CN114282360A - Ontology-driven workshop unsafe state semantic reasoning method under digital twin environment - Google Patents
Ontology-driven workshop unsafe state semantic reasoning method under digital twin environment Download PDFInfo
- Publication number
- CN114282360A CN114282360A CN202111534942.4A CN202111534942A CN114282360A CN 114282360 A CN114282360 A CN 114282360A CN 202111534942 A CN202111534942 A CN 202111534942A CN 114282360 A CN114282360 A CN 114282360A
- Authority
- CN
- China
- Prior art keywords
- unsafe
- workshop
- ontology
- semantic
- unsafe state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention provides a semantic reasoning method for an unsafe state of a workshop driven by a body under a digital twin environment. Firstly, establishing a semantic ontology model of unsafe states of workshop production sites through a web ontology language (OWL), and establishing inference rules of unsafe states of different types of workshops by using a Semantic Web Rule Language (SWRL); secondly, encoding the semantic ontology of the unsafe state of the established workshop production site by using an ontology editor Prot g; secondly, simulating the unsafe state virtual scene of the twin workshop in Unity 3D, and recording a high-fidelity simulation animation into a video serving as a data set source for subsequent target detection; then, identifying an example of an unsafe state in a workshop production field by using an example segmentation algorithm, and mapping the example of the unsafe state into an example of the established ontology; and finally, executing a reasoning rule by using a reasoning engine to realize automatic reasoning of the potential dangerous state and relevant information of the workshop production field.
Description
Technical Field
The invention relates to the technical field of digitization and intelligence of production management of a manufacturing workshop, in particular to a semantic reasoning method for unsafe states of a workshop driven by a body under a digital twin environment.
Background
The workshop is a basic unit for production and operation of manufacturing enterprises and consists of people, equipment, materials, environment and other factors. The safety management of the workshop production field is the basis for ensuring that all workshop production activities are smoothly carried out, and the traditional workshop production field safety management work is mainly carried out by human experience and is assisted by safety prompts. For example, the safety of personnel, equipment, production lines, environment and the like is ensured by means of camera monitoring, regular patrol of security personnel, field safety warning slogans, safety doors, safety equipment setting such as gratings and the like. However, in a production workshop of a complex product, because parts are complex, the process is complex, the working procedures are various, the number of personnel is large, the environment is complex, the environment of part of the workshop is severe, and toxic, harmful, flammable and explosive substances exist, so that unsafe factors on the production site of the workshop are complex and variable. For example, conditions that may cause a hazard to occur in an automotive welding shop include: the method comprises the following steps that a worker mistakenly enters a safe working range of the welding robot, the worker does not wear a safety helmet and wear a tool according to the regulations, arc light and laser are used in welding, the skin of welding operation is exposed, a mobile phone is played and chatted for a long time, climbing operation does not operate according to the specifications, foreign matters break into a production site, and the like.
For the management and control of the unsafe state of the workshop site, the conventional workshop production site safety management mode based on the subjective experience of people has the possibility of misjudgment, misjudgment and missed judgment, and although the occurrence of partial safety accidents can be reduced by establishing a safety management working group, strengthening safety training and education, adding a safety inspector, additionally installing a monitoring camera, increasing punishment strength and other measures, the occurrence of the misjudgment, the missed judgment and the like of the unsafe state of the workshop production site mainly based on the subjective experience of people cannot be fundamentally solved. In addition, the fault tolerance rate of the extreme production environment is low, so that the production safety training is high in cost and risk, and the input and output are low.
The emergence of the Digital Twin (DT) provides a way for solving the problems, the Digital Twin technology emphasizes the establishment of a Digital object equivalent to an actual object, and the virtual-real linkage, virtual control and real control, iterative optimization and intelligent feedback are realized through methods and technologies such as the Internet of things, big data, artificial intelligence and the like. For example, a Digital Twin plant (DTW) implements integration and fusion of a physical plant, a virtual plant, and a plant service system by bidirectional mapping and real-time interaction between the physical plant and the virtual plant, and implements a new plant operation mode of plant production element management, production activity planning, production process control, and the like.
