Method and device for determining operation and maintenance efficiency loss information of unit, equipment and medium

Abstract

The invention relates to a method, a device and equipment for determining operation and maintenance efficiency loss information of a unit and a medium. In response to the loss information question request, obtaining a to-be-processed named entity; target named entities matched with the to-be-processed named entities and target association relationships among different target named entities are obtained from a pre-constructed loss information knowledge graph, and nodes of the loss information knowledge graph are associated with the named entities and the association relationships thereof; and determining operation and maintenance efficiency loss information of the target unit according to identification information and loss information of the target unit corresponding to different target named entities and semantic information between different target named entities represented by the target association relationship. Therefore, the named entities and the association relationship thereof are integrated by utilizing the knowledge graph, and the operation and maintenance efficiency loss information of the unit is accurately and efficiently determined in a named entity matching mode, so that the wind power industry is promoted to develop towards the low-cost and high-benefit direction.

Description

Technical Field

[0001] The present disclosure relates to the technical field of wind farm operation and maintenance, and in particular to a method, device, equipment and medium for determining operation and maintenance efficiency loss information of a unit. Background Art

[0002] Tracing the operation and maintenance efficiency loss information of the units in the wind farm provides an effective means to evaluate the operation level of the wind farm, so as to promote the production and operation activities of the wind power industry.

[0003] The operation and maintenance efficiency loss information of the units in the wind farm mainly comes from many aspects such as the decline in the power generation performance of the units, the fault response time process, and the omissions in operation and maintenance work. However, the operation and maintenance efficiency loss information of the units currently has many problems such as wide sources, diverse structures, and scattered storage, and has not been effectively integrated, which has brought great obstacles to tracing the operation and maintenance efficiency loss information of the units. Therefore, it is necessary to provide a method for determining the operation and maintenance efficiency loss data of the units to efficiently and accurately trace the operation and maintenance efficiency loss information of the wind turbines. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a method, device, equipment and medium for determining operation and maintenance efficiency loss information of a unit.

[0005] In a first aspect, the present disclosure provides a method for determining operation and maintenance efficiency loss information of a unit, comprising:

[0006] In response to a loss information query request, obtaining a to-be-processed named entity;

[0007] From a pre-constructed loss information knowledge graph, a target named entity matching the to-be-processed named entity and a target association relationship between different target named entities are obtained, wherein the nodes of the pre-constructed loss information knowledge graph are associated with named entities of historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is an association relationship between different named entities;

[0008] The operation and maintenance efficiency loss information of the target unit is determined according to the identification information and loss information of the target units corresponding to the different target named entities, and the semantic information between the different target named entities represented by the target association relationship.

[0009] In a second aspect, the present disclosure provides a device for determining operation and maintenance efficiency loss information of a unit, comprising:

[0010] A first acquisition module, configured to acquire a to-be-processed named entity in response to a loss information inquiry request;

[0011] The second acquisition module is used to obtain the target named entity matching the to-be-processed named entity and the target association relationship between different target named entities from the pre-constructed loss information knowledge graph, wherein the nodes of the pre-constructed loss information knowledge graph are associated with the named entities of the historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is the association relationship between different named entities;

[0012] A determination module is used to determine the operation and maintenance efficiency loss information of the target unit according to the identification information and loss information of the target units corresponding to the different target named entities, and the semantic information between the different target named entities represented by the target association relationship.

[0013] In a third aspect, an embodiment of the present disclosure further provides an electronic device, the device comprising:

[0014] one or more processors;

[0015] a storage device for storing one or more programs,

[0016] When one or more programs are executed by one or more processors, the one or more processors implement the method provided in the first aspect.

[0017] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method provided in the first aspect is implemented.

