A knowledge graph construction and management method, device, and storage medium

Abstract

A method, apparatus, and storage medium for constructing and managing a knowledge graph. The construction method is applied to an oilfield water gathering, transportation, and injection system. The knowledge graph includes a concept graph. The method includes: acquiring text data related to energy consumption control in the oilfield water gathering, transportation, and injection system, wherein the text data records anomaly analysis information and resolution process information; treating parameter anomalies, anomaly causes, handling measures, and at least one piece of information used to characterize the event localization process as entities in the concept graph; performing entity recognition on the text data to obtain naming information for each type of entity in the text data; obtaining entity relationships between different types of entities in the text data based on the entity naming information; and generating the concept graph based on the entity relationships in the text data.

Description

Technical Field

[0001] This article relates to the field of computer technology, and in particular to a method, apparatus and storage medium for constructing and managing knowledge graphs. Background Technology

[0002] Oilfield water gathering, transportation, and injection systems involve multiple technological processes and are large and dispersed. As oilfield development continues, single-well production declines and water cut increases. Pump units and other equipment are prone to corrosion and aging after years of use, resulting in high energy consumption, low operating efficiency, and significant energy waste in oilfield water gathering, transportation, and injection systems.

[0003] Energy management in oilfield water gathering, transportation, and injection systems largely relies on in-depth analysis of accumulated field operational data, utilizing theoretical optimization algorithms or process software, supplemented by manual analysis of existing management documents. In recent years, the rapid development of emerging information technologies such as data acquisition and sensing, and big data analytics has led to continuous optimization of oilfield energy management systems, resulting in the accumulation of massive amounts of operational data and complex management documents. Manually extracting the necessary knowledge from these vast and complex documents and coordinating it with operational data to make decisions is time-consuming and cannot meet the intelligent decision-making needs for real-time optimization and control of oilfield energy.

[0004] Therefore, proposing a processing method that automatically extracts the required knowledge from large and complex management documents, coordinates operational data to complete timely and effective decision-making, and meets the needs of real-time optimization control and intelligent decision-making in oilfield energy has become an urgent problem to be solved. Summary of the Invention

[0005] This application provides a method, apparatus, and storage medium for constructing and managing knowledge graphs.

[0006] A method for constructing a knowledge graph, applied to an oilfield water gathering, transportation, and injection system, wherein the knowledge graph includes a concept graph, and the method includes: Acquire text data on energy consumption control in the oilfield water gathering, transportation and injection system, wherein the text data records anomaly analysis information and resolution process information; The concept graph uses parameter anomalies, anomaly causes, handling measures, and at least one piece of information used to characterize the event localization process as entities. Entity recognition is performed on the text data to obtain the naming information of each type of entity in the text data. Herein, parameter anomalies represent abnormal situations in which parameters occur, anomaly causes represent the reasons that lead to parameter anomalies, and handling measures represent targeted measures taken. Based on the naming information of the entities, obtain the entity relationships between different types of entities in the text data, wherein the entity relationships in the text data include bidirectional relationships between parameter anomalies and anomaly causes, bidirectional relationships between anomaly causes and handling measures, and bidirectional relationships between parameter anomalies and at least one piece of information used to characterize the event localization process. The concept map is generated based on the entity relationships in the text data.

[0007] A knowledge graph management method, characterized in that it is applied to an oilfield water gathering, transportation, and injection system, the method comprising: Obtain the query request for the oilfield water gathering, transportation and injection system; Obtain the query results corresponding to the query request from the knowledge graph, wherein the knowledge graph is constructed using the method described above; Output the query results.

[0008] A knowledge graph construction device, applied to an oilfield water gathering, transportation, and injection system, wherein the knowledge graph includes a concept graph, and the device comprises: The first acquisition module is configured to acquire text data on energy consumption control in the oilfield water gathering and injection system, wherein the text data records anomaly analysis information and resolution process information; The entity recognition module is configured to treat parameter anomalies, anomaly causes, handling measures, and at least one piece of information used to characterize the event localization process as entities in the concept graph, and to perform entity recognition on the text data to obtain naming information for each type of entity in the text data; wherein, parameter anomalies represent abnormal situations in which parameters occur, anomaly causes represent the reasons that lead to the parameter anomalies, and handling measures represent targeted measures taken. The relationship extraction module is configured to obtain entity relationships between different types of entities in the text data based on the entity naming information. The entity relationships in the text data include bidirectional relationships between parameter anomalies and anomaly causes, bidirectional relationships between anomaly causes and handling measures, and bidirectional relationships between parameter anomalies and at least one piece of information used to characterize the event localization process. The generation module is configured to generate the concept graph based on the entity relationships in the text data.

[0009] A knowledge graph management device, applied to an oilfield water gathering, transportation, and injection system, the device comprising: The second acquisition module is set up as a query request for the knowledge graph; The processing module is configured to obtain the query results corresponding to the query request from the knowledge graph, wherein the knowledge graph is constructed using the method described above; The output module is configured to output the query results.

[0010] A storage medium having a computer program stored thereon, characterized in that the program, when executed by a processor, implements the steps of any of the methods described above.

[0011] In this embodiment, the process of analyzing consumption anomalies in the oilfield water gathering, transportation and injection system, namely "parameter anomaly - event location - anomaly cause - handling measures", is presented in a structured form, generating bidirectional reasoning logic, which facilitates knowledge reasoning using concept graphs.

[0012] By utilizing knowledge graphs to respond to query requests from oilfield water gathering, transportation, and injection systems, timely, comprehensive, and intelligent decision recommendations can be made, thereby improving the energy management level of oilfield water gathering, transportation, and injection systems and providing reference opinions for energy management decisions.

