CN107885747B

CN107885747B - Semantic relation generation method and equipment

Info

Publication number: CN107885747B
Application number: CN201610868517.1A
Authority: CN
Inventors: 余明; 袁勇; 王琪
Original assignee: Siemens AG
Current assignee: Siemens AG
Priority date: 2016-09-29
Filing date: 2016-09-29
Publication date: 2022-06-28
Anticipated expiration: 2036-09-29
Also published as: CN107885747A

Abstract

A semantic relationship generation method and apparatus are provided for generating a semantic relationship between at least two components in a production system, the method comprising: extracting at least one target keyword required for analyzing the operating condition from data of at least one stage of a product life cycle of the production system; and generating a semantic relation between the at least two components according to the at least one target keyword based on a semantic coupling rule. Wherein the semantic coupling rule is used for specifying a condition that the at least one target keyword should satisfy when the at least two components have the semantic relationship therebetween.

Description

Semantic relation generation method and equipment

Technical Field

The present disclosure relates to the field of industrial automation technologies, and in particular, to a method and an apparatus for generating semantic relationships between components in a production system.

Background

In the field of industrial automation, various production systems (production systems) exist. Such as: a water supply network, etc. In order to achieve an optimization of the production system, various operating conditions of the production system need to be analyzed. Taking a water supply network as an example, the leakage condition of the water supply network may need to be analyzed to prevent the pipeline from leaking; it may also be necessary to analyze the energy consumption of the water supply network for energy optimization.

In one aspect, during at least one phase of a Product Life Cycle (PLC) of a production system, various tools are required to describe the production system. Data models are often used in these tools to describe the individual components (components) in a production system and, optionally, the interrelationships between the components.

On the other hand, the operation of a production system can be analyzed by using a network system (cyber system) or the like.

When analyzing the operation of a production system, it is necessary to know information such as the composition of the production system. Therefore, a description is needed of a production system in a network system such as the above. In order to meet the requirement of the analysis of the running condition, not only the description of each component of the production system but also the description of the relationship between the components are required, and we refer to the relationship between the components of the production system described in the network system as "semantic relationship".

How to generate semantic relationships between components of a production system used by a network system becomes an urgent problem to be solved when analyzing the operation conditions of the production system.

Disclosure of Invention

In view of this, the present invention provides a semantic relationship generating method and apparatus for generating a semantic relationship between at least two components in a production system.

In a first aspect, an embodiment of the present invention provides a semantic relationship generating method, where the method is configured to generate a semantic relationship between at least two components in a production system, where the semantic relationship is used to analyze a type of operation condition of the production system, and the method includes:

extracting at least one target keyword required for analyzing the operating condition from data of at least one stage of a product life cycle of the production system;

and generating a semantic relation between the at least two components according to the at least one target keyword based on a semantic coupling rule, wherein the semantic coupling rule is used for stipulating a condition which the at least one target keyword should meet when the at least two components have the semantic relation.

According to the technical scheme of the embodiment of the invention, the automatic generation of the semantic relation by the equipment is realized, the effectiveness and the accuracy of the semantic relation are fully ensured, and the defects caused by the manual determination of the semantic relation in the prior art are overcome. Target keywords are extracted according to the operation condition to be analyzed, useless keywords are prevented from being extracted, and the efficiency of the whole semantic relation generation process is improved. The extracted target keywords are required for analyzing the operation condition, so that the accuracy of the analysis result is ensured. The semantic coupling rule is used for stipulating the conditions which are required to be met by the target keywords when the at least two components have the semantic relationship, and a feasible scheme for automatically generating the semantic relationship is provided.

According to a first aspect, in a first implementation manner, the at least one target keyword includes at least one first keyword for describing a mutual relationship between the at least two components, and generating a semantic relationship between the at least two components according to the at least one target keyword based on a semantic coupling rule includes:

and when the at least one first keyword meets the condition specified by the semantic coupling rule, determining the semantic relation according to the mutual relation described by the at least one first keyword.

According to the technical scheme of the embodiment of the invention, the automatic generation of the semantic relation by the equipment based on the keywords describing the mutual relation between at least two components is realized, the effectiveness and the accuracy of the semantic relation are fully ensured, and the flexibility is improved.

According to the first aspect, in a second implementation manner, the generating a semantic relationship between the at least two components according to the at least one target keyword includes, for each of the at least two components, a second keyword having an attribute and used for describing a value of the attribute, and based on a semantic coupling rule, generating the semantic relationship between the at least two components according to the at least one target keyword, including:

and when the mutual relation between the values of the second keywords corresponding to the at least two components meets the condition specified by the semantic coupling rule, determining the semantic relation according to the value of the second keyword of each component of the at least two components.

According to the technical scheme of the embodiment of the invention, the semantic relation is automatically generated by the equipment based on the keywords describing the attributes of the single component, the semantic relation between the components is determined according to the value of the attributes, the semantic relation can be deeply mined, and the generated semantic relation is more comprehensive and accurate.

According to the first aspect as well as the first or second implementation manner of the first aspect, in a third implementation manner, the extracting, from data of at least one stage of a product life cycle of the production system, at least one target keyword required for analyzing the operation condition includes:

extracting the at least one target keyword from data of a design phase and/or a planning phase of the production system.

According to the technical scheme of the embodiment of the invention, the automatic generation of the semantic relation by the equipment based on the data of the design stage and/or the planning stage of the production system is realized.

According to the first aspect, the first, second or third implementation manner of the first aspect, in a fourth implementation manner, before generating a semantic relationship between the at least two components according to the at least one target keyword based on a semantic coupling rule, the method further includes:

and generating the semantic coupling rule according to the analysis requirement of the running condition.