[0018] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0019] A method, device, equipment and medium for determining the operation and maintenance efficiency loss information of a unit according to an embodiment of the present disclosure include: obtaining a to-be-processed named entity in response to a loss information question request; obtaining a target named entity matching the to-be-processed named entity and a target association relationship between different target named entities from a pre-constructed loss information knowledge graph, wherein the nodes of the pre-constructed loss information knowledge graph are associated with named entities of historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is an association relationship between different named entities; determining the operation and maintenance efficiency loss information of the target unit according to the identification information and loss information of the target units corresponding to different target named entities, and the semantic information between different target named entities represented by the target association relationship. Thus, the historical operation and maintenance efficiency loss data and their association relationships with a wide range of sources, diverse structures and dispersed storage are integrated through the loss information knowledge graph, and the operation and maintenance efficiency loss information of the unit is accurately and efficiently determined by matching the named entities in the loss information question request with the named entities in the loss information knowledge graph, which is conducive to promoting the development of the wind power industry in the direction of low cost and high efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0022] Figure 1 A flow chart of a method for determining operation and maintenance efficiency loss information of a unit provided in an embodiment of the present disclosure;

[0023] Figure 2 A schematic diagram of a process for generating a loss information knowledge graph provided in an embodiment of the present disclosure;

[0024] Figure 3 A schematic diagram of the structure of a named entity extraction model provided in an embodiment of the present disclosure;

[0025] Figure 4 A schematic diagram of the structure of a device for determining operation and maintenance efficiency loss information of a unit provided in an embodiment of the present disclosure;

[0026] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0028] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0029] In order to solve the above problems, the following Figure 1-Figure 3 The method for determining the operation and maintenance efficiency loss information of the unit provided in the embodiment of the present disclosure is described. In the embodiment of the present disclosure, the method for determining the operation and maintenance efficiency loss information of the unit can be executed by an electronic device or a server. Among them, the electronic device may include a device with communication function such as a tablet computer, a desktop computer, a laptop computer, and may also include a virtual machine or a device simulated by a simulator. The server may include a single server or a server cluster. This embodiment is specifically explained with the electronic device as the execution subject.

[0030] Figure 1 A flow chart of a method for determining operation and maintenance efficiency loss information of a unit provided in an embodiment of the present disclosure is shown.

[0031] like Figure 1 As shown, the method for determining the operation and maintenance efficiency loss information of the unit may include the following steps.

[0032] S110. In response to a loss information inquiry request, obtain a to-be-processed named entity.

[0033] In this embodiment, when the user wants to query the operation and maintenance efficiency loss information of the unit, the query function provided by the interactive interface of the electronic device is triggered so that the electronic device obtains a loss information question request. Alternatively, the user inputs a loss information question request on the interactive interface, and the electronic device parses the loss information question request to determine the named entity to be processed.

[0034] Optionally, the interactive interface can be a web interface of a wind turbine operation and maintenance efficiency loss tracing question and answer system built based on a Gradio framework, which provides controls or search boxes corresponding to the query function to obtain loss information question requests.

[0035] The named entity to be processed is the noun in the loss information query request.

[0036] Among them, the specific implementation method of S110 includes but is not limited to the following methods: extracting keywords from the loss information question request to determine the current keyword corresponding to the loss information question request; based on the part of speech corresponding to the current keyword, obtaining the named entity to be processed from the current keyword.

[0037] Specifically, the electronic device uses a string search algorithm (such as the Ahocorasick algorithm) to extract current keywords from the loss question request, and then extracts target keywords whose part of speech is a noun from the current keywords as named entities to be processed.

[0038] For example, if the loss information query request is "What is the power generation consumed by Unit 1 in July", the current keywords include "Unit 1", "July", "consumption", and "power generation", and the named entities to be processed include "Unit 1", "July", and "power generation".

[0039] In this way, by performing keyword extraction and part-of-speech analysis on the loss information question request, the named entities to be processed are accurately obtained from the loss information question request.

[0040] S120. Obtain target named entities that match the named entities to be processed and target association relationships between different target named entities from a pre-constructed loss information knowledge graph, wherein named entities of historical operation and maintenance performance loss information of multiple units are associated with the nodes of the pre-constructed loss information knowledge graph, and the connection relationships between different nodes of the pre-constructed loss information knowledge graph are association relationships between different named entities.

[0041] It can be understood that the loss information knowledge graph integrates historical operation and maintenance efficiency loss data and their associations that come from a wide range of sources, have diverse structures, and are stored in a dispersed manner. The named entities to be processed are matched in sequence with the named entities of the historical operation and maintenance efficiency loss information associated with the nodes of the loss information knowledge graph to obtain the target named entities that match the named entities to be processed and the target associations between different target named entities.