Aiming at the problems, the invention provides a body-driven semantic reasoning method for unsafe states of a workshop under a digital twin environment, which comprises the steps of establishing a semantic body model and a reasoning rule of unsafe states of a workshop production field according to the workshop field of a physical space; on the basis, a twin workshop of a virtual space is used for simulating a vivid scene of an unsafe state on the workshop site, a video is recorded as a data set source, and an ontology instance set required by semantic reasoning is automatically identified from the data set by combining an instance segmentation algorithm; and finally, executing a predefined inference rule by using an inference engine, and automatically inferring the type of unsafe states of a workshop production field, the reasons for the unsafe states, involved personnel, the hazards brought by the unsafe states and the like, so that the defects of a workshop production field safety management mode based on human subjective experience, high cost and great risk of safety training caused by extreme production environments are overcome, and irrecoverable production safety accidents are avoided.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a semantic reasoning method for an unsafe state of a workshop driven by a body in a digital twin environment, aiming at the defects in the prior art, the method comprises two large stages, wherein the first stage is a semantic body modeling stage for the unsafe state of a workshop production site, the second stage is a semantic reasoning stage for the unsafe state based on deep learning in the digital twin workshop, and the specific steps comprise:
s1, modeling the semantic ontology of unsafe states in the workshop production field:
s11: establishing a concept set;
s12: establishing a relation set;
s13: establishing an attribute set;
s14: establishing an axiom;
s15: establishing an inference rule;
s16: body coding;
s2: carrying out semantic reasoning on unsafe states based on deep learning in a workshop through a digital twin method:
s21: simulating the virtual scene of the unsafe state of the twin workshop;
s22: reading a video of a virtual scene in an unsafe state of the twin workshop as a data set;
s23: identifying instances of unsafe conditions at a production site of the plant based on instance separation for deep learning;
s24: mapping the identified instances to an instance set of an unsafe state ontology in a workshop production field;
s25: and (4) automatic reasoning, identification of unsafe state types, production reasons, potential dangers, involved personnel and equipment and a danger early warning processing method.
Further, step S11 specifically includes: classifying and sorting unsafe states of a workshop production site into a dangerous area entering class, an irregular wearing class, an unsafe behavior class, a dangerous substance leakage class and a man-machine interaction safety class, subdividing according to specific conditions, and establishing a concept set of a semantic ontology of the unsafe states of the production site; step S12 specifically includes: analyzing semantic relations among different concepts of unsafe states of a workshop production site, defining the semantic relations by using object attributes in an Ontology modeling Language (OWL) and establishing a semantic relation set of an unsafe state Ontology of the workshop production site; step S13 specifically includes: defining data type attributes of all concepts related to the unsafe state semantic ontology in the workshop production field by using the data type attributes in the OWL, and establishing an attribute set; step S14 specifically includes: constraining concepts, relations and attributes in the unsafe state ontology of the workshop production site by using an axiom, wherein the axiom of the unsafe state of all the workshop production sites forms an axiom set of the unsafe state semantic ontology model of the workshop production site; step S15 specifically includes: defining conditions required by automatic reasoning in detail by using Semantic network Rule Language (SWRL), wherein reasoning rules of all unsafe states form a reasoning Rule set of a workshop production field unsafe state Semantic ontology model; step S16 specifically includes: and encoding a concept set, a relation set, an attribute set, an axiom set and an inference rule set of the semantic ontology of unsafe states of the workshop production site established in the steps S11 to S15 by using the ontology editor Prot g.