According to the technical scheme of the embodiment of the invention, the semantic coupling rule can be automatically generated according to the analysis requirement of the operating condition, and the semantic relation is automatically generated according to the generated semantic coupling rule, so that the semantic relation is more targeted according to the analysis requirement of the specific operating condition in the generated semantic relation, and the effectiveness and the reliability of the generated semantic relation are further improved.

According to the first aspect as well as the first, second, third or fourth implementation manners of the first aspect, in a fifth implementation manner, the semantic relationship includes at least one of a physical connection relationship, a dependency relationship, and a control relationship.

According to the technical scheme of the embodiment of the invention, more flexible and diversified automatic generation of the semantic relation is realized.

In a second aspect, an embodiment of the present invention provides a semantic relationship generating device, where the device is configured to generate a semantic relationship between at least two components in a production system, where the semantic relationship is used to analyze a type of operation condition of the production system, and the semantic relationship generating device includes:

a target keyword extraction module for extracting at least one target keyword required for analyzing the operation condition from data of at least one stage of a product life cycle of the production system;

and the semantic relation generating module is used for generating the semantic relation between the at least two components according to the at least one target keyword extracted by the target keyword extracting module based on a semantic coupling rule, wherein the semantic coupling rule is used for stipulating the condition which the at least one target keyword should meet when the at least two components have the semantic relation.

According to the technical scheme of the embodiment of the invention, the automatic generation of the semantic relation by the equipment is realized, the effectiveness and the accuracy of the semantic relation are fully ensured, and the defects caused by the manual determination of the semantic relation in the prior art are overcome. Target keywords are extracted according to the operation condition to be analyzed, useless keywords are avoided being extracted, and the efficiency of the whole semantic relation generation process is improved. The extracted target keywords are required for analyzing the operation condition, so that the accuracy of the analysis result is ensured. The semantic coupling rule is used for stipulating the conditions which are required to be met by the target keywords when the at least two components have the semantic relationship, and a feasible scheme for automatically generating the semantic relationship is provided.

According to the second aspect, in a first implementation manner, the at least one target keyword extracted by the target keyword extraction module includes at least one first keyword for describing a relationship between the at least two components, and the semantic relationship generation module is specifically configured to:

According to the second aspect, in a second implementation manner, the at least one target keyword extracted by the target keyword extraction module includes a second keyword which has an attribute for each of the at least two components and is used to describe a value of the attribute, and the semantic relationship generation module is specifically configured to:

According to the technical scheme of the embodiment of the invention, the automatic generation of the semantic relationship by the equipment based on the key words for describing the attributes of the single component is realized, the effectiveness and the accuracy of the semantic relationship are fully ensured, the semantic relationship among the components is determined according to the values of the attributes, the semantic relationship can be deeply mined, and the generated semantic relationship is more comprehensive and accurate.

According to the second aspect and the first or second implementation manner of the second aspect, in a third implementation manner, the target keyword extraction module is specifically configured to:

According to the second aspect, the first, second or third implementation manner of the second aspect, in a fourth implementation manner, the system further includes a semantic coupling rule generating module, configured to generate the semantic coupling rule according to an analysis requirement of the operating condition before the semantic relationship generating module generates the semantic relationship between the at least two components according to the at least one target keyword based on the semantic coupling rule.

According to the technical scheme of the embodiment of the invention, the semantic coupling rule can be automatically generated according to the analysis requirement of the operation condition, and the semantic relation is automatically generated according to the generated semantic coupling rule, so that the semantic relation is more targeted according to the analysis requirement of the specific operation condition in the generated semantic relation, and the effectiveness and the reliability of the generated semantic relation are further improved.

According to the second aspect and the first, second, third or fourth implementation manners of the second aspect, in a fifth implementation manner, the semantic relationship generated by the semantic relationship generation module includes at least one of a physical connection relationship, a dependency relationship and a control relationship.

In a third aspect, an embodiment of the present invention provides a semantic relationship generating apparatus, where the apparatus is configured to generate a semantic relationship between at least two components in a production system, where the semantic relationship is used to analyze a type of operation condition of the production system, and the apparatus includes:

at least one memory for storing computer instructions;

at least one processor, configured to invoke the computer instruction, to execute the method provided in the first aspect or any implementation manner of the first aspect.

In a fourth aspect, the present invention provides a computer-readable medium, on which computer instructions are stored, and when executed by a processor, the computer instructions cause the processor to execute the method described in the first aspect or any implementation manner of the first aspect.

Drawings

The above and other features and advantages of the present method will become more apparent to those of ordinary skill in the art by describing in detail preferred embodiments thereof with reference to the attached drawings, in which:

fig. 1 shows an example of a network topology of a water supply network described using the Arcsis Pipeline Data Model (APDM);

FIG. 2 illustrates an exemplary data structure employed in describing a network topology for the water supply network shown in FIG. 1 using an APDM;

FIG. 3 illustrates another exemplary data structure employed in describing a network topology for the water supply network shown in FIG. 1 using APDM;

FIG. 4 shows an example of a water supply network logical topology reconstructed in the analysis software (EPANET);

FIG. 5 illustrates an example of a Petri Net generated by the semantic coupling method according to an embodiment of the invention;

FIG. 6 is a flowchart of a semantic relationship generation method according to an embodiment of the present invention;

fig. 7 is a schematic structural diagram of a semantic relationship generating device according to an embodiment of the present invention;

fig. 8 is a schematic structural diagram of a semantic relationship generation apparatus according to another embodiment of the present invention;

fig. 9 is a schematic structural diagram of a semantic relationship generating apparatus according to yet another embodiment of the present invention.

