device fault knowledge recommendation method according to claim 4, wherein The fault knowledge of different levels of the device is pushed in the form of augmented reality (AR) or virtual reality (VR).

6. The device fault knowledge recommendation method according to any one of claims 1 to 5, wherein The fault knowledge includes fault information and corresponding fault maintenance information.

7. A device fault knowledge recommendation system, comprising: a product network establishing unit configured to establish a product network comprising a plurality of levels, including: establishing a product node for each device, and obtaining operation data of each device, establishing a product level database capable of storing data and fault knowledge of each product node, performing cluster analysis on a plurality of devices to obtain a plurality of product families, establishing a product family node for each product family, and establishing a connection relationship between each product node and the product family node to which it belongs, establishing a product family level database capable of storing data and product level fault knowledge of each product family node according to the connection relationship between each product node and the product family node to which it belongs, establishing a product system node, and establishing a connection relationship between the product system node and each product family node, and establishing a product system level database capable of storing product system level fault knowledge according to the connection relationship between the product system node and each product family node; a fault knowledge obtaining unit configured to analyze the operation data of each device based on an artificial intelligence algorithm, obtain fault knowledge of each device, simulate the operation state of each device based on the operation data of each device using a digital twin model, obtain the fault state of each device, verify the fault knowledge of each device based on the fault state of each device, integrate and fuse the fault knowledge of each device according to the product network, obtain fault knowledge of different levels, and store the fault knowledge of different levels in the corresponding level database; and The fault knowledge recommendation unit is configured to identify a fault state corresponding to a real-time operation parameter of a to-be-processed device based on a digital twin model, and recommend different levels of fault knowledge of the to-be-processed device obtained by querying in a database, wherein the matching process of the querying comprises: , represents product model information, represents the model of the device, represents a digital twin simulation result, represents a fault that occurs, represents a set of key operation parameters that appear to be abnormal, represents a key abnormal parameter of a product family device operation, represents true.

8. A device fault knowledge recommendation system, comprising: a memory; and a processor coupled to the memory, the processor being configured to execute the device fault knowledge recommendation method according to any one of claims 1 to 6 based on instructions stored in the memory.

9. A computer readable storage medium having computer program instructions stored thereon, the instructions being executed by a processor to implement the device fault knowledge recommendation method according to any one of claims 1 to 6.

Images