[0013] Other features and advantages of this application will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the application. Other advantages of this application can be realized and obtained by means of the solutions described in the description and the accompanying drawings. Attached Figure Description

[0014] The accompanying drawings are used to provide an understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0015] Figure 1 A flowchart illustrating the knowledge graph construction method provided in this application embodiment; Figure 2 A schematic diagram illustrating the knowledge graph construction process provided in this application embodiment; Figure 3 This is a schematic diagram of the knowledge graph pattern layer provided in an embodiment of this application; Figure 4 A schematic diagram illustrating entity recognition using a BiGRU-CRF model provided in an embodiment of this application; Figure 5 A schematic diagram illustrating the BiGRU-ATT model relation extraction provided in this application embodiment; Figure 6 A flowchart illustrating the knowledge graph management method provided in this application embodiment; Figure 7 A flowchart illustrating the application of knowledge graphs in an embodiment of this application; Figure 8 A schematic diagram illustrating the query results for energy consumption concept relationships provided in this application embodiment; Figure 9 A schematic diagram illustrating the query results of pipeline entity relationships provided in this application embodiment; Figure 10 A schematic diagram illustrating the reasoning process of alarm information provided in the embodiments of this application; Figure 11 A schematic diagram of the structure of the knowledge graph construction apparatus provided in the embodiments of this application; Figure 12 A schematic diagram of the structure of the knowledge graph management device provided in the embodiments of this application. Detailed Implementation

[0016] The intelligent decision-making process for energy management in oilfield water gathering and injection systems faces the challenge of handling massive amounts of heterogeneous data from multiple sources. This application proposes a knowledge graph-based intelligent auxiliary decision-making scheme for energy consumption anomalies in oilfield water gathering and injection systems, providing holistic support and guidance for energy consumption control from both data and knowledge perspectives. Based on the constructed knowledge graph, a business-demand-oriented energy management knowledge graph visualization platform (hereinafter referred to as the "visualization platform") is built using Web technology to achieve functions such as attribution analysis and decision recommendation for energy consumption anomalies. Verification shows that the constructed visualization platform can provide operators with rapid, comprehensive, and accurate energy consumption control guidance in task scenarios and effectively handle energy consumption anomaly events. Using the proposed knowledge graph, intelligent auxiliary decision-making for energy consumption anomalies in oilfield water gathering and injection systems can provide timely, holistic, and intelligent decision recommendations, improving the energy management level of oilfield water gathering and injection systems and providing reference opinions for energy management decisions in oilfield water gathering and injection systems.

[0017] The construction of a knowledge graph requires a large amount of data, which originates from energy consumption management data of the oilfield's water gathering, transportation, and injection system in recent years. After several years of digital management construction, the oilfield has achieved digitization of its joint stations, transfer stations, oil gathering and injection distribution rooms, and oil and water wells, initially realizing large-scale digital oilfield construction. Digital management is adopted in the oilfield's production process, and a relatively comprehensive data foundation exists. The data comes from structured data such as pipeline topology and equipment asset data, as well as unstructured data such as energy consumption anomaly ledgers, expert experience and knowledge, and system operation and maintenance manuals; among which:

[0018] Structured data is stored in tabular form. Pipeline topology information includes pipeline name, pipeline number, pipeline code, commissioning date, originating station type, originating station name, ending station type, and ending station name, as detailed in Table 1. Equipment asset data includes pipelines, mixing pumps, mixing boilers, and injection pumps both inside and outside the stations (areas). For example, basic information for injection pumps includes the station name, pump name, station number, pump type, commissioning date, pump manufacturer, rated flow rate, rated speed, and rated operating pressure.

[0019]

[0020] Table 1 Unstructured data is text data written in natural language. The energy consumption anomaly log is natural text compiled by staff, containing information on energy consumption anomaly events and their handling processes. It typically consists of four parts: parameter anomaly, event location, cause of the anomaly, and handling measures. Parameter anomaly refers to abnormal parameter events in the oilfield's water gathering, transportation, and injection system; event location is the node where the parameter anomaly event occurred; the cause of the anomaly includes abnormal factors in the equipment itself and the influence of upstream and downstream nodes; handling refers to the targeted measures taken to mitigate the anomaly. The system operation and maintenance manual introduces the overall architecture of the production system and the anomaly event handling process.

[0021] For example, one entry in the energy consumption anomaly log is as follows: System monitoring detected a continuous increase in the inlet pressure of the Wei-1 dehydration station's inlet manifold. A review of the system operation and maintenance manual indicated possible causes including excessively high total return oil pressure between the station's oil collecting valve groups, increased liquid production, excessive water mixing in the oil collecting rings between valve groups, or pipeline pressure buildup. Upon inspection, personnel found that the pipeline flow rate remained unchanged, and the inlet pressure was already higher than the water mixing pressure, confirming a pressure buildup in the pipeline. Following operating procedures, the pipeline was flushed and descaled.

[0022] For example, one entry in the system operation and maintenance manual is as follows: The status of a ring is influenced not only by the ring itself but also by the status of the wells it governs. Once a ring alarm is triggered, the ring itself must be troubleshooted first, followed by troubleshooting the wells it governs. The ring-well relationship acquisition process requires data from three tables: the ring basic information table, the well basic information table, and the oil gathering room / station basic information table. After a series of filtering, merging, and summarizing operations, the data from these three tables establishes a one-to-many correspondence between the ring information and its subordinate well information. Finally, the ring-well relationship is output and used in the ring alarm process.

[0023] The construction of the knowledge graph based on the above data sources includes: Figure 1 This is a flowchart illustrating the method for constructing a knowledge graph provided in an embodiment of this application. Figure 1 As shown, the method is applied to an oilfield water gathering, transportation, and injection system. The knowledge graph includes a concept graph, and the method includes:

[0024] S101: Obtain text data on energy consumption control in the oilfield water gathering, transportation and injection system, wherein the text data records anomaly analysis information and resolution process information; S102: The parameter anomaly, the cause of the anomaly, the handling measures, and at least one piece of information used to characterize the event localization process are all taken as entities in the concept graph. Entity recognition is performed on the text data to obtain the naming information of each type of entity in the text data. Wherein, the parameter anomaly represents an abnormal situation in which the parameter occurs, the cause of the anomaly represents the reason that caused the parameter to become abnormal, and the handling measures represent the targeted handling measures taken. S103: Based on the naming information of the entities, obtain the entity relationships between different types of entities in the text data, wherein the entity relationships in the text data include a bidirectional relationship between parameter anomalies and anomaly causes, a bidirectional relationship between anomaly causes and handling measures, and a bidirectional relationship between parameter anomalies and at least one piece of information used to characterize the event localization process. S104: Generate the concept graph based on the entity relationships in the text data.