List of reference numerals:

p1, P2, P3: pipe section V1: valve with a valve body

21: water supply equipment 22: connection class

221: diameter 222: material of

223: connection type 31: building data structure

32: pipeline closest point location data structure 33: pipe segment data structure

311: building name 312: type of building

313: whether or not the person owns 321: building event ID

331: line ring event ID 332: parent event ID

333: last station column event ID 334: last station train connection coefficient

335: the next station is listed as event ID 336: next station connection coefficient

337: reference model 338: name of station

338: station column ID

p1, p2, p3, p4, p 5: resource/behavior t1, t2, t3, t 4: state change of resource

601: extracting at least one target keyword required for analyzing the operation condition from data of at least one stage of a product life cycle of the production system

602: generating a semantic relationship between the at least two components according to the at least one target keyword based on a semantic coupling rule, wherein the semantic coupling rule is used for stipulating a condition which the at least one target keyword should satisfy when the semantic relationship is formed between the at least two components

71: target keyword extraction module

72: semantic relationship generation module 83: semantic coupling rule generation module

90: the memory 91: processor with a memory having a plurality of memory cells

Detailed Description

In the embodiment of the invention, the semantic relation among the components of the production system is generated based on the analysis requirement of the running condition of the production system. There may be differences in the semantic relationships needed to analyze different operating conditions. Such as: when the leakage condition of a water supply network is analyzed, more connection relations among pipelines and position relations between the pipelines and a water pump, a valve and the like are required to be considered; when analyzing the energy consumption condition of the water supply network, the mutual relationship between the energy consumption devices such as the water pump and the like in the whole water supply network and other devices in the flow process of water supply may need to be considered more. Therefore, the generated semantic relation aims at the operation condition to be analyzed, and the accuracy of operation condition analysis can be ensured.

In the process of generating the semantic relation, target keywords required by analyzing the operation condition of the production system can be extracted from data of a Product Life Cycle (PLC) of the production system; and then generating semantic relations among the components of the production system according to the extracted target keywords based on the semantic coupling rules. Target keywords are extracted according to the operation condition to be analyzed, useless keywords are avoided being extracted, and the efficiency of the whole semantic relation generation process is improved. The extracted target keywords are required for analyzing the operation condition, so that the accuracy of the analysis result is ensured. The semantic coupling rule is used for stipulating the conditions which are required to be met by the target keywords when the at least two components have the semantic relationship, and a feasible scheme for automatically generating the semantic relationship is provided.

In the following, some descriptions and concepts related to the embodiments of the present invention are explained for easy understanding. It should be noted that these explanations should not be construed as limiting the scope of the present invention.

1. Production system

A system that converts a production demand into a final product and/or service can be considered a production system.

2. Assembly

Production systems typically include at least one component. Taking production systems in the process industry as an example, these production systems can generally include an execution component, a control component, and an environment component.

Taking the water supply network as an example of the production system, the actuating assembly may include a pump (pump) and a valve (valve); the control component may include a Programmable Logic Controller (PLC) or the like; the environmental components may include a reservoir (tank) and a pipeline (pipe), among others.

3. Life cycle of the product

Product lifecycle refers to the entire process from demand analysis, product planning, product design, manufacturing, testing, delivery through maintenance, and scrapping.

4. The operation condition of the production system and the semantic description of the production system in the network system.

To optimize, control and manage the production system, it is necessary to analyze the operation of the production system. The data on which the analysis is based may include information on components of the production system, and in embodiments of the present invention, information on semantic relationships between components of the production system.

Information required for analyzing the operation of the production system may be stored in the network system, and the operation of the production system may be analyzed based on the information by software in the network system. The network system may collect field data (e.g., data collected by sensors) of the production system, send control commands to the production system, submit analysis results to the information/control system of the upper layer, and so on.

In the past, analysis of production system operation has relied primarily on the experience of field experts, engineers, or workers. However, as the complexity of production systems continues to increase, analysis faces more and more challenges. Therefore, network systems have emerged that analyze the operation of production systems.

In order to analyze the operation condition of the network system, a basic requirement is to make the network system know the actual condition of the production system in a real environment. To fulfill this need, a standardized description of the production system needs to be established in the network system, so as to convert the real environment of the production system into a semantic description that can be understood by the network system based on the standardized description.

Specifically, in the network system, information such as performance and parameters of a component of the production system can be known through keywords for describing the component. In the embodiment of the invention, the network system can also acquire semantic relations (also called as virtual semantic relations) among different components of the production system, and the operation condition of the production system is analyzed based on the acquired semantic relations among the components.

5. Semantic relationships

The semantic relationships between the production system components described in the network system may include, but are not limited to, the following relationships: physical connection relationships, control relationships, dependency relationships, and the like.

In the embodiment of the invention, when the network system is used for analyzing the operation condition of the production system, the network system not only knows the performance, parameters and the like of each system component, but also knows the semantic relation among the components.

Because the semantic relationship between the concerned system components may also change when the analysis requirements for different operating conditions are met, in the embodiment of the present invention, the semantic relationship between the corresponding components is obtained according to the analysis requirements for different operating conditions.

6. Description of production systems in various stages of the product lifecycle

1) Description of production systems in Geographic Information Systems (GIS)

As an example, the network system may analyze the operation of the production system based on the information of the production system stored in the GIS. Therefore, the description of the production system will be described with reference to the information of the production system stored in the GIS as an example. A GIS system uses data models to describe production systems, and the data models used to describe different production systems may vary. When describing the production system of the water supply network described above, the GIS system describes at least one phase of the product life cycle of the water supply network, for example using the ArcGIS Pipeline Data Model (APDM) of ArcGIS. More specifically, FIG. 1 shows an example of a network topology for a water supply network described using APDM. Fig. 2 and 3 illustrate two exemplary data structures employed in describing the network topology of the water supply network shown in fig. 1 using APDM. A detailed description of the use of an APDM to describe a water supply network is described herein in connection with fig. 1-3.