[0042] Among them, historical operation and maintenance performance loss information includes a large amount of unstructured data and structured data.

[0043] The named entity of the historical operation and maintenance performance loss information refers to the noun in the historical operation and maintenance performance loss information.

[0044] The association relationship between different named entities can be determined based on the semantic information between the named entities.

[0045] In this way, by uniformly aggregating the historical operation and maintenance performance loss data into the loss information knowledge graph, and utilizing the advantages of the knowledge graph such as high flexibility, good compatibility and strong scalability, it is convenient to quickly determine the historical operation and maintenance performance loss information, while also eliminating erroneous and redundant information.

[0046] S130, determining the operation and maintenance efficiency loss information of the target unit according to the identification information and loss information of the target units corresponding to different target named entities, and the semantic information between different target named entities represented by the target association relationship.

[0047] In this embodiment, the electronic device obtains identification information and loss information of the target unit from different target named entities, and obtains semantic information between different target named entities, and then generates operation and maintenance efficiency loss data of the target unit based on the semantic information, identification information and loss information.

[0048] The identification information of the target unit includes the name or code of the target named entity. The loss information includes the loss time and loss amount.

[0049] For example, if the target named entities include "Unit 1", "July", and "Generation Power", the identification information of the target unit is "Unit 1", and the loss information includes "July" and "Generation Power".

[0050] Among them, the specific implementation method of S130 includes but is not limited to the following methods: based on semantic information, a sentence is composed of identification information and loss information of the target unit; and the sentence meaning information represented by the sentence is used as the operation and maintenance efficiency loss information of the target unit.

[0051] Specifically, based on semantic information, the identification information and loss information of the target unit are connected to form a sentence, and the sentence meaning information represented by the sentence is directly used as the operation and maintenance efficiency loss information of the target unit.

[0052] In this way, after determining the target named entity and target association relationship of the target unit, the statement is reversely constructed through semantic analysis, so as to accurately and quickly determine the operation and maintenance efficiency loss information of the target unit.

[0053] A method for determining the operation and maintenance efficiency loss information of a unit in an embodiment of the present disclosure includes: obtaining a to-be-processed named entity in response to a loss information question request; obtaining a target named entity matching the to-be-processed named entity and a target association relationship between different target named entities from a pre-constructed loss information knowledge graph, wherein the nodes of the pre-constructed loss information knowledge graph are associated with named entities of historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is an association relationship between different named entities; determining the operation and maintenance efficiency loss information of the target unit according to the identification information and loss information of the target units corresponding to different target named entities, and the semantic information between different target named entities represented by the target association relationship. Thus, the historical operation and maintenance efficiency loss data and their association relationships with a wide range of sources, diverse structures and dispersed storage are integrated through the loss information knowledge graph, and the operation and maintenance efficiency loss information of the unit is accurately and efficiently determined by matching the named entities in the loss information question request with the named entities in the loss information knowledge graph, which is conducive to promoting the development of the wind power industry in the direction of low cost and high efficiency.

[0054] In another embodiment of the present disclosure, the method of loss information knowledge graph is explained in detail.

[0055] Figure 2 A flow chart of a method for generating a loss information knowledge graph provided in an embodiment of the present disclosure is shown.

[0056] like Figure 2 As shown, the method for generating the loss information knowledge graph may include the following steps.

[0057] S210. Obtain historical operation and maintenance efficiency loss data of multiple units.

[0058] In this embodiment, the electronic device obtains historical operation and maintenance efficiency loss data of multiple units including the target unit and other units, so as to integrate a large amount of structured data and unstructured data.

[0059] Among them, the historical operation and maintenance efficiency loss data refers to the actual operation and maintenance efficiency loss data of the unit in the past period of time.

[0060] S220: Use a pre-trained named entity extraction model to process the historical operation and maintenance efficiency loss data to obtain named entities of the historical operation and maintenance efficiency loss data.