[0025] The method provided in this application presents the consumption anomaly analysis process in the oilfield water gathering, transportation and injection system in a structured form, which includes "parameter anomaly - event location - anomaly cause - handling measures", and generates bidirectional reasoning logic, which facilitates knowledge reasoning using concept graphs.

[0026] The method provided in the embodiments of this application is described below: In an exemplary embodiment, at least one piece of information used to characterize the event localization process includes a station, related equipment, event location, and related parameters; wherein, the station represents different types of stations in the oilfield water gathering, transportation, and injection system, the related equipment represents the main process equipment in the process flow, the event location represents the location node where the abnormal event occurs, and the related parameters are the event location or process parameters monitored by the related equipment; Among them, the bidirectional relationship between parameter anomalies and at least one piece of information used to characterize the event localization process includes the bidirectional relationship between parameter anomalies and related parameters; The bidirectional relationships between entities used to characterize the event localization process include the bidirectional relationships between relevant parameters and relevant equipment, the bidirectional relationships between relevant parameters and event location, the bidirectional relationships between event location and station, and the bidirectional relationships between event location and relevant equipment.

[0027] Using the above information, we can further describe the event location process more accurately and in detail, improve the completeness and detail of information records, and improve the accuracy of subsequent knowledge reasoning.

[0028] In an exemplary embodiment, the knowledge graph further includes an entity graph that records the pipeline topology of an oilfield water gathering, transportation and injection system; The entity map uses stations, equipment, and pipelines as entities, and the relationships between entities are used to represent connection information between facilities. The entities in the entity map include at least the stations, equipment, and pipelines involved in the entities of the conceptual map. The stations in the entity map provide configuration information for the stations in the conceptual map within the oilfield's water gathering and injection system. The equipment in the entity map provides configuration information for the relevant equipment in the conceptual map within the oilfield's water gathering and injection system. At least one of the equipment and pipelines in the entity map provides definition information for parameter anomalies in the conceptual map.

[0029] By adding entity graphs, we can not only provide support for locating the fault locations of abnormal events, but also provide information support for knowledge reasoning in concept graphs, thereby improving the accuracy of subsequent knowledge reasoning.

[0030] Figure 2 This is a schematic diagram illustrating the construction process of the knowledge graph provided in an embodiment of this application. For example... Figure 2 As shown, based on the functions of the knowledge graph described above, it can be logically divided into a schema layer and a data layer. Therefore, the knowledge graph is constructed jointly by the schema layer and the data layer. The schema layer provides rules to guide the construction of the data layer, and the data layer is an instantiation of the schema layer of the knowledge graph. The two are combined to obtain the knowledge graph of the oilfield gathering, transportation, and injection water system. Wherein:

[0031] The schema layer is the knowledge organization architecture of the knowledge graph. By analyzing the text content in the energy management domain of the oilfield water gathering and injection system, it performs operations such as entity type determination, entity relationship determination, and entity attribute determination to form the domain knowledge system corresponding to the schema layer, thus realizing the construction of the schema layer.

[0032] The data layer utilizes knowledge extraction methods to perform entity extraction, relation extraction, and attribute extraction under the guidance of the knowledge architecture of the schema layer of the knowledge graph, thereby obtaining all the information of the entities and realizing the construction of the data layer.

[0033] The following explains the knowledge graph schema layer: Figure 3 This is a schematic diagram of a knowledge graph pattern layer provided in an embodiment of this application. Figure 3 As shown, considering the data characteristics of energy management in the oilfield water gathering, transportation, and injection system and the business logic of operators under abnormal energy consumption conditions, the knowledge graph is divided into a concept graph and an entity graph. The concept graph records at least the information of abnormal energy consumption events, energy consumption influencing factors, and abnormality prevention and control measures; the entity graph records at least the information of the water gathering, transportation, and injection network, related equipment, and node process parameters.

[0034] The concept graph is the core content of the energy management knowledge graph. It is constructed from unstructured text such as energy consumption management ledgers and some structured data (e.g., equipment asset data). The concept graph displays the energy consumption anomaly analysis process of oilfield gathering, transportation and injection water systems in a structured form, namely "parameter anomaly - event location - anomaly cause - handling measures". It mainly includes 7 types of entities and the relationships between entities. Bidirectional relationships are set between nodes, that is, a reverse reasoning mechanism is added on the basis of "problem-location-cause-measure" reasoning to generate reasoning logic, which facilitates subsequent knowledge reasoning.

[0035] exist Figure 3 In the conceptual diagram shown, the seven entities are station, location, equipment, parameter, variation, reason, and measure. The term "station" refers to different types of stations in the oilfield's water gathering, transportation, and injection system, such as the Wei-2 oil transfer station and the oil gathering valve assembly room. Event location indicates the location node where an abnormal event may occur, such as the oil collection manifold between oil collection valve groups, the water mixing manifold between oil collection valve groups, etc. Related equipment refers to the main process equipment in the process flow, such as water mixing pumps and water mixing furnaces; Relevant parameters indicate the location of the event or the process parameters monitored by the main equipment, such as total return oil pressure and water-mixed outlet temperature; Parameter anomalies indicate possible abnormalities in the parameters, such as a decrease in total water injection pressure or a low temperature of the ring return oil. The cause of the abnormality indicates the reason that may cause a certain parameter to be abnormal. For example, the reason for the increase in total return oil pressure may be the increase in production volume, excessive water mixing in the oil collecting ring, reduction in diameter of the inter-station return oil pipeline, or poor insulation effect of the inter-station return oil pipeline. The treatment measures indicate that targeted measures need to be taken, such as measures to alleviate the narrowing of pipelines between stations, including pipeline flushing, descaling, or replacement.