Further, step S21 specifically includes: in a digital twin workshop of a virtual space, simulating an unsafe state virtual scene of the twin workshop by using Unity 3D; step S22 specifically includes: recording the vivid simulation animation into a video as a data set source for subsequent target detection; step S23 specifically includes: marking, training and testing the data set in the S21 by using a target detection algorithm and an example segmentation algorithm based on deep learning, taking a video shot by workshop monitoring as a detection object, and identifying a concept example and an attribute example required by a semantic ontology for producing an unsafe state on site in a workshop in a physical space; step S24 specifically includes: establishing mapping from the instances identified in the step S23 to the semantic ontology of unsafe states in the workshop production field, and completing instantiation of the semantic ontology of unsafe states in the workshop production field; step S25 specifically includes: and (4) executing the defined inference rule by the S15 on the instanced unsafe state ontology instance of the workshop production field by using a rule inference engine, and automatically inferring the potential danger of the workshop production field, the involved personnel and the corresponding processing method.
Further, the semantic relationship set of the on-site unsafe state ontology generated in the workshop in step S12 specifically includes: having a subtype is represented by "hasSubType," e.g., "wear irregular class" having subtypes "no helmet", "no glove", "no goggles", "no tooling"; the 'unsafe behaviors' have sub-types 'run and jump in the workshop', 'fall', 'use mobile phone for a long time' and 'chat for a long time'.
By adopting the technical scheme, the invention can bring the following beneficial effects:
1) establishing a workshop production field unsafe state ontology model, solving the workshop production field unsafe state by a calculation mechanism, reasoning out potential dangers, reasons for generating the unsafe state, related personnel and corresponding processing methods according to semantic relations of the states, and the like, so that the defects of misjudgment, misjudgment and missed judgment existing in the traditional workshop production field safety management mode based on human subjective experience can be overcome;
2) the method has the advantages that the examples required by the unsafe state ontology model of the workshop production site are automatically identified by using an example segmentation algorithm, so that the method is suitable for different production workshops, meanwhile, the error probability of the manually identified examples is reduced, and the cost of the manually identified examples is saved;
3) in a twin workshop of a virtual space, 3DMax and Unity 3D are utilized to establish a vivid scene of an unsafe state on a simulation workshop site, the real unsafe scene on the workshop site is replaced and serves as a data set source of an example segmentation algorithm, and the problems of high production safety training cost, high risk and low investment and output caused by low fault tolerance rate of an extreme production environment are solved.
Drawings
FIG. 1 is a general flow diagram of the present invention;
FIG. 2 is a block diagram of the system of the present invention;
FIG. 3 is an example segmentation algorithm flow based on a digital twin plant;
fig. 4 is a framework for reasoning using Prot g and a reasoning engine.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and the application background of the embodiments of the present invention: in a production site of a manufacturing workshop, a plurality of unsafe states causing production accidents exist, including abnormal behaviors such as running, alarming, falling, chatting, playing mobile phones for a long time, fatigue and the like of workers in the production site, the workers do not wear safety helmets, gloves, tools, goggles and the like according to regulations and are worn irregularly, people and objects which are not in the workshop break into the production site, open flame, harmful gas, liquid and the like are not found and processed in time, the unsafe states are potential factors causing the production accidents, the generation reasons, related personnel, caused hazards and the like need to be found in time, safety managers are informed to carry out early warning and processing, and irrecoverable safety production accidents are avoided. As shown in fig. 1, the invention provides a semantic reasoning method for unsafe states of a workshop based on example segmentation under digital twin, which comprises two major stages, wherein the first stage is a semantic ontology modeling stage for unsafe states in a workshop production site, and the second stage is a semantic reasoning stage for unsafe states based on deep learning in a digital twin workshop, and the specific steps are as follows:
s11, modeling the semantic ontology of unsafe states in the workshop production field, and firstly establishing a concept set: aiming at a specific workshop production field, such as an automobile body-in-white welding workshop, a machining workshop and the like, analyzing an actual production flow, and investigating a production safety state to obtain classification, a generation reason, a corresponding process flow, work activities, a distribution area, related physical objects and a corresponding processing mode of an unsafe state, wherein in the automobile body-in-white welding workshop, a loading area of a certain welding workstation has the risk that a loading person enters a working space of a welding robot and is collided by a mechanical arm, the type of the unsafe state belongs to 'invasion danger', specifically 'enters a dangerous area', related ontology concepts comprise the loading person, the welding robot, the dangerous area and the like, and concepts related to all types of unsafe states form a concept set of an unsafe semantic state ontology model; s12 analyzes and defines semantic relations between different concepts of unsafe states in the workshop production site, establishes a semantic relation set of an unsafe state ontology in the workshop production site, further analyzes binary semantic relations between the concepts of each unsafe state on the basis of S11, lists definition fields and value fields of the semantic relations, forms the semantic relation set of an unsafe state ontology model in the workshop production site, defines phrases required by the binary semantic relations in a format of ' verb + noun ', for example, for unsafe states such as ' enter dangerous area ', the binary semantic relations comprise ' enter into dangerous area ' enterDangerou zone ', generate potential risk ' causePottentialHazard ', generate reasons ' causedBy ', act on ' workkOn ', occur in ' happeddIn ', and the like, two concepts corresponding to entering the dangerous area "enter dangeruszone" are loading personnel and the dangerous area;
s13, further defining data type attributes of all concepts related to the semantic ontology of unsafe states in each workshop production field, representing the data type attributes by using the data type attributes in OWL, and establishing an attribute set: the attribute set refers to a relationship between an ontology concept and a data type, for example, for entering an unsafe state such as "enter DangerusZone", a material person is one of the concepts, the material person has a name attribute whose definition field is a person, and a value field is data of a character string type. The data type attributes of the concepts in all the unsafe state semantic ontologies form an attribute set of the unsafe state semantic ontology model, and since OWL does not provide any predefined data types, the required attributes need to be customized, or data types in other semantic web languages are referred to, such as the data type xsd provided by XML Schema: int or xsd: true, etc.; meanwhile, a Resource Description Framework Schema (RDFS) is used for representing the semantic relationship among the concepts, the relationships and the attributes of the established semantic ontology under the unsafe state in the production field;
s14, establishing an axiom, and constraining concepts, relations and attributes in the unsafe state ontology of the workshop production field by using the axiom: including domain, value domain, notational axiom, transitive axiom, inverse axiom, disjoint axioms, class axioms, attribute axioms, full name constraints, presence constraints, value constraints, cardinality constraints, and data type constraints, etc., for example, with owl: AnnotationProperty represents the Annotation axiom, with owl: taswitveproperty stands for delivery axiom, with owl: invertseof stands for inverse axiom, with owl: oneOf stands for enumeration, with own: the disJointProerty represents non-intersecting axioms and the like, and the axioms of the unsafe states of all workshop production sites form an axiom set of the semantic ontology model of the unsafe states of the workshop production sites;
s15, further, defining inference rules required by automatic inference of unsafe states of a workshop production field by using Semantic Web Rule Language (SWRL): on the basis of establishing a concept set, a relation set and an attribute set, conditions required by automatic reasoning are defined in detail. For example, a rule for deducing the potential risk of a certain loader x entering a dangerous area is that the minimum distance minimumDistance z between a certain loader x and the welding robot weldingRobot y is less than a certain threshold m, when the welding robot needs to be stopped, which can be expressed as SWRL:
worker(？x)^weldingRobot(？y)^hasminimum Distance(？z)^swrlb：lessTan(？z，？m)→isEnteringDangerousZone(？x，true)^stop(？y)
(ii) a In addition, when the loading personnel has the potential risk of entering a dangerous area, a corresponding processing method reasoning rule is established, for example, the movement speed of the mechanical arm is slowed down according to the distance, or the mechanical arm is stopped, and the reasoning rules of all unsafe states form a reasoning rule set of a workshop production field unsafe state semantic ontology model;