[0061] In this embodiment, in order to improve the extraction accuracy of named entities in historical operation and maintenance performance loss data, the electronic device calls a pre-trained named entity extraction model to extract named entities from the historical operation and maintenance performance loss data.

[0062] The named entity extraction model is obtained by training the initial model using training sample pairs. Specifically, the training sample pairs include training operation and maintenance efficiency loss data and classification labels of the training operation and maintenance efficiency loss data. The training operation and maintenance efficiency loss data in the training samples are input into the initial model to obtain predicted named entities. The predicted named entities and classification labels are then used to iteratively train the initial model until the model meets the iterative training conditions to obtain the named entity extraction model.

[0063] The classification labels can be obtained by sequentially labeling the named entities in the training operation and maintenance efficiency loss data using the BIOES labeling method. Specifically, the BIOES labeling method labels the beginning named entities in the training operation and maintenance efficiency loss data using B-begin, labels the middle named entities in the training operation and maintenance efficiency loss data using I-inside, labels the end named entities in the training operation and maintenance efficiency loss data using O-outside, and labels the single character entities in the training operation and maintenance efficiency loss data using S-single.

[0064] Optionally, the named entity extraction model includes but is not limited to one of an entity and relationship joint extraction model based on a RoBERTa pre-trained deep learning model, a bidirectional long short-term memory network, and a conditional random field (RoBERTa-BiLSTM-CRF).

[0065] The specific implementation method of S220 includes but is not limited to the following methods:

[0066] S1. Based on the word vector extraction network in the pre-trained named entity extraction model, extract a first word vector sequence and a second word vector sequence from the historical operation and maintenance performance loss data, wherein the first word vector sequence and the second word vector sequence contain the same word vectors in reverse arrangement order;

[0067] S2. Based on the context extraction network in the pre-trained named entity extraction model, the first word vector sequence and the second word vector sequence are processed to obtain the context features of the historical operation and maintenance efficiency loss data;

[0068] S3. Based on the named entity prediction network in the pre-trained named entity extraction model, the context features are processed to obtain the named entities of the historical operation and maintenance performance loss data.

[0069] Among them, the word vector extraction network is specifically a pre-trained network, such as a RoBERTa network.

[0070] Among them, the specific implementation method of S1 includes but is not limited to the following methods: based on the word embedding network in the word vector extraction network, the historical operation and maintenance efficiency loss data is processed to obtain the word embedding features of the historical operation and maintenance efficiency loss data; based on the segment embedding network in the word vector extraction network, the word embedding features are processed to obtain the segment embedding features of the historical operation and maintenance efficiency loss data; based on the position embedding network in the word vector extraction network, the segment embedding features are processed to obtain the first word vector sequence and the second word vector sequence.

[0071] Among them, the word embedding network can be Token Embeddings, which is used to extract the vector representation of the word itself as the word embedding feature. The segment embedding network can be Segment Embeddings, which is used to encode the position information of the word into a feature vector to obtain the segment embedding feature. The position embedding network can be Position Embeddings, which uses the vector representation to distinguish two sentences to obtain the first word vector sequence and the second word vector sequence.

[0072] The first word vector sequence and the second word vector sequence include the same word vectors but in opposite order. For example, if the first word vector sequence is a forward word vector sequence, the second word vector sequence is a reverse word vector sequence; conversely, if the first word vector sequence is a reverse word vector sequence, the second word vector sequence is a forward word vector sequence.

[0073] Among them, the context extraction network is specifically a recurrent network, such as a bidirectional long short-term memory network (Bidirectional Long Short-Term Memory, BiLSTM).

[0074] Among them, the specific implementation method of S2 includes but is not limited to the following methods: based on the word vector processing network in the context extraction network, the first word vector sequence and the second word vector sequence are processed to obtain the word vector bidirectional dependency feature; based on the linear layer in the context extraction network, the word vector bidirectional dependency feature is processed to obtain the linear feature of the word vector bidirectional dependency feature; based on the output layer in the context extraction network, the linear feature of the word vector bidirectional dependency feature is processed to obtain the contextual feature of the historical operation and maintenance efficiency loss data.