[0036] exist Figure 3 In the concept map shown, the entity relationships are as follows: In the bidirectional relationship between parameter anomalies and their causes, one is that parameter anomalies are the cause of the cause of the anomalies, and the other is that the cause of the anomalies leads to parameter anomalies. The bidirectional relationship between the cause of an anomaly and the handling measures includes one aspect of recommending handling measures based on the cause of the anomaly, and the other aspect of inferring the cause of the anomaly that can be resolved based on the handling measures. The bidirectional relationship between parameter anomalies and related parameters involves two aspects: one is that the energy consumption anomaly information corresponding to the parameter anomaly causes other parameters to become anomaly, thus obtaining related parameters; the other is that the corresponding parameter anomaly is determined based on the change of related parameters. The bidirectional relationship between event location and related parameters involves locating the event location based on the related parameters and monitoring the related parameters corresponding to that event location. The bidirectional relationship between relevant equipment and relevant parameters involves two aspects: one is locating relevant equipment based on relevant parameters, and the other is monitoring the relevant parameters corresponding to the relevant equipment. The bidirectional relationship between event location and related devices involves two parts: one is to obtain the event location based on the location associated with the related devices, and the other is to obtain the related devices based on the location associated with the event. The two-way relationship between event location and site involves one aspect: determining which site the event location belongs to, and the other aspect: determining which event locations a site includes.

[0037] For example, see Table 2 for the entity information in the concept graph of a certain application scenario:

[0038] Table 2 The entity map represents the actual pipeline topology of the oilfield water gathering, transportation and injection system, providing entity querying and matching for the conceptual map.

[0039] exist Figure 3 In the entity map shown, the three entities are the station (corresponding to the station in this application), equipment, and pipeline. For example, the connection between the entities and facilities can be represented by the joint dehydration station, oil transfer station, equipment in the station, oil gathering valve group, injection and preparation room, oil production well, water injection well, and various connecting pipelines. The specific parameters of the wellhead and pipeline to which the oil gathering ring belongs, as well as the specific parameters of the equipment, are stored in the form of entity attributes.

[0040] In addition, more detailed pipeline topology information about a site can be obtained by querying and matching the site (or space) in the entity map based on the site in the conceptual map; similarly, pipeline topology information related to a device can be obtained by querying and matching the related device in the entity map based on the related device in the conceptual map.

[0041] In addition, the concepts of technical terms involved in parameter anomalies in the conceptual map can be defined based on the attribute information of equipment and pipelines in the physical map, thereby describing energy consumption anomaly information in more detail and accurately.

[0042] The following explains the data layer of the knowledge graph: Different data processing methods are required for structured and unstructured data types. For structured data such as pipeline topology in energy management data related to oilfield water gathering and injection systems, entity relationship triples and entity attributes can be directly extracted. The extracted pipeline entities include 15 types such as junction / transfer stations and locations within stations (intervals), as detailed in Table 3.

[0043]

[0044] Table 3 For unstructured text-based data such as energy consumption anomaly ledgers, named entity recognition (NER) and relation extraction (RE) are used to transform them into structured data. Named entity recognition (NER) and relation extraction (RE) are important parts of knowledge extraction tasks, aiming to extract entity-relation triples from semi-structured or unstructured text data in the water collection and transportation domain.

[0045] Text related to energy management in oilfield gathering, transportation, and water injection systems is characterized by numerous proper nouns, complex knowledge structures, and entities often consisting of nested nouns such as "site," "facility," and "parameter changes," resulting in blurred entity boundaries. Currently used Chinese word segmentation tools, suitable for general applications, increase segmentation errors in the water injection field, impacting subsequent extraction of entity relationships.

[0046] In this embodiment, a Bidirectional Gated Recurrent Unit-Conditional Random Field (BiGRU-CRF) model based on word vector representation is used to identify entities in the field of oilfield water gathering, transportation, and injection. Building upon entity recognition, a Bidirectional Gated Recurrent Unit-Attention (BiGRU-ATT) model based on an attention mechanism is employed to extract relationships between entities. By combining the advantage of BiGRU in considering the contextual semantic information of characters with the characteristics of CRF in ensuring the reasonableness of output results and the characteristics of the Attention mechanism in improving model attention, the extraction effect of textual knowledge in the energy management field of oilfield water gathering, transportation, and injection systems is improved.

[0047] (1) Entity recognition based on BiGRU-CRF Named entity recognition is essentially a sequence labeling problem, which involves assigning a label to each character in a given text. This application adopts the BIO (B-begin, I-inside, O-outside, not belonging to any entity) sequence labeling method. The BIO labeling rules stipulate that an entity must begin with B, end with I, and O indicates that it does not belong to any entity.

[0048] To address the challenges of ambiguous entity boundaries and difficult word segmentation in entity recognition tasks within the energy management field of oilfield water gathering and injection systems, this paper employs a single character as input to the BiGRU-CRF model for entity extraction, avoiding the impact of word segmentation errors. Chinese characters are processed into vector representations using a character-to-vector conversion model. These vectors are then input into the BiGRU model for deep feature extraction, yielding a preliminary label sequence. The BiGRU model can simultaneously process the contextual information of the predicted words, improving the model's preservation of textual features and enhancing the accuracy of prediction results, while also having lower computational cost and better convergence. The preliminary output labels need further refinement through a CRF layer. The CRF layer constrains the output entity labels by defining feature functions, addressing the issue of potentially invalid labels output by BiGRU due to its lack of consideration for mutual constraints between output labels, thus improving the accuracy of the output results.

[0049] Figure 4 This is a schematic diagram illustrating entity recognition using a BiGRU-CRF model provided in an embodiment of this application. Figure 4 As shown, the BiGRU-CRF model has an embedding layer, a BiGRU layer, and a CRF layer; the input parameters are... Figure 4 The word sequence in the middle, the output parameter is Figure 4 The output sequence in the code, where the input parameters are processed on a per-character basis.