To describe the network topology of a water supply network as shown in fig. 1, it is necessary, on the one hand, to describe the equipment, such as pumps, valves, etc. in the water supply network, at the connection point of two lines in the water supply network where there is some kind of transition. Wherein, pipeline, pump, valve are all the components of this production system of water supply network. In the APDM, the connection (fixing) class is used for this description, and the data structure shown in fig. 2 is the data structure of the connection (fixing) class. As shown in fig. 2, the connection class 22 of the water supply equipment class (WaterFacility)21 in the water supply network may specifically define the Diameter (Diameter)221 of the component, the Material (Material)222 of the component, and the type of connection (jointype) 223 between the two lines forming the connection point where the component is located. Connection types 223 may specifically include bends, crosses, couplings, and the like. If the connection type is coupling, it means that two small lengths of pipeline are simply connected to form a new structure, and the characteristics of the two pipelines are identical, so the two pipelines should be regarded as one pipeline. By defining the connection type in the connection class, the components at the connection point of two pipelines with different connection types can be described by the same connection class data structure, so that the number of network feature classes is reduced, and the performance of the GIS can be improved.

In another aspect of the embodiment of the present invention, the relationship between the pipe line in the water supply network and the geographical location information of the equipment, such as the pump and the valve, is also described, and fig. 3 shows an exemplary data structure in the APDM for describing the relationship between the pipe line in the water supply network and the geographical location information of the equipment. As shown in FIG. 3, the data structures involved include a building data Structure (Structure)31, a pipeline closest point location data Structure (Structure location)32, and a pipe segment data Structure (Stationseries) 33.

Wherein the building data structure 31 is used to describe the geographical location points closest to the pipeline, in particular, the geographical location point for each pump or valve etc. in the water supply network, and the building data structure is used to describe the geographical location of the equipment, such as pumps and valves, etc. on the pipeline. More specifically, the location point type in the building data structure 31 is an Environmental Systems Research Institute (ESRI) geographic data point, which is false if there is an M value and true if there is a Z value. As shown in fig. 3, the building data structure 31 includes: a building name 311 set to an ESRI field data type string type; a building type 312, setting the building type 312 as unknown; if owned by person 313, set option Yes/No to NO.

The pipe segment data structure 33 is used to describe the position of the start point and the end point of each pipe segment, wherein a pipe segment is a section of a pipeline, and the pipeline is formed by connecting a plurality of pipe segments, so that the position of the pipeline can be calculated based on the position of the start point and the end point of each pipe segment. Specifically, the location type in the pipe segment data structure 33 is an ESRI geographic data point, which is false if there is an M value and true if there is a Z value. As shown in FIG. 3, the pipe segment data structure 33 includes: pipeline ring event Identification (ID) 331, which takes the ESRI field data type Global ID; a father station column event ID 332 which adopts an ESRI field data type global ID; last station column event ID 333, which is ESRI field data type global ID; a last station train connection coefficient 334 set to an ESRI field data type double precision number; the next station column event ID 335, which is an ESRI field data type global ID; a next-train connection coefficient 336 set to the ESRI field data type double-precision number; a reference model 337, setting the reference model 337 as a continuous model; a station list name 338, wherein the station list name 338 employs an ESRI field data type string; station column ID 339, which is an ESRI field data type integer.

The pipe segment closest point location data structure 32 is used to describe the pipeline and the closest geographical location point (e.g., pump or valve, etc.) to the pipeline. Specifically, the position of the pipeline, i.e., the Route Structure Location, may be calculated based on information described in the pipe segment data Structure, and the relationship of pipe segments and geographical Location points, i.e., the station train Structure Location, may be calculated based on information described in the building data Structure and the pipe segment data Structure. More specifically, the location type in the pipe segment closest point location data structure 32 is an ESRI geographic data point, which is false if there is an M value and true if there is a Z value. As shown in FIG. 3, the pipe segment closest point location data structure 32 includes: the building event ID 321, which is an ESRI field data type global ID.

The connection class and the geographic position information are keywords which are stored in the GIS and used for describing the water supply pipe network system.

2) Description of a production System by a Petri Net Process model

The Petri Net process model comprises two nodes of a PLACE (PLACE) and a TRANSITION (TRANSITION). When describing a work process, the decisive component is a resource and an activity (activity). Resources and behaviors can be mapped to libraries in Petri Net and state changes of resources can be mapped to transitions in Petri Net. Taking an Automatic Guided Vehicle (AGV) for picking up goods, the AGV has four states P1-P4 (namely, garage), where P1 means that the AGV is idle, P2 means that goods on a buffer station are waiting to be sent, P3 means that the AGV picks up goods and sends the goods to a warehouse, and P4 means a warehousing platform for unloading the goods. Two transitions may be defined, T1 indicates the event that the AGV picks a load from the loading dock, and T2 indicates the event that the AGV unloads a load from the loading dock.

The entire flow can be described as follows using Petri Net:

|--<-P1<-----|

||

P2→T1-->P3-->T2-->P4

the network system can perform analysis based on the Petri Net process model when describing the working process of the production system, such as workflow management.

7. Target keywords

The network system uses different keywords to describe different characteristics of the production system, and since the characteristics of the production system concerned by different analysis requirements may be different, the target keywords required by different analysis requirements may also be different. Thus, when analyzing a production system, target keywords are determined according to specific analysis requirements, the target keywords being one or more of the keywords used in the network system to describe the production system.

For example, when leakage detection is performed on a water supply network as shown in fig. 2, the network system needs to establish a logic topology of the water supply network that the network system can understand based on physical topology information of the water supply network stored in the GIS, so as to calculate a hydraulic model based on the logic topology, and analyze where leakage occurs. Fitting and Location should be used as the target keywords to be extracted to obtain the pumps/valves described by the connection class at the connection point of two pipelines in the water supply network where there is some transition, and the pipelines described by the geographical Location point and pipe section relation data structure and the pumps/valves closest to the pipelines.