[0075] The word vector processing network may include an input layer and a hidden layer, which are used to capture both the forward and reverse information of the data and obtain the bidirectional dependency features of the word vector. The linear layer is used to map the bidirectional dependency features of the word vector to a specific interval to improve the fitting ability of the model and the accuracy of the context features.

[0076] Optionally, the named entity prediction network is specifically a probabilistic model, such as a CRF network.

[0077] For ease of understanding, Figure 3 Figure 2 shows a schematic diagram of the structure of the named entity extraction model. Figure 3 First, based on the RoBERTa network in the named entity extraction model, the first word vector sequence and the second word vector sequence are extracted from the historical operation and maintenance performance loss data. The first word vector sequence includes: The second word vector sequence includes: Then, the first word vector sequence and the second word vector sequence are processed based on the BiLSTM network in the named entity extraction model to obtain the context features of the historical operation and maintenance efficiency loss data. The context features include: h1, h2, ..., h n-1 、h n Then, based on the CRF network in the named entity extraction model, the context features are processed to obtain the named entities of the historical operation and maintenance performance loss data, which include C1, C2, ..., C n-1 , C n .

[0078] S230: Determine the association relationship between different named entities based on the semantic information between different named entities in the historical operation and maintenance performance loss data.

[0079] Since the semantic information represents whether there is an association relationship between different named entities, the electronic device obtains the semantic information from the historical operation and maintenance performance loss data to obtain the named entities with the association relationship.

[0080] Through the above method, the named entity extraction model can be used to accurately determine the named entities of the historical operation and maintenance performance loss data, and based on the semantic information between different named entities, the association relationship between different named entities can be quickly determined.

[0081] S240. Generate a loss information knowledge graph using the named entities of the historical operation and maintenance performance loss data as nodes and the association relationships between different named entities as the connection relationships between different nodes.

[0082] In this embodiment, after determining the named entities and the association relationships between different named entities, the named entities of the historical operation and maintenance performance loss data are used as nodes of the initial knowledge graph, and the association relationships between different named entities are used as connection relationships between different nodes on the initial knowledge graph, thereby generating a loss information knowledge graph.

[0083] Therefore, we first use the named entity extraction model to determine the named entities of historical operation and maintenance efficiency loss data, and then accurately construct a loss information knowledge graph by performing semantic analysis on different named entities, thereby improving the reliability of integrating historical operation and maintenance efficiency loss data and their correlation relationships with wide sources, diverse structures and dispersed storage.

[0084] The disclosed embodiment also provides a device for determining the operation and maintenance efficiency loss information of a unit for implementing the above-mentioned method for determining the operation and maintenance efficiency loss information of the unit, and the device is configured in an electronic device or a server. The electronic device may include a tablet computer, a desktop computer, a laptop computer, and other devices with communication functions, and may also include a virtual machine or a device simulated by a simulator. The server may include a single server or a server cluster. This embodiment is specifically explained using an electronic device as a device for determining the operation and maintenance efficiency loss information of a unit. The following is combined with Figure 4 Provide explanation.

[0085] Figure 4 A schematic diagram of the structure of a device for determining operation and maintenance efficiency loss information of a unit provided in an embodiment of the present disclosure is shown.

[0086] like Figure 4 As shown, the unit operation and maintenance efficiency loss information determination device 400 may include:

[0087] A first acquisition module 410 is used to obtain a to-be-processed named entity in response to a loss information query request;

[0088] The second acquisition module 420 is used to obtain the target named entity matching the to-be-processed named entity and the target association relationship between different target named entities from the pre-constructed loss information knowledge graph, wherein the nodes of the pre-constructed loss information knowledge graph are associated with the named entities of the historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is the association relationship between different named entities;

[0089] The determination module 430 is used to determine the operation and maintenance efficiency loss information of the target unit according to the identification information and loss information of the target units corresponding to the different target named entities, and the semantic information between the different target named entities represented by the target association relationship.