s16, encoding the semantic ontology of the unsafe state of the established workshop production site: after defining the nonsecure state semantic ontology model, selecting an ontology modeling tool Prot g to encode the nonsecure state semantic ontology established in the workshop production site in S11, S12, S13 and S14, and simultaneously, using a rule editor in the Prot g to create an inference rule defined by SWRL in S15;
s21, simulating an unsafe state virtual scene of the twin workshop, using 3dMax software to perform mapping rendering on three-dimensional models and scenes of existing personnel, plants, materials, equipment and the like to achieve a vivid effect, importing the scene into Unity 3D, designing a corresponding scene according to a defined unsafe state type, wherein the corresponding scene comprises the setting of parameters such as illumination, materials, textures, colors and the like, and the set scene file is the digital twin model of the unsafe state of the workshop;
s22: reading a video of a virtual scene in an unsafe state of the twin workshop as a data set; simulating the designed unsafe state scene to obtain a vivid simulation animation, and recording the simulation animation as a video serving as a source of a data set segmented by a subsequent example;
s23, marking and training the picture of the video stream of the simulation animation by using a Mask R-CNN (case segmentation method), reading the video stream data by using a monitoring camera installed in a specific workshop production field and a Software Development Kit (SDK) corresponding to the monitoring camera, extracting the picture in the video stream data as a test set for case segmentation, and identifying the case required by the semantic ontology in the workshop production field in an unsafe state from the test set
S24, representing the identified concept instances and attribute instances by using RDF/XML language, establishing mapping of RDF/XML to unsafe state semantic ontologies of workshop production sites, and completing instantiation of unsafe state ontologies of the workshop production sites, wherein OWL can be represented by using all legal RDFS because OWL is established on the RDFS, and can be organized by using RDFS and RDF/XML documents, so that the conversion of RDF/XML documents to OWL can be completed by using programming;
s25, using a rule engine Drools or Jess carried by Prot g to execute inference rules on the instanced unsafe state ontology instances of the workshop production field, and automatically identifying potential dangers, involved personnel and corresponding processing methods.
The system structure of the automatic inference method for the unsafe state of the workshop based on semantic and example segmentation shown in fig. 2 mainly comprises a workshop production site of a physical space, a digital twin workshop of a virtual space, a semantic ontology model for the unsafe state of the workshop production site, and target detection based on deep learning. Different types of unsafe states exist in different workshop production fields for different products, such as dangerous areas for workers, random running of workers, careless falling of workers, incapability of wearing safety helmets by the workers according to regulations, fatigue caused by long-time work of the workers, chatting at the last moment, playing mobile phones, oil leakage of workshop pipelines, intrusion of foreign matters in the workshop fields and the like, the unsafe states are potential dangerous factors causing production safety, and the requirements of specific workshops need to be analyzed and classified; on the basis, establishing a corresponding workshop production field unsafe state semantic ontology model which comprises a concept set, an attribute set, a relation set, an axiom set, an inference rule set and an instance set, and coding the ontology by using an ontology editor, wherein the instance set is intelligently identified by an object detection algorithm based on deep learning through instance segmentation and is converted into an instance corresponding to the production field unsafe state semantic ontology; and after the semantic ontology of the unsafe state in the workshop production field is instantiated, executing the defined inference rule by using an inference engine, and finishing automatic inference of the unsafe state in the production field. The method comprises the steps of establishing a semantic ontology model of the unsafe state of a workshop production field, establishing an inference rule of the unsafe state of the workshop production field, identifying examples related to the unsafe state of the workshop by using an example segmentation algorithm, mapping the examples identified by the example segmentation to the examples of the unsafe state ontology of the workshop production field, and executing the inference rule of the unsafe state of the workshop production field based on an inference engine to realize automatic inference of potential dangers, related personnel, corresponding processing modes and the like.