[0050] When the character sequence is "if the water content is too small", the embedding layer processes the six Chinese characters separately, obtaining six character vectors x1, x2, x3, x4, x5, and x6. The BiGRU layer further processes the character vectors output by the embedding layer, obtaining the preliminary label sequence h1, h2, h3, h4, h5, and h6. The CRF layer further refines the preliminary label sequence output by the BiGRU, obtaining the entity labels y1, y2, y3, y4, y5, and y6. The output sequence is O, B-PRA, B-PRA, I-PRA, I-PRA, O, O. Therefore, the entity recognition result is "water content".

[0051] (2) Relation extraction based on BiGRU-ATT The extraction of entity relationships is based on named entity recognition and relies heavily on contextual information. Its purpose is to extract entity relationship triples from unstructured text. This application introduces an attention mechanism into the relationship extraction task based on the BiGRU model, so as to fully learn the textual context information.

[0052] Attention mechanisms are deep learning models that mimic the human brain's attention, improving feature extraction quality by giving more attention to certain keywords in the text. The BiGRU-ATT model, based on text data and entity recognition results from energy consumption control in oilfield water supply and transportation systems, leverages the BiGRU model's ability to simultaneously perform forward and backward propagation to acquire contextual information from the text data. Using an attention mechanism, it assigns different weights to the states corresponding to each piece of information in the input corpus, summing the weights to obtain the new state and updating the output sequence. This improves the causal relationship extraction effect for abnormal energy consumption events in oilfield water supply and transportation systems. Based on the output sequence, it determines whether a predefined relationship exists between entities, extracting the relationship and obtaining relation triples, where the predefined relationship is... Figure 3 The relationships between entities in the concept map shown.

[0053] Figure 5 This is a schematic diagram illustrating the relation extraction of the BiGRU-ATT model provided in an embodiment of this application. For example... Figure 5 As shown, the BiGRU-ATT model has an embedding layer, a BiGRU layer, and an Attention layer; the input parameters are... Figure 5 The word sequence in the middle, the output parameter is Figure 5 The output sequence in the code, where the input parameters are processed on a per-character basis.

[0054] In the word sequence "X1, X2, X3, X4, ..., X n When the embedding layer processes each Chinese character separately, it obtains n character vectors x1, x2, x3, x4, ..., xn. n The BiGRU layer further processes the word vectors output by the embedding layer to obtain the preliminary tag sequence h1, h2, h3, h4, ..., h n The Attention layer assigns weights to the initial label sequence output by the BiGRU, with weight values ​​a1, a2, a3, a4, ..., a n， Then perform a weighted calculation to obtain the sequence T, where n is a positive integer.

[0055] For example, the content of text data 1 is as follows: Locations prone to abnormal events between oil collecting valve groups include the oil collecting manifold between oil collecting valve groups and the water mixing manifold of the oil collecting valve assembly.

[0056] From text data 1, two sets of entity relation triples can be extracted: one set is "oil gathering valve group - including - oil gathering manifold between oil gathering valve groups"; the other set is "oil gathering valve group - including - water mixing manifold between oil gathering valve groups".

[0057] For example, the content of text data 2 is as follows: Increased water mixing volume, reduced air intake volume, and significant temperature differences at the outlet of individual water mixing boilers can all lead to an increase in the water mixing outlet temperature. From text data 2, three sets of entity relation triples can be extracted. The first set is: increased water addition leads to a decrease in the water addition outlet temperature; the second set is: decreased air intake leads to a decrease in the water addition outlet temperature; and the third set is: large differences in outlet temperature among individual water addition boilers lead to a decrease in the water addition outlet temperature.

[0058] For example, the content of text data 3 is as follows: The location of the event is determined by the incoming water pressure at the water manifold.

[0059] From text data 3, one entity relation triple can be extracted, namely, water pressure location of water manifold.

[0060] For example, the content of text data 4 is as follows: For water injection pumps, the main parameters to monitor are the inlet manifold pressure and the outlet manifold pressure. From text data 4, two sets of entity relation triples can be extracted: one set is water injection pump - monitoring - water injection pump inlet manifold pressure; the other set is water injection pump - monitoring - water injection pump outlet manifold pressure.

[0061] In one exemplary embodiment, the generation of at least one of the concept graph and the entity graph includes: According to the preset data storage format, each relationship between entities is recorded to obtain triple data, wherein the data storage format is entity-relation-entity; A graph corresponding to the triplet data is generated using a graph database, wherein the data storage format of the graph database is node-relation-node.

[0062] The graph database in question is Neo4j, an open-source, high-performance NoSQL graph database based on Java (a computer programming language). Neo4j employs a native graph-based knowledge graph storage and management approach. Unlike traditional relational databases, graph databases visualize and store structured data in a network as "node-relation-node" triples. This unique storage structure enables it to efficiently handle complex knowledge graph queries, making it particularly suitable for representing complex, interconnected knowledge such as text in the water supply and demand domain.

[0063] Given the large amount of data involved, Py2neo (a Python library for Neo4j) + Neo4j was chosen for batch creation of nodes, relationships, and attributes during knowledge storage and graph construction. This allows for knowledge storage and retrieval from Python applications and command lines by calling the Neo4j graph database, improving the ease of use of Neo4j. A knowledge entity graph and concept graph for the energy management of the oilfield water gathering and injection system were constructed, with different colors used to distinguish different types of nodes in the graph.

[0064] Knowledge extraction is performed on domain-specific text data, and the extracted triples are stored in the Neo4j graph database to construct a knowledge graph for energy management in oilfield water gathering, transportation, and injection systems. A visualization platform is then built based on this knowledge graph and integrated into the field production management system. In oilfield energy management scenarios, the visualization platform can be used for energy consumption anomaly diagnosis.