[0090] A device for determining the operation and maintenance efficiency loss information of a unit according to an embodiment of the present disclosure includes: obtaining a to-be-processed named entity in response to a loss information question request; obtaining a target named entity matching the to-be-processed named entity and a target association relationship between different target named entities from a pre-constructed loss information knowledge graph, wherein the nodes of the pre-constructed loss information knowledge graph are associated with named entities of historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is an association relationship between different named entities; determining the operation and maintenance efficiency loss information of the target unit according to the identification information and loss information of the target units corresponding to different target named entities, and the semantic information between different target named entities represented by the target association relationship. Thus, the historical operation and maintenance efficiency loss data and their association relationships with a wide range of sources, diverse structures and dispersed storage are integrated through the loss information knowledge graph, and the operation and maintenance efficiency loss information of the unit is accurately and efficiently determined by matching the named entities in the loss information question request with the named entities in the loss information knowledge graph, which is conducive to promoting the development of the wind power industry in the direction of low cost and high efficiency.

[0091] In some embodiments of the present disclosure, the first acquisition module 410 includes:

[0092] A keyword extraction unit, configured to extract keywords from the loss information question request and determine a current keyword corresponding to the loss information question request;

[0093] The first determining unit is used to obtain the to-be-processed named entity from the current keyword based on the part of speech corresponding to the current keyword.

[0094] In some embodiments of the present disclosure, the determination module 430 includes:

[0095] A sentence generating unit, configured to form a sentence based on the semantic information and the identification information of the target unit and the loss information;

[0096] The second determining unit is used to use the sentence meaning information represented by the sentence as the operation and maintenance efficiency loss information of the target unit.

[0097] In some embodiments of the present disclosure, the device further includes:

[0098] The third acquisition module is used to obtain historical operation and maintenance efficiency loss data of multiple units;

[0099] A fourth acquisition module is used to process the historical operation and maintenance efficiency loss data using a pre-trained named entity extraction model to obtain named entities of the historical operation and maintenance efficiency loss data;

[0100] An association relationship determination module, used to determine the association relationship between different named entities based on the semantic information between different named entities in the historical operation and maintenance efficiency loss data;

[0101] The knowledge graph generation module is used to generate the loss information knowledge graph using the named entities of the historical operation and maintenance performance loss data as nodes and the association relationships between different named entities as the connection relationships between different nodes.

[0102] In some embodiments of the present disclosure, the fourth acquisition module includes:

[0103] A word vector extraction unit, configured to extract a first word vector sequence and a second word vector sequence from the historical operation and maintenance performance loss data based on a word vector extraction network in the pre-trained named entity extraction model, wherein the first word vector sequence and the second word vector sequence include the same word vectors in reverse arrangement order;

[0104] A context extraction unit, configured to process the first word vector sequence and the second word vector sequence based on the context extraction network in the pre-trained named entity extraction model to obtain context features of the historical operation and maintenance efficiency loss data;

[0105] A context processing unit is used to process the context features based on a named entity prediction network in the pre-trained named entity extraction model to obtain named entities of the historical operation and maintenance performance loss data.

[0106] In some embodiments of the present disclosure, the word vector extraction unit is specifically used to:

[0107] Based on the word embedding network in the word vector extraction network, the historical operation and maintenance efficiency loss data is processed to obtain word embedding features of the historical operation and maintenance efficiency loss data;

[0108] Based on the segment embedding network in the word vector extraction network, the word embedding features are processed to obtain the segment embedding features of the historical operation and maintenance efficiency loss data;

[0109] Based on the position embedding network in the word vector extraction network, the segment embedding features are processed to obtain the first word vector sequence and the second word vector sequence.

[0110] In some embodiments of the present disclosure, the context extraction unit is specifically configured to:

[0111] Based on the word vector processing network in the context extraction network, the first word vector sequence and the second word vector sequence are processed to obtain a word vector bidirectional dependency feature;

[0112] Based on the linear layer in the context extraction network, the word vector bidirectional dependency feature is processed to obtain the linear feature of the word vector bidirectional dependency feature;

[0113] Based on the output layer in the context extraction network, the linear features of the bidirectional dependency features of the word vector are processed to obtain the context features of the historical operation and maintenance efficiency loss data.

[0114] It should be noted that Figure 4 The unit operation and maintenance efficiency loss information determination device 400 shown in the figure can be executed Figure 1-Figure 3 The various steps in the method embodiment shown in the figure are implemented Figure 1-Figure 3 The various processes and effects in the method embodiment shown are not described in detail here.