For another example, when the analysis software of the network system analyzes the workflow of the production system, since the state change of the resource is the key of the analysis, the transition may be determined as the target keyword.

8. Semantic coupling rules

Semantic coupling rules are used to define relationships that at least two components need to meet in a real-world environment when they have a certain semantic relationship, namely: for establishing a mapping between the actual relationship of at least two components in the real-world environment and the semantic relationship that the network system is interested in and understands when analyzing the production system. Therefore, when the analysis requirements are different, the semantic coupling rules are usually different. The semantic coupling rule is defined by domain experts and qualified staff according to the analysis requirement, or is automatically analyzed and selected from a rule set of a corresponding application domain according to the analysis requirement by an upper information/control system.

Taking the analysis requirement for leak detection of a water supply network as shown in fig. 2 as an example, when a network system establishes a logical topology of the water supply network that can be understood by the network system based on physical topology information of the water supply network stored in the GIS, the following semantic coupling rules may be defined:

rule one is as follows: if the connection type of the connection point between two pipelines is two-way, the two pipelines are regarded as a whole; if the connection type of the connection point between two pipelines is tee (tee), a node is considered to be arranged between the two pipelines;

rule two: if the geographical position of a valve coincides with a pipeline, the valve is considered to be interconnected with the pipeline, and the valve is considered to control the on-off of the pipeline, and accordingly, the pipeline is divided into two new pipelines which are interconnected in the generated logical topology. For example, the geographic location of valve V1 coincides with spool piece P1 (as shown in FIG. 1), then in the logical topology, spool piece P1 is split into spool pieces P2 and P3 as shown in FIG. 4, where P2 is connected to V1 and V1 is connected to P3.

When leakage detection is performed in a water supply pipe network system, a pipe network physical topological structure needs to be converted into a component logical topological structure, a hydraulic model is calculated on the basis of the logical topology, and the position of leakage is analyzed.

Based on the above rules, the whole water supply network shown in fig. 1 is traversed repeatedly:

1) extracting target keywords 'fixing' and 'Location' of each component;

2) determining the current semantic relationship among the components according to a keyword matching algorithm, such as the connection type and the position relationship among the current components;

3) and judging whether the semantic coupling rule I and/or the semantic coupling rule II are/is met or not based on whether the connection type between the current components is 'two-way' or 'three-way' and whether the geographic position of the valve is superposed with the pipe section or not, and if so, coupling in the logic topology according to the corresponding rule to generate the semantic relationship between the components.

Based on the above traversal, for example, a logical topology can be established as shown in FIG. 4, where FIG. 4 shows an example of a water supply network logical topology map reconstructed in analysis software (EPANET).

In addition, when the network system plans the working process of the production system based on Petri Net, for the two behaviors a1 and a2, for example, the following semantic coupling rules are defined:

rule one is as follows: if the output of A1 is the input of A2, it means that the relationship of A1 and A2 is continuous, i.e. A1 and A2 occur chronologically one after the other;

and a second rule: if the input of A1 is the same as the input of A2, it means that the relationship of A1 to A2 is concurrent;

rule three: if the output of A1 is the same as the output of A2, this means that A1 is in sync with A2.

When semantic coupling is performed, the resource state mapped in transport is extracted for the resources and behaviors mapped in PLACE, namely: and extracting the target keywords of the component. Thereafter, the relationships of the inputs and outputs of the different behaviors are determined according to domain knowledge, such as a resource state matching algorithm, so that semantic relationships between the different behaviors, such as semantic relationships that are continued, concurrent or synchronized a1 and a2, are generated according to the mappings between the relationships of the inputs, outputs and semantic relationships of the different behaviors defined by the semantic coupling rules.

Based on the above semantic coupling relationship, for example, a Petri Net example as shown in fig. 5 may be generated, where fig. 5 shows one Petri Net example generated by the semantic coupling method according to the embodiment of the present invention. In fig. 5, p1, p2, p3, p4 and p5 are used for representing resources/behaviors, and t1, t2, t3 and t4 are used for representing state changes of the resources, which can directly reflect the logical relationship between the resources/behaviors and the state changes of the resources in the flow. With the Petri Net shown in fig. 5, it is possible to analyze whether there is a fault in the workflow, for example, determine whether there is a deadlock in the workflow according to whether the running flow directions of the resource/behavior and the resource state in fig. 5 are unified, so as to solve the fault in the workflow.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Fig. 6 is a flowchart of a semantic relationship generation method according to an embodiment of the present invention. The method is used for generating a semantic relationship between at least two components in a production system. As shown in fig. 1, the process includes the following steps:

601. extracting at least one target keyword required for analyzing the operating condition from data of at least one stage of a product life cycle of the production system;

the input information of this step is the name of at least one target keyword to be extracted, and the output is the information of the component/component relationship of the production system characterized by the extracted at least one target keyword matching the name.

In this step, target keywords can be extracted from the software tools or documents in the system design phase, which are used to characterize the key performance of the component that is affected by the current analysis requirements, such as the factors of interest or important system variables.

In the process industry, the main content of interest is simulation and optimization of a process flow, and the factors of interest may include, for example, the operating characteristics, the working timing sequence, and the like of each component, and specifically may include, for example, operating parameters, configuration parameters, system characteristics, and the like; important system variables may include, for example, application information, the values of which may help an application predict or infer the system environment.