FIG. 3 shows the specific steps of obtaining an example using the example segmentation algorithm Mask R-CNN: firstly, reading a video of a virtual scene in an unsafe state of a twin workshop, and taking frames from a video stream; further, adding a gray bar to the current frame picture to realize undistorted size adjustment (Resize); further, the preprocessed pictures are placed into a trunk feature extraction network and a feature pyramid, and five effective feature layers are generated; further, the effective characteristic layer passes through a Region candidate Network (Region probable Network) to obtain a suggestion box; further, the obtained suggestion frame is subjected to interception of a shared feature layer through ROI Align and Resize is carried out; further, carrying out classification regression processing on the obtained Resize picture to obtain a prediction frame, wherein the classification regression is divided into two parts, the first part is used for judging whether the suggestion frame contains an object, and the second part is used for adjusting the suggestion frame to obtain the prediction frame; further, the obtained prediction frame is subjected to shared feature layer interception and Resize through ROI Align; finally, processing a Mask R-CNN semantic segmentation model on the Resize picture, performing target detection on a video stream shot in a workshop site by using the processed model, realizing example segmentation, and identifying an example of the concept, relationship and attribute of the semantic ontology in the unsafe state of the workshop production site;
the example set for mapping the identified examples to the unsafe state ontology of the workshop production field is specifically as follows: firstly, storing the instances which are segmented and identified by the instances as RDF/XML documents, wherein the concepts, the relationships, the attributes and the like of the unsafe state ontology of the production site are defined by using an OWL language, so that the next step is to define the mapping relationship of elements between the RDF/XML documents and the OWL documents, and finally converting the elements in the XML documents into concept object instances and attribute instances of the unsafe state ontology of the production site of a workshop according to the mapping relationship;
FIG. 4 presents a framework for implementing automated reasoning of unsafe conditions based on inference engine execution of inference rules of unsafe conditions at a plant production site: and importing the production field unsafe state body and the reasoning engine into the body editor, starting the reasoning engine to execute a reasoning rule, automatically reasoning out the current workshop field unsafe state including potential dangers, involved personnel and corresponding processing methods, and feeding a reasoning result back to the body editor.
The invention provides a semantic reasoning method for an unsafe state of a body-driven workshop in a digital twin environment, and a plurality of methods and ways for realizing the technical scheme are provided. All the components not specified in the present embodiment can be realized by the prior art.
Claims (4)
1. A semantic reasoning method for an unsafe state of a workshop driven by a body under a digital twin environment is characterized by comprising the following steps:
s1, modeling the semantic ontology of unsafe states in the workshop production field:
s11: establishing a concept set;
s12: establishing a relation set;
s13: establishing an attribute set;
s14: establishing an axiom;
s15: establishing an inference rule;
s16: body coding;
s2: carrying out semantic reasoning on unsafe states based on deep learning in a workshop through a digital twin method:
s21: simulating the virtual scene of the unsafe state of the twin workshop;
s22: reading a video of a virtual scene in an unsafe state of the twin workshop as a data set;
s23: identifying instances of unsafe conditions at a production site of the plant based on instance separation for deep learning;
s24: mapping the identified instances to an instance set of an unsafe state ontology in a workshop production field;
s25: and (4) automatic reasoning, identification of unsafe state types, production reasons, potential dangers, involved personnel and equipment and a danger early warning processing method.