[0115] Figure 5 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown.

[0116] like Figure 5 As shown, the electronic device may include a processor 501 and a memory 502 storing computer program instructions.

[0117] Specifically, the processor 501 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0118] The memory 502 may include a large capacity memory for information or instructions. By way of example and not limitation, the memory 502 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory 502 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 502 may be inside or outside the integrated gateway device. In a particular embodiment, the memory 502 is a non-volatile solid-state memory. In a particular embodiment, the memory 502 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (Electrically Erasable Programmable ROM, EEPROM), an electrically rewritable ROM (EAROM) or a flash memory, or a combination of two or more of these.

[0119] The processor 501 reads and executes the computer program instructions stored in the memory 502 to perform the steps of the method for determining the operation and maintenance efficiency loss information of the unit provided in the embodiment of the present disclosure.

[0120] In one example, the electronic device may further include a transceiver 503 and a bus 504. Figure 5 As shown, the processor 501, the memory 502 and the transceiver 503 are connected via a bus 504 and communicate with each other.

[0121] The bus 504 includes hardware, software, or both. For example, but not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 504 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0122] The following is an embodiment of a computer-readable storage medium provided in an embodiment of the present disclosure. The computer-readable storage medium and the method for determining the operation and maintenance efficiency loss information of the unit in the above-mentioned embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the computer-readable storage medium, reference can be made to the embodiment of the method for determining the operation and maintenance efficiency loss information of the unit in the above-mentioned embodiments.

[0123] This embodiment provides a storage medium including computer executable instructions. When the computer executable instructions are executed by a computer processor, they are used to execute a method for determining operation and maintenance efficiency loss information of a unit. The method is applied to an electronic device corresponding to a mechanical arm, and an end effector of the mechanical arm is installed with an external shaft. The method includes:

[0124] In response to a loss information query request, obtaining a to-be-processed named entity;

[0125] From a pre-constructed loss information knowledge graph, a target named entity matching the to-be-processed named entity and a target association relationship between different target named entities are obtained, wherein the nodes of the pre-constructed loss information knowledge graph are associated with named entities of historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is an association relationship between different named entities;

[0126] The operation and maintenance efficiency loss information of the target unit is determined according to the identification information and loss information of the target units corresponding to the different target named entities, and the semantic information between the different target named entities represented by the target association relationship.

[0127] Of course, the storage medium containing computer executable instructions provided in the embodiment of the present disclosure is not limited to the above method operations, and the computer executable instructions can also execute related operations in the method for determining the operation and maintenance efficiency loss information of the unit provided in any embodiment of the present disclosure.

[0128] Through the above description of the implementation methods, the technical personnel in the relevant field can clearly understand that the present disclosure can be implemented with the help of software and necessary general hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present disclosure can be essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer cloud platform (which can be a personal computer, server, or network cloud platform, etc.) to execute the unit operation and maintenance efficiency loss information determination method provided in each embodiment of the present disclosure.

[0129] Note that the above are only preferred embodiments of the present disclosure and the technical principles used. Those skilled in the art will understand that the present disclosure is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present disclosure. Therefore, although the present disclosure is described in more detail through the above embodiments, the present disclosure is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present disclosure, and the scope of the present disclosure is determined by the scope of the appended claims.

Claims

1. A method for determining operation and maintenance efficiency loss information of a unit, characterized in that: include: In response to a loss information query request, obtaining a to-be-processed named entity; From a pre-constructed loss information knowledge graph, a target named entity matching the to-be-processed named entity and a target association relationship between different target named entities are obtained, wherein the nodes of the pre-constructed loss information knowledge graph are associated with named entities of historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is an association relationship between different named entities; The operation and maintenance efficiency loss information of the target unit is determined according to the identification information and loss information of the target units corresponding to the different target named entities, and the semantic information between the different target named entities represented by the target association relationship.

2. The method according to claim 1, characterized in that The step of obtaining a to-be-processed named entity in response to the loss information inquiry request includes: Extracting keywords from the loss information question request to determine current keywords corresponding to the loss information question request; Based on the part of speech corresponding to the current keyword, the to-be-processed named entity is obtained from the current keyword.