[0065] Figure 6 This is a flowchart illustrating the knowledge graph management method provided in an embodiment of this application. Figure 6 As shown, the method is applied to an oilfield water gathering, transportation, and injection system, and the method includes:

[0066] S601: Obtain the query request for the oilfield water gathering, transportation and injection system; S602: Obtain the query result corresponding to the query request from the knowledge graph, wherein the knowledge graph is constructed using the method described above; When the query request is used to query the energy consumption concept relationship of the target site, the concept graph in the knowledge graph is searched to obtain the event location corresponding to the target site; When the query request is used to query the pipeline entity relationship of the target facility, the entity graph in the knowledge graph is searched to obtain the connection relationship of the target facility; When the query request includes alarm information, wherein the alarm information records energy consumption anomaly information of the oilfield gathering, transportation and injection water system, the query result corresponding to the alarm information is obtained, wherein the query result includes at least one of location information, anomaly cause and handling measures, wherein the anomaly cause and handling measures are obtained from the concept graph in the knowledge graph, and the location information is obtained by performing entity query in the entity graph in the knowledge graph using event location process information determined in the concept graph in the knowledge graph; S602: Output the query results.

[0067] The method provided in this application utilizes a knowledge graph to respond to query requests from oilfield water gathering and injection systems. It enables timely, comprehensive, and intelligent decision recommendations, thereby improving the energy management level of oilfield water gathering and injection systems and providing reference opinions for energy management decisions.

[0068] By combining physical maps and conceptual maps, more specific and comprehensive information can be obtained, providing reference for energy management decisions in oilfield water gathering, transportation and injection systems.

[0069] The following explains how knowledge graphs can be applied: Figure 7 This is a flowchart illustrating the knowledge graph application provided in an embodiment of this application. Figure 7 As shown, when an energy consumption anomaly occurs in the oilfield's water gathering and injection system, the production management system returns an alarm message. This alarm message includes the alarm location, alarm time, alarm terminology, and alarm description. The visualization platform performs a knowledge graph retrieval operation to obtain the location information, anomaly cause, and recommended measures corresponding to the alarm message. The location information indicates the position of the alarm location within the pipeline network topology. The anomaly cause can be determined based on the anomaly causes in the concept graph, and the recommended measures can be determined based on the handling measures in the concept graph.

[0070] Analyze alarm information, locate the corresponding entity node in the entity graph based on the alarm location, and return the reason for the abnormal monitoring parameters of the node and the corresponding handling measures in combination with the concept graph.

[0071] Driven by business needs, a visualization platform was built in a Python 3.6 development environment on Windows 10, including both front-end and back-end implementations. The front-end primarily used the Bootstrap framework, along with HTML, CSS (Cascading Style Sheets), and JavaScript to create and display the visualization platform interface. The back-end used the FLASK framework to construct business logic such as knowledge retrieval and to facilitate interaction between the front-end and back-end. The FLASK object's `route` method sends requests to the server via GET or POST interfaces. After obtaining the front-end query request (GET request) data using request.args.get (a computer operation command), the request-related nodes and relationships can be obtained in real time by accessing the Neo4j graph database through the Py2neo interface; or entity queries can be completed in Neo4j offline mode using JSON format data extracted from the graph database in advance, and finally the query results can be returned to the front-end for display via jsonify.

[0072] The visualization platform enables convenient intelligent information retrieval, anomaly cause reasoning, and decision recommendation functions; among which: For intelligent information retrieval, since knowledge graphs represent the semantic relationships between knowledge related to energy consumption anomalies in the form of graphs, after obtaining search keywords from the search box or detecting clicked page display tags, they are mapped to specific concepts or entities, and all related information is returned, thus realizing intelligent information retrieval.

[0073] Among them, intelligent information retrieval includes retrieval of energy consumption concept relationships based on concept graphs and retrieval of pipeline entity relationships based on entity graphs.

[0074] Figure 8 This is a schematic diagram illustrating the query results for the energy consumption concept relationship provided in an embodiment of this application. For example... Figure 8 As shown, Figure 8 The query results shown are retrieved using the term "installation room". The entity type for "installation room" is "site", while the entity type for this query result is "event location". Figure 8 The locations of possible abnormal events in the filling room are indicated as the inlet water manifold, the water injection valve assembly, the water injection pump inlet manifold, and the water injection pump outlet manifold.

[0075] Figure 9 This is a schematic diagram illustrating the query results of pipeline entity relationships provided in an embodiment of this application. For example...Figure 9 As shown, Figure 9 The query results shown are obtained by searching for "1-2# Injection and Dispensing Room". The results indicate that the upstream of the 1-2# Injection and Dispensing Room is the water supply pipe to room 1-2#, and the downstream is 14 injection pipes, including Wei 1-27-3. These 14 injection pipes are not listed in the provided text. Figure 9 All are shown in the image.

[0076] For anomaly cause reasoning and decision recommendation, a relationship chain of energy consumption anomalies can be established based on the knowledge graph of the oilfield's water gathering, transportation, and injection system. When abnormal parameter fluctuations are detected, path analysis is performed through the relationship chain to locate possible factors, thereby taking targeted safety measures to control energy consumption factors and avoid large amounts of energy waste or production accidents caused by continuous parameter fluctuations.

[0077] In practical applications, interactive queries between entity maps and conceptual maps can be provided. By entering any facility, the monitoring parameters related to that facility can be queried. Clicking on the query results will take you to the "Search Energy Consumption Concept Relationships" or "Search Pipeline Entity Relationships" page for further retrieval.

[0078] Figure 10 This is a schematic diagram illustrating the reasoning process for alarm information provided in the embodiments of this application. For example... Figure 10 As shown, taking the alarm message of a continuous decrease in pressure at the outlet manifold of the water injection pump in the 1-2# injection preparation room as an example, the location of the event may be due to changes in its own conditions, or it may be affected by fluctuations in its upstream and downstream nodes. Figure 10 The upstream and downstream causes were determined based on the water injection pump outlet manifold and the actual process flow. Upstream of the water injection pump outlet manifold is the water injection pump, and downstream is the water injection well. The actual cause of the anomaly could be one factor or a combination of multiple factors. Based on the identified possible factors according to the relationship chain, adjustments were made to the corresponding equipment units in conjunction with the actual site conditions to effectively prevent a continuous increase in energy consumption.

[0079] Figure 11 This is a schematic diagram of the structure of the knowledge graph construction apparatus provided in an embodiment of this application. Figure 11 As shown, the device is applied to an oilfield water gathering, transportation, and injection system. The knowledge graph includes a concept graph, and the device includes:

[0080] The first acquisition module 1101 is configured to acquire text data of energy consumption control in the oilfield water gathering and injection system, wherein the text data records anomaly analysis information and resolution process information. The entity recognition module 1102 is configured to treat parameter anomalies, anomaly causes, handling measures, and at least one piece of information used to characterize the event localization process as entities in the concept graph, and to perform entity recognition on the text data to obtain naming information for each type of entity in the text data; wherein, the parameter anomaly indicates an abnormal situation in which the parameter occurs, the anomaly cause indicates the reason that causes the parameter to become abnormal, and the handling measures indicate the targeted handling measures taken. The relation extraction module 1103 is configured to obtain entity relationships between different types of entities in the text data based on the entity naming information, wherein the entity relationships in the text data include bidirectional relationships between parameter anomalies and anomaly causes, bidirectional relationships between anomaly causes and handling measures, and bidirectional relationships between parameter anomalies and at least one piece of information used to characterize the event localization process. The generation module 1104 is configured to generate the concept graph based on the entity relationships of the text data.

[0081] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0082] Figure 12 This is a schematic diagram of the structure of a knowledge graph management device provided in an embodiment of this application. Figure 12 As shown, the device is used in an oilfield water gathering, transportation, and injection system, and the device includes:

[0083] The second acquisition module 1201 is configured to handle query requests for the knowledge graph. The processing module 1202 is configured to obtain the query result corresponding to the query request from the knowledge graph, wherein the knowledge graph is constructed using the method described above; Output module 1203 is configured to output the query results.

[0084] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0085] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the present invention according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0086] Based on the same inventive concept, embodiments of the present invention also provide a storage medium storing a computer program thereon, wherein when the program is executed by a processor, it implements the steps of the knowledge graph construction method or the knowledge graph management method in any of the above possible implementations.

[0087] Optionally, the storage medium may be a non-transitory computer-readable storage medium, such as a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.

[0088] Based on the same inventive concept, embodiments of the present invention also provide a computer program product, including a computer program, wherein when the program is executed by a processor, it implements the steps of the knowledge graph construction method or the steps of the knowledge graph management method in any of the above possible implementations.

[0089] Based on the same inventive concept, embodiments of this application also provide an electronic device, including a memory (e.g., non-volatile memory), a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the knowledge graph construction method or the knowledge graph management method in any of the above possible implementations, which can be equivalent to the knowledge graph construction device or knowledge graph management device described above. Of course, the processor can also be used to process other data or perform calculations. This electronic device can be a PC, server, terminal, or other similar device.

[0090] It should be noted that the aforementioned knowledge graph construction device can be implemented by software. As a logical device, it is formed by the processor of the electronic device in which it is located reading the computer program instructions stored in the non-volatile memory into the memory for execution.

[0091] The embodiments of the subject matter and functional operation described in this specification can be implemented in the following ways: digital electronic circuits, tangibly embodied computer software or firmware, computer hardware including the structures disclosed in this specification and their structural equivalents, or combinations thereof. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible, non-transitory program carrier for execution by a data processing apparatus or for controlling the operation of a data processing apparatus. Alternatively or additionally, the program instructions may be encoded on artificially generated propagation signals, such as machine-generated electrical, optical, or electromagnetic signals, which are generated to encode information and transmit it to a suitable receiving device for execution by the data processing apparatus. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or combinations thereof.

[0092] The processing and logic flow described in this specification can be executed by one or more programmable computers that execute one or more computer programs to perform corresponding functions by operating on input data and generating output. The processing and logic flow can also be executed by dedicated logic circuitry—such as FPGAs (Field-Programmable Gate Arrays) or ASICs (Application-Specific Integrated Circuits), and the device can also be implemented as dedicated logic circuitry.

[0093] Suitable computers for executing computer programs include, for example, general-purpose and / or special-purpose microprocessors, or any other type of central processing unit. Typically, the central processing unit receives instructions and data from read-only memory and / or random access memory. The basic components of a computer include a central processing unit for implementing or executing instructions and one or more memory devices for storing instructions and data. Typically, a computer will also include one or more mass storage devices for storing data, such as disks, magneto-optical disks, or optical disks, or the computer will be operatively coupled to such mass storage devices to receive data from or transfer data to them, or both. However, a computer is not required to have such devices. Furthermore, a computer can be embedded in another device, such as a mobile phone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a global positioning system (GPS) receiver, or a portable storage device such as a universal serial bus (USB) flash drive, to name a few.

[0094] Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media, and memory devices, such as semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and CD-ROM and DVD-ROM disks. Processors and memory may be supplemented by or incorporated into dedicated logic circuitry.

[0095] While this specification contains numerous specific implementation details, these should not be construed as limiting the scope of any invention or the scope of the claims, but rather are primarily intended to describe features of specific embodiments of a particular invention. Certain features described in the various embodiments herein may also be implemented in combination in a single embodiment. Conversely, various features described in a single embodiment may also be implemented separately in various embodiments or in any suitable sub-combination. Furthermore, while features may function in certain combinations as described above and even initially claimed in this way, one or more features from a claimed combination may be removed from that combination in some cases, and a claimed combination may refer to a sub-combination or a variation thereof.

[0096] Similarly, although the operations are depicted in a specific order in the accompanying drawings, this should not be construed as requiring these operations to be performed in the specific order shown or sequentially, or requiring all illustrated operations to be performed to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system modules and components in the above embodiments should not be construed as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0097] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims may be performed in a different order and still achieve the desired result. Furthermore, the processes depicted in the drawings are not necessarily shown in a specific order or sequence to achieve the desired result. In some implementations, multitasking and parallel processing may be advantageous.