For example, in applications where leak detection of a water supply network is performed, important system variables may include, for example, flow rate, and a line may be determined to be free of leaks if the inflow and outflow rates of the line are the same. As another example, for applications where control relationships between components in a water supply network are of concern, important system variables may include, for example, variables from which control relationships between components can be determined, such as the geographic locations of the components, where valves and lines having the same location are identified as having a control relationship, i.e.: identifying the valve controls the disconnection of the line at the same location as the valve. Generally, the target keyword is, for example, a geographic location of the component or a resource status (also referred to as a component status), wherein the resource status may include status parameters such as an operating parameter, an operating flow rate, a pressure, and a liquid level of the component.

In this step, a certain/some phase(s) of the product lifecycle of the production system having the target keyword(s) may be traversed, such as a planning phase and/or a design phase for the production system, etc., to find relationships between the components. The relationship between the components may include a logical relationship representing temporal or spatial precedence between the components during operation, or a physical connection relationship between the components, and the like. The planning stage And/or the design stage of the production System may include at least one of a Relational Database Management System (DBMS), a spatial Database, And a System structure in an XML file, for example, And the specific operation of retrieving the internal information of the planning stage And/or the design stage of the production System And extracting the target keyword may be implemented by using a Query Language (Simple Protocol And rf Query Language, rql), for example.

At this step, the name of the target keyword to be extracted may be input to the execution subject of the semantic relationship generation method (i.e., semantic relationship generation device) by a user or an external device, and specific information described by the target keyword is extracted from the stored information of the components and/or component relationships of the production system described in various keywords by the semantic relationship generation device.

In addition, the name of the target keyword to be extracted can also be obtained by analyzing operation data or storage information of the production system by the semantic relation generating device, wherein the operation data can comprise data obtained in the running process of the component, and the storage information can comprise information statically configured in advance for the component. For example, the semantic relation generating device learns that all components have a LOCATION (LOCATION) attribute through retrieval, and if some of the components have the same LOCATION, the LOCATION is taken as a target keyword, which is not limited in this embodiment.

602. And generating a semantic relation between the at least two components according to the at least one target keyword based on a semantic coupling rule, wherein the semantic coupling rule is used for stipulating a condition which the at least one target keyword should meet when the at least two components have the semantic relation.

The input information of this step is the semantic coupling rule and the information of the component/component relationship characterized by the at least one target keyword extracted in step 601, and the output information is the semantic relationship between at least two components.

The semantic coupling rule is defined by a domain expert and a senior staff according to requirements and input into the semantic relation generating equipment, or is automatically analyzed and selected from a rule set of a corresponding application domain according to requirements by an upper information/control system.

In this step, the semantic relationship generating device obtains the semantic relationship between the at least two components in the current analysis requirement according to the semantic coupling rule and the information of the component/component relationship represented by the target keywords of the at least two components extracted in step 601. For example, one specific example of the semantic coupling rule may be: if two components are in an electrically conductive relationship, the two components are considered as a new integrated component, i.e. the semantic relationship is the same integrated component. Then at this step, the client determines whether an electrical conduction relationship exists between at least two components based on the information of the component/building relationship characterized by the target keyword extracted at step 601, and if so, outputs a semantic relationship that the at least two components are the same integrated component.

The semantic relationship is generated according to the information of the component/component relationship represented by the target keyword in the real environment, which may include aggregating the relationship in the real environment as in the above example, adjusting the relationship in the real environment according to a specific semantic coupling rule, and/or adding a new relationship based on the relationship in the real environment. The embodiments of the present invention are not limited in this respect.

In the technical scheme of the embodiment of the invention, the semantic relation between the components does not need to be artificially determined by application engineers, the target keywords of the components are automatically extracted from a production system by the semantic relation generating equipment, and the semantic relation is generated based on the target keywords and the acquired semantic coupling rule. Namely: the semantic relationship generation method according to the embodiment of the invention can provide the semantic relationship understood by the network system by using the known component information based on the obtained semantic coupling rule.

For example, the keywords stored in the network system for describing the production system include connection (CONNECTING), which characterizes whether at least two components are connected in the actual environment. When different running state analysis is carried out on the production system, connection can be used as a target keyword so as to provide a semantic relation meeting the analysis requirement according to the existing connection information.

For example, when the production system is a water supply network and the component A, B, C, D, E are pipelines of the water supply network, if the reason for overflow of the pipeline a needs to be analyzed, the following semantic coupling rules can be determined:

if component A (hereinafter referred to as A) is connected to component B (hereinafter referred to as B), it means that liquid can flow from A to B and vice versa;

if A has a connection to B and B has a connection to component C (hereinafter referred to as C), it means that liquid can flow from A to C and vice versa;

from this semantic coupling rule and the information of component A, B, C, D, E described by the target keyword CONNECTING of component A, B, C, D, E, the pipeline flow value of component A, B, C, D, E can be fed back to analyze the cause of pipeline A overflow.

For another example, if the upper information/control system needs to obtain the system topology of the water supply network, all the hardware-level pipe segment connection relationships can be fed back based on the target keyword "connection".

Further, in the semantic relationship generating method of the above embodiment, the generating the semantic relationship between the at least two components according to the at least one target keyword includes:

and when the at least one first keyword meets the condition specified by the semantic coupling rule, determining the semantic relation according to the mutual relation described by the at least one first keyword. In particular, the target keywords may be used to characterize a semantic relationship between at least two components. For example, the target keyword is "connection (CONNECTING)", and the specific content of the target keyword defines the connection object of the component, for example, the target keyword CONNECTING of component a defines the connection between a and B.

Further, in the semantic relationship generating method according to the above embodiment, the generating of the semantic relationship between the at least two components according to the at least one target keyword includes, for each of the at least two components, a second keyword having an attribute and used for describing a value of the attribute, and based on a semantic coupling rule, generating the semantic relationship between the at least two components according to the at least one target keyword, including:

In particular, the target keywords may also be used to characterize the attributes of the individual components. When the target keyword is used to represent the attribute of a single component, for example, the target keyword is a geographic location, according to the semantic relationship generation method of the above embodiment, the semantic relationship between the components can be determined according to the values of the target keyword of at least two components, that is, the relationship between the geographic locations, and according to the corresponding semantic coupling rule. For example: the semantic coupling rule is as follows: if the geographic locations of component A and component B coincide, then it is deemed that component A and component B are physically connected.

According to the technical scheme of the embodiment, the semantic relation can be generated more flexibly and automatically based on different types of target keywords.

Another aspect of the embodiments of the present invention further provides a semantic relationship generating device.

Fig. 7 is a schematic structural diagram of a semantic relationship generating device according to an embodiment of the present invention. The apparatus is configured to generate a semantic relationship between at least two components in a production system, wherein the semantic relationship is configured to analyze a type of operation of the production system. As shown in fig. 7, the apparatus includes:

a target keyword extraction module 71, configured to extract at least one target keyword required for analyzing an operation condition from data of at least one stage of a product life cycle of the production system;

a semantic relationship generating module 72, configured to generate a semantic relationship between the at least two components according to the at least one target keyword extracted by the target keyword extracting module based on a semantic coupling rule, where the semantic coupling rule is used to specify a condition that the at least one target keyword should satisfy when the at least two components have the semantic relationship.

Specifically, the input information of the target keyword extraction module 71 is a name of at least one target keyword to be extracted, and the output is information of a component/component relationship of the production system characterized by the extracted at least one target keyword matching the name. As shown in fig. 7, the output of the target keyword extraction module 71 is one input of the semantic relationship generation module 72. The other input of the target keyword extraction module 71 is a semantic coupling rule, which is defined by a domain expert and qualified staff according to requirements and input to a semantic relation generation device, or automatically analyzed and selected by an upper information/control system from a rule set of a corresponding application domain according to requirements. The semantic relationship generating module 72 obtains the semantic relationship between the at least two components in the current analysis requirement according to the input semantic coupling rule and the information of the component/component relationship represented by the target keywords of the at least two components input by the target keyword extracting module 71.

The specific process of generating the semantic relationship by the semantic relationship generation device provided by the embodiment of the present invention is the same as the semantic relationship generation method provided by the embodiment of the present invention, and therefore, the detailed description thereof is omitted here.

Further, in the semantic relationship generating apparatus of the above embodiment, the at least one target keyword extracted by the target keyword extracting module 71 includes at least one first keyword for describing a mutual relationship between at least two components, and the semantic relationship generating module 72 is specifically configured to:

Further, in the semantic relationship generating device of the foregoing embodiment, the at least one target keyword extracted by the target keyword extraction module 71 includes a second keyword which has an attribute for each of the at least two components and is used to describe a value of the attribute, and the semantic relationship generating module 72 is specifically configured to:

According to the technical scheme of the embodiment of the invention, the automatic generation of the semantic relationship by the equipment based on the key words for describing the attributes of the single component is realized, the effectiveness and the accuracy of the semantic relationship are fully ensured, the semantic relationship among the components is determined according to the values of the attributes, the semantic relationship can be deeply mined, and the generated semantic relationship is more comprehensive and accurate. .

Further, in the semantic relationship generating device according to the above embodiment, the target keyword extraction module 71 is specifically configured to:

Further, in the semantic relation generating device of the above embodiment, a semantic coupling rule generating module 83 is further included, as shown in fig. 8.

Fig. 8 is a schematic structural diagram of a semantic relationship generating device according to another embodiment of the present invention. In the semantic relationship generating apparatus shown in fig. 8, the semantic coupling rule generating module 83 is configured to generate the semantic coupling rule according to the analysis requirement of the operating condition before the semantic relationship generating module 72 generates the semantic relationship between at least two components according to at least one target keyword based on the semantic coupling rule.

Further, in the semantic relation generating apparatus of the above embodiment, the semantic relation generated by the semantic relation generating module 72 includes at least one of a physical connection relation, a dependency relation, and a control relation.

The embodiment of the invention also provides semantic relation generating equipment.

Fig. 9 is a schematic structural diagram of a semantic relationship generating apparatus according to yet another embodiment of the present invention. The apparatus is configured to generate a semantic relationship between at least two components in a production system, wherein the semantic relationship is configured to analyze a type of operation of the production system. As shown in fig. 9, the apparatus includes:

at least one memory 90 for storing computer instructions;

at least one processor 91 for extracting at least one target keyword required for analyzing the operation condition from data of at least one stage of a product life cycle of the production system; and generating a semantic relation between the at least two components according to the at least one target keyword based on a semantic coupling rule, wherein the semantic coupling rule is used for stipulating a condition which should be met by the at least one target keyword when the at least two components have the semantic relation.

Further, in the semantic relation generating device of the above embodiment, the at least one target keyword includes at least one first keyword for describing a mutual relation between at least two components, and the processor 91 is specifically configured to:

Further, in the semantic relationship generating device of the above embodiment, the at least one target keyword includes a second keyword that has an attribute for each of the at least two components and is used to describe a value of the attribute, and the processor 91 is specifically configured to:

According to the technical scheme of the embodiment of the invention, the automatic generation of the semantic relation by the equipment based on the key words describing the attributes of the single component is realized, the effectiveness and the accuracy of the semantic relation are fully ensured, and the flexibility is improved.

Further, in the semantic relationship generating device according to the foregoing embodiment, the processor 91 is specifically configured to:

at least one target keyword is extracted from data of a design phase and/or a planning phase of the production system.

Further, in the semantic relation generating device of the above embodiment, the processor 91 is further configured to: generating a semantic coupling rule according to analysis requirements of an operating condition before generating a semantic relationship between at least two components according to at least one target keyword based on the semantic coupling rule.

According to the technical scheme of the embodiment of the invention, the semantic coupling rule can be automatically generated according to the analysis requirement of the operation condition, and the semantic relationship can be automatically generated according to the generated semantic coupling rule, so that the validity and reliability of the generated semantic relationship are further improved.

Further, in the semantic relation generating apparatus of the above embodiment, the semantic relation includes at least one of a physical connection relation, a subordinate relation, and a control relation.

Yet another aspect of the embodiments of the present invention further provides a computer-readable medium, on which computer instructions are stored, and when the computer instructions are executed by a processor, the computer instructions cause the processor to execute the semantic relation generating method provided in any one of the embodiments of the present invention.

Furthermore, an embodiment of the present invention further provides a computer-readable storage medium storing computer program instructions, where the computer program instructions are used to execute the semantic coupling method according to any embodiment of the present invention.

As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed terms. In addition, the singular forms "a", "an" and "the" are to be construed as "at least one" and thus may include plural referents unless expressly stated otherwise. It will be further understood that the terms "comprises," "comprising," "includes" and/or "including," when used in this specification, specify the presence of stated features, operations, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, operations, integers, steps, operations, elements, components, and/or groups thereof. Certain features are described in mutually different dependent claims, but this does not imply that these measures cannot be used in combination to advantage. A computer program may be stored/distributed on a suitable non-transitory medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the internet or other wired or wireless telecommunication systems.

The above description is only a preferred embodiment of the present method and is not intended to limit the present method, and any modification, equivalent replacement, or improvement made within the spirit and principle of the present method should be included in the protection scope of the present method.

Claims

1. A semantic relationship generation method for generating a semantic relationship between at least two components in a production system, wherein the semantic relationship is used to analyze a class of operating conditions of the production system, the method comprising:

determining at least one target keyword according to an analysis requirement for the operating condition of the production system;

extracting information of the components/component relationships characterized by the at least one target keyword required for analyzing the operating condition from data of at least one stage of a product life cycle of the production system;

and generating a semantic relation between the at least two components according to the information of the component/component relation represented by the at least one target keyword based on a semantic coupling rule, wherein the semantic coupling rule is used for stipulating a condition which is satisfied by the information of the component/component relation represented by the at least one target keyword when the semantic relation exists between the at least two components, and the semantic coupling rule is used for defining a relation which is required to be satisfied in a real environment when the semantic relation exists between the at least two components.

2. The method of claim 1, wherein the at least one target keyword comprises at least one first keyword for describing a relationship between the at least two components, and wherein generating a semantic relationship between the at least two components from the at least one target keyword based on a semantic coupling rule comprises:

3. The method of claim 1, wherein the at least one target keyword comprises a second keyword having an attribute for each of the at least two components for describing a value of the attribute, and the generating the semantic relationship between the at least two components according to the at least one target keyword based on a semantic coupling rule comprises:

4. The method of any one of claims 1 to 3, wherein extracting at least one target keyword required for analyzing the operating condition from data of at least one stage of a product lifecycle of the production system comprises:

5. The method of any of claims 1-3, further comprising, prior to generating a semantic relationship between the at least two components from the at least one target keyword based on a semantic coupling rule:

6. The method of any of claims 1-3, wherein the semantic relationships include at least one of physical connection relationships, dependency relationships, and control relationships.

7. A semantic relationship generation apparatus for generating a semantic relationship between at least two components in a production system, wherein the semantic relationship is used for analyzing a type of operation of the production system, comprising:

a target keyword extraction module (71) for extracting information of the component/component relation represented by at least one target keyword required for analyzing the operation condition from data of at least one stage of the product life cycle of the production system; the at least one target keyword is determined according to analysis requirements for the operating conditions of the production system;

a semantic relation generating module (72) for generating a semantic relation between the at least two components according to the information of the component/component relation represented by the at least one target keyword extracted by the target keyword extracting module (71) based on a semantic coupling rule, wherein the semantic coupling rule is used for specifying a condition that the information of the component/component relation represented by the at least one target keyword should satisfy when the semantic relation exists between the at least two components, and the semantic coupling rule is used for defining a relation which needs to be satisfied in a real environment when the semantic relation exists between the at least two components.

8. The device according to claim 7, wherein the at least one target keyword extracted by the target keyword extraction module (71) comprises at least one first keyword for describing a mutual relationship between the at least two components, the semantic relationship generation module (72) being particularly configured to:

9. The device according to claim 7, wherein the at least one target keyword extracted by the target keyword extraction module (71) includes a second keyword having an attribute for each of the at least two components for describing a value of the attribute, and the semantic relationship generation module (72) is specifically configured to:

10. The device according to any one of claims 7 to 9, wherein the target keyword extraction module (71) is specifically configured to:

11. The apparatus according to any of claims 7 to 9, further comprising a semantic coupling rule generating module (83) configured to generate the semantic coupling rule according to the analysis requirement of the operating condition before the semantic relationship between the at least two components is generated by the semantic relationship generating module (72) according to the at least one target keyword based on the semantic coupling rule.

12. The apparatus of any of claims 7-9, wherein the semantic relationship generated by the semantic relationship generation module (72) comprises at least one of a physical connection relationship, a dependency relationship, and a control relationship.

13. A semantic relationship generation apparatus for generating a semantic relationship between at least two components in a production system, wherein the semantic relationship is used to analyze a type of operation of the production system, the apparatus comprising:

at least one memory (90) for storing computer instructions;

at least one processor (91) for invoking the computer instructions to perform the method of any of claims 1-6.

14. A computer readable medium having stored thereon computer instructions which, when executed by a processor, cause the processor to perform the method of any of claims 1-6.