3. The method according to claim 1, characterized in that The determining the operation and maintenance efficiency loss information of the target unit according to the identification information and loss information of the target units respectively corresponding to the different target named entities and the semantic information between the different target named entities represented by the target association relationship includes: Based on the semantic information, a sentence is formed by the identification information of the target unit and the loss information; The sentence meaning information represented by the sentence is used as the operation and maintenance efficiency loss information of the target unit.

4. The method according to claim 1, characterized in that: Also includes: Obtain historical operation and maintenance efficiency loss data of multiple units; Using a pre-trained named entity extraction model, the historical operation and maintenance efficiency loss data is processed to obtain named entities of the historical operation and maintenance efficiency loss data; Determining association relationships between different named entities based on semantic information between different named entities in the historical operation and maintenance efficiency loss data; The loss information knowledge graph is generated by taking the named entities of the historical operation and maintenance performance loss data as nodes and taking the association relationships between different named entities as the connection relationships between different nodes.

5. The method according to claim 4, characterized in that The method of using a pre-trained named entity extraction model to process the historical operation and maintenance efficiency loss data to obtain named entities of the historical operation and maintenance efficiency loss data includes: Based on the word vector extraction network in the pre-trained named entity extraction model, extracting a first word vector sequence and a second word vector sequence from the historical operation and maintenance performance loss data, wherein the first word vector sequence and the second word vector sequence contain the same word vectors in reverse order; Based on the context extraction network in the pre-trained named entity extraction model, the first word vector sequence and the second word vector sequence are processed to obtain context features of the historical operation and maintenance efficiency loss data; Based on the named entity prediction network in the pre-trained named entity extraction model, the context features are processed to obtain named entities of the historical operation and maintenance performance loss data.

6. The method according to claim 5, characterized in that The word vector extraction network based on the pre-trained named entity extraction model extracts a first word vector sequence and a second word vector sequence from the historical operation and maintenance performance loss data, including: Based on the word embedding network in the word vector extraction network, the historical operation and maintenance efficiency loss data is processed to obtain word embedding features of the historical operation and maintenance efficiency loss data; Based on the segment embedding network in the word vector extraction network, the word embedding features are processed to obtain the segment embedding features of the historical operation and maintenance efficiency loss data; Based on the position embedding network in the word vector extraction network, the segment embedding features are processed to obtain the first word vector sequence and the second word vector sequence.

7. The method according to claim 5, characterized in that The context extraction network based on the pre-trained named entity extraction model processes the first word vector sequence and the second word vector sequence to obtain context features of the historical operation and maintenance efficiency loss data, including: Based on the word vector processing network in the context extraction network, the first word vector sequence and the second word vector sequence are processed to obtain a word vector bidirectional dependency feature; Based on the linear layer in the context extraction network, the word vector bidirectional dependency feature is processed to obtain the linear feature of the word vector bidirectional dependency feature; Based on the output layer in the context extraction network, the linear features of the bidirectional dependency features of the word vector are processed to obtain the context features of the historical operation and maintenance efficiency loss data.

8. A device for determining operation and maintenance efficiency loss information of a unit, characterized in that: include: A first acquisition module, configured to acquire a to-be-processed named entity in response to a loss information inquiry request; The second acquisition module is used to obtain the target named entity matching the to-be-processed named entity and the target association relationship between different target named entities from the pre-constructed loss information knowledge graph, wherein the nodes of the pre-constructed loss information knowledge graph are associated with the named entities of the historical operation and maintenance efficiency loss information of multiple units, and the connection relationship between different nodes of the pre-constructed loss information knowledge graph is the association relationship between different named entities; A determination module is used to determine the operation and maintenance efficiency loss information of the target unit according to the identification information and loss information of the target units corresponding to the different target named entities, and the semantic information between the different target named entities represented by the target association relationship.

9. An electronic device, characterized in that: include: processor; A memory for storing executable instructions; The processor is used to read the executable instructions from the memory and execute the executable instructions to implement the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the method according to any one of claims 1 to 7.

Images