[0098] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0099] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for constructing a knowledge graph, characterized in that, Applied to oilfield water gathering, transportation, and injection systems, the knowledge graph includes a concept graph, and the method includes: Acquire text data on energy consumption control in the oilfield water gathering, transportation and injection system, wherein the text data records anomaly analysis information and resolution process information; The concept graph uses parameter anomalies, anomaly causes, handling measures, and at least one piece of information used to characterize the event localization process as entities. Entity recognition is performed on the text data to obtain the naming information of each type of entity in the text data. Herein, parameter anomalies represent abnormal situations in which parameters occur, anomaly causes represent the reasons that lead to parameter anomalies, and handling measures represent targeted measures taken. Based on the naming information of the entities, obtain the entity relationships between different types of entities in the text data, wherein the entity relationships in the text data include bidirectional relationships between parameter anomalies and anomaly causes, bidirectional relationships between anomaly causes and handling measures, and bidirectional relationships between parameter anomalies and at least one piece of information used to characterize the event localization process. The concept map is generated based on the entity relationships in the text data.

2. The method according to claim 1, characterized in that: At least one piece of information used to characterize the event localization process includes a station, related equipment, event location, and related parameters; wherein, the station refers to different types of stations in the oilfield water gathering, transportation, and injection system, the related equipment refers to the main process equipment in the process flow, the event location refers to the location node where the abnormal event occurred, and the related parameters are the process parameters monitored by the event location or the related equipment; Among them, the bidirectional relationship between parameter anomalies and at least one piece of information used to characterize the event localization process includes the bidirectional relationship between parameter anomalies and related parameters; The bidirectional relationships between entities used to characterize the event localization process include the bidirectional relationships between relevant parameters and relevant equipment, the bidirectional relationships between relevant parameters and event location, the bidirectional relationships between event location and station, and the bidirectional relationships between event location and relevant equipment.

3. The method according to claim 2, characterized in that, The knowledge graph also includes an entity graph, wherein the entity graph records the equipment asset data and pipeline topology of the oilfield water gathering, transportation and injection system. The entity map uses stations, equipment, and pipelines as entities, and the relationships between entities are used to represent connection information between facilities. The entities in the entity map include at least the stations, equipment, and pipelines involved in the entities of the conceptual map. The stations in the entity map provide configuration information for the stations in the conceptual map within the oilfield's water gathering and injection system. The equipment in the entity map provides configuration information for the relevant equipment in the conceptual map within the oilfield's water gathering and injection system. At least one of the equipment and pipelines in the entity map provides definition information for parameter anomalies in the conceptual map.

4. The method according to any one of claims 1 to 3, characterized in that: The text data was obtained through at least one of the following: energy consumption anomaly log, system operation and maintenance manual, and expert experience and knowledge. Specifically, the Chinese characters in the text data are converted into vectors, and a single Chinese character is used as the input of the BiGRU-CRF model to obtain the entity recognition result of the BiGRU-CRF model as the entity naming information of the concept graph. Specifically, based on the entity recognition results of the BiGRU-CRF model, the text data is processed using the BiGRU-ATT model to obtain the entity relationships between different types of entities in the text data.

5. The method according to claim 3, characterized in that, The generation methods of at least one of the concept graph and the entity graph include: According to the preset data storage format, each relationship between entities is recorded to obtain triple data, wherein the data storage format is entity-relation-entity; A graph corresponding to the triplet data is generated using a graph database, wherein the data storage format of the graph database is node-relation-node.

6. A method for managing knowledge graphs, characterized in that, The method, applied to oilfield water gathering, transportation, and injection systems, includes: Obtain the query request for the oilfield water gathering, transportation and injection system; Obtain the query result corresponding to the query request from the knowledge graph, wherein the knowledge graph is constructed using the method described in any one of claims 1 to 5; Output the query results.

7. The method according to claim 6, characterized in that, The step of obtaining the query results corresponding to the query request from the knowledge graph includes: When the query request is used to query the energy consumption concept relationship of the target site, the concept graph in the knowledge graph is searched to obtain the event location corresponding to the target site; When the query request is used to query the pipeline entity relationship of the target facility, the entity graph in the knowledge graph is searched to obtain the connection relationship of the target facility; When the query request includes alarm information, wherein the alarm information records energy consumption anomaly information of the oilfield gathering, transportation and injection water system, the query result corresponding to the alarm information is obtained, wherein the query result includes at least one of location information, anomaly cause and handling measures, wherein the anomaly cause and handling measures are obtained from the concept graph in the knowledge graph, and the location information is obtained by performing an entity query in the entity graph in the knowledge graph using event location process information determined in the concept graph in the knowledge graph.

8. A knowledge graph construction apparatus, characterized in that, Applied to oilfield water gathering, transportation, and injection systems, the knowledge graph includes a concept graph, and the device includes: The first acquisition module is configured to acquire text data on energy consumption control in the oilfield water gathering and injection system, wherein the text data records anomaly analysis information and resolution process information; The entity recognition module is configured to treat parameter anomalies, anomaly causes, handling measures, and at least one piece of information used to characterize the event localization process as entities in the concept graph, and to perform entity recognition on the text data to obtain naming information for each type of entity in the text data; wherein, parameter anomalies represent abnormal situations in which parameters occur, anomaly causes represent the reasons that lead to the parameter anomalies, and handling measures represent targeted measures taken. The relationship extraction module is configured to obtain entity relationships between different types of entities in the text data based on the entity naming information. The entity relationships in the text data include bidirectional relationships between parameter anomalies and anomaly causes, bidirectional relationships between anomaly causes and handling measures, and bidirectional relationships between parameter anomalies and at least one piece of information used to characterize the event localization process. The generation module is configured to generate the concept graph based on the entity relationships in the text data.

9. A knowledge graph management device, characterized in that, The device, used in oilfield water gathering, transportation, and injection systems, includes: The second acquisition module is set up as a query request for the knowledge graph; The processing module is configured to obtain the query result corresponding to the query request from the knowledge graph, wherein the knowledge graph is constructed using the method described in any one of claims 1 to 5; The output module is configured to output the query results.

10. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5 or the steps of the method according to claim 6 or 7.

Images