2. The ontology-driven semantic reasoning method for unsafe states of workshops under the digital twin environment as claimed in claim 1, wherein step S11 specifically comprises: classifying and sorting unsafe states of a workshop production site into a dangerous area entering class, an irregular wearing class, an unsafe behavior class, a dangerous substance leakage class and a man-machine interaction safety class, subdividing according to specific conditions, and establishing a concept set of a semantic ontology of the unsafe states of the production site; step S12 specifically includes: analyzing semantic relations among different concepts of unsafe states in a workshop production field, defining the semantic relations by using object attributes in an ontology modeling language OWL, and establishing a semantic relation set of an unsafe state ontology in the workshop production field; step S13 specifically includes: defining data type attributes of all concepts related to the unsafe state semantic ontology in the workshop production field by using the data type attributes in the OWL, and establishing an attribute set; step S14 specifically includes: constraining concepts, relations and attributes in the unsafe state ontology of the workshop production site by using an axiom, wherein the axiom of the unsafe state of all the workshop production sites forms an axiom set of the unsafe state semantic ontology model of the workshop production site; step S15 specifically includes: defining conditions required by automatic reasoning in detail by using a semantic network rule language SWRL, and forming a reasoning rule set of a workshop production field unsafe state semantic ontology model by using reasoning rules of all unsafe states; step S16 specifically includes: and encoding a concept set, a relation set, an attribute set, an axiom set and an inference rule set of the semantic ontology of unsafe states of the workshop production site established in the steps S11 to S15 by using the ontology editor Prot g.
3. The ontology-driven semantic reasoning method for unsafe states of workshops under the digital twin environment as claimed in claim 1, wherein step S21 specifically comprises: in a digital twin workshop of a virtual space, simulating an unsafe state virtual scene of the twin workshop by using Unity 3D; step S22 specifically includes: recording the vivid simulation animation into a video as a data set source for subsequent target detection; step S23 specifically includes: marking, training and testing the data set in the S21 by using a target detection algorithm and an example segmentation algorithm based on deep learning, taking a video shot by workshop monitoring as a detection object, and identifying a concept example and an attribute example required by a semantic ontology for producing an unsafe state on site in a workshop in a physical space; step S24 specifically includes: establishing mapping from the instances identified in the step S23 to the semantic ontology of unsafe states in the workshop production field, and completing instantiation of the semantic ontology of unsafe states in the workshop production field; step S25 specifically includes: and (4) executing the defined inference rule by the S15 on the instanced unsafe state ontology instance of the workshop production field by using a rule inference engine, and automatically inferring the potential danger of the workshop production field, the involved personnel and the corresponding processing method.
4. The method for semantic reasoning on unsafe conditions of workshops driven by ontologies in a digital twin environment as claimed in claim 1, wherein the semantic relationship set of on-site unsafe condition ontologies generated in a workshop in step S12 specifically includes: having a subtype is represented by "hasSubType," e.g., "wear irregular class" having subtypes "no helmet", "no glove", "no goggles", "no tooling"; the 'unsafe behaviors' have sub-types 'run and jump in the workshop', 'fall', 'use mobile phone for a long time' and 'chat for a long time'.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111534942.4A CN114282360B (en) | 2021-12-15 | 2021-12-15 | Ontology-driven workshop unsafe state semantic reasoning method under digital twin environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111534942.4A CN114282360B (en) | 2021-12-15 | 2021-12-15 | Ontology-driven workshop unsafe state semantic reasoning method under digital twin environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114282360A true CN114282360A (en) | 2022-04-05 |
| CN114282360B CN114282360B (en) | 2023-04-18 |
Family
ID=80872349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111534942.4A Active CN114282360B (en) | 2021-12-15 | 2021-12-15 | Ontology-driven workshop unsafe state semantic reasoning method under digital twin environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114282360B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115169855A (en) * | 2022-06-29 | 2022-10-11 | 郑州轻工业大学 | Unsafe state detection method based on digital twin workshop mixed data set |
| CN115937626A (en) * | 2022-11-17 | 2023-04-07 | 郑州轻工业大学 | Automatic generation method of semi-virtual data set based on instance segmentation |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113340344A (en) * | 2021-05-27 | 2021-09-03 | 中国建筑第四工程局有限公司 | Hydraulic climbing mould intelligent monitoring control early warning system based on digital twin technology |
| CN113658325A (en) * | 2021-08-05 | 2021-11-16 | 郑州轻工业大学 | Intelligent identification and early warning method for uncertain objects of production line in digital twin environment |
-
2021
- 2021-12-15 CN CN202111534942.4A patent/CN114282360B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113340344A (en) * | 2021-05-27 | 2021-09-03 | 中国建筑第四工程局有限公司 | Hydraulic climbing mould intelligent monitoring control early warning system based on digital twin technology |
| CN113658325A (en) * | 2021-08-05 | 2021-11-16 | 郑州轻工业大学 | Intelligent identification and early warning method for uncertain objects of production line in digital twin environment |
Non-Patent Citations (2)
| Title |
|---|
| 李浩等: "基于数字孪生的复杂产品设计制造一体化开发框架与关键技术", 《计算机集成制造系统》 * |
| 王昊琪等: "基于本体的公理化系统设计语义建模与推理规则", 《机械工程学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115169855A (en) * | 2022-06-29 | 2022-10-11 | 郑州轻工业大学 | Unsafe state detection method based on digital twin workshop mixed data set |
| CN115169855B (en) * | 2022-06-29 | 2023-05-26 | 郑州轻工业大学 | Unsafe state detection method based on digital twin workshop mixed data set |
| CN115937626A (en) * | 2022-11-17 | 2023-04-07 | 郑州轻工业大学 | Automatic generation method of semi-virtual data set based on instance segmentation |
| CN115937626B (en) * | 2022-11-17 | 2023-08-08 | 郑州轻工业大学 | Automatic generation method of paravirtual data set based on instance segmentation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114282360B (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114282360B (en) | Ontology-driven workshop unsafe state semantic reasoning method under digital twin environment | |
| CN104267711B (en) | Running state monitoring and failure diagnosis method for tobacco logistics system | |
| CN102881125B (en) | Alarm monitoring system based on multi-information fusion centralized processing platform | |
| CN111310330B (en) | Method, system, storage medium and computer equipment for designing digital parameters of safety emergency plan | |
| Monferini et al. | A compound methodology to assess the impact of human and organizational factors impact on the risk level of hazardous industrial plants | |
| Ayhan et al. | An association rule mining model for the assessment of the correlations between the attributes of severe accidents | |
| CN116681250A (en) | Building engineering progress supervisory systems based on artificial intelligence | |
| CN116228186A (en) | Ship cabin intelligent operation and maintenance system based on human engineering | |
| CN113627784A (en) | Enterprise asset management intelligent decision-making system based on industrial internet | |
| CN115169855A (en) | Unsafe state detection method based on digital twin workshop mixed data set | |
| Zhang et al. | Data-driven AI emergency planning in process industry | |
| CN114841660A (en) | Enterprise intelligent safety management and control cloud platform based on field information | |
| CN115309913A (en) | Deep learning-based financial data risk identification method and system | |
| CN111062350B (en) | Artificial intelligence based firework recognition algorithm | |
| CN111882141A (en) | Safety management and control system applied to high-energy explosive preparation system | |
| CN111325119B (en) | Video monitoring method and system for safe production | |
| CN115410335A (en) | Building engineering safety monitoring and early warning system based on machine vision and semantic network | |
| Chen et al. | Knowledge graph improved dynamic risk analysis method for behavior-based safety management on a construction site | |
| CN115691044A (en) | Dynamic risk assessment early warning method, system and device | |
| Duan et al. | A science mapping approach-based review of near-miss research in construction | |
| CN113673171B (en) | Docking mechanism digital twin test system based on historical data driving and machine learning and operation method thereof | |
| CN113657788A (en) | Enterprise safety production standardization intelligent supervision system and method | |
| Junjia et al. | Identification and analysis of hoisting safety risk factors for IBS construction based on the AcciMap and cases study | |
| CN113657348A (en) | Intelligent analysis method and system for operation violation behaviors of transformer substation | |
| CN116665305A (en) | Method and system for detecting worker behaviors based on computer vision and knowledge graph |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |