How to handle Industrial Internet operating systems and products

Abstract

The present application belongs to the field of industrial Internet technology, and specifically relates to an industrial Internet operating system and a product processing method. The present application aims to solve the problem that the existing industrial Internet operating system has insufficient data mining and analysis application capabilities, and cannot achieve the integration of heterogeneous data, optimal scheduling of tasks, and coordination of resources. The industrial Internet operating system includes a heterogeneous data integration engine, a digital twin model engine, and a dynamic multi-task scheduling engine. The heterogeneous data integration engine defines and expresses the data of the product to be processed in a unified format to achieve the integration of heterogeneous data. The digital twin model engine monitors the status of the flexible production line when the flexible production line produces the product to be processed. At the same time, when the dynamic multi-task scheduling engine finds problems with the flexible production line during monitoring by the digital twin model engine, it constructs a new flexible production line for continuing to produce the product to be processed, thereby achieving optimal scheduling of tasks and coordination of resources.

Description

Technical Field

[0001] The embodiments of the present application belong to the field of industrial Internet technology, and specifically relate to an industrial Internet operating system and product processing method. Background Art

[0002] With the development of emerging technologies, leading companies in the discrete manufacturing industry and intelligent manufacturing pilot demonstration companies are accelerating their progress towards intelligence and actively carrying out intelligent layout.

[0003] The production transition and industrial upgrading of the discrete manufacturing industry are closely dependent on the development of information technology, and the core of this is the Industrial Internet operating system. The Industrial Internet operating system empowers discrete manufacturing production processes and related industrial products, adding intelligence to industrial products and production systems, effectively improving the intelligence of products and production processes, and meeting the discrete manufacturing industry's requirements for high efficiency, reliability, real-time operation, and environmental protection, ultimately achieving automation and intelligentization of products or processes.

[0004] However, the existing industrial Internet operating system is still in its early stages of development, with insufficient data mining and analysis application capabilities, and is unable to achieve the integration of heterogeneous data, optimal task scheduling, and resource coordination. Summary of the Invention

[0005] In order to solve the above-mentioned problems in the prior art, that is, to solve the problem that the data mining and analysis application capabilities of the industrial Internet operating system in the prior art are insufficient and cannot achieve the integration of heterogeneous data, optimal scheduling of tasks and coordination of resources, the embodiment of the present application provides a processing method for an industrial Internet operating system and product.

[0006] In a first aspect, an embodiment of the present application provides an industrial Internet operating system, including:

[0007] Heterogeneous data integration engine, digital twin model engine, and dynamic multi-task scheduling engine;

[0008] The heterogeneous data integration engine processes the initial production data of the product to be processed according to a preset data format to generate target production data;

[0009] The digital twin model engine obtains the production scheduling information of the product to be processed based on the flexible production line data used to produce the product to be processed and the historical production scheduling information of other products of the same category as the product to be processed; and can also monitor the status of the flexible production line when the flexible production line produces the product to be processed based on the product information of the product to be processed, the production scheduling information, and the target production data, to obtain the status data of the flexible production line;

[0010] When the status data indicates that there is a problem with the flexible production line, the dynamic multi-task scheduling engine constructs a new flexible production line to enable continued production of the product to be processed on the new flexible production line.

[0011] In the preferred technical solution of the above-mentioned industrial Internet operating system, the system further includes:

[0012] Industrial big data knowledge engine and industrial edge intelligent CPS management shell;

[0013] The industrial big data knowledge engine obtains historical product information of the other products;

[0014] The industrial edge intelligent CPS management shell determines the product information of the product to be processed based on the historical product information and the user's functional requirements and / or appearance requirements for the product to be processed.

[0015] In the preferred technical solution of the above-mentioned industrial Internet operating system, the system further includes:

[0016] The scheduling algorithm library stores scheduling algorithms, and the scheduling algorithms are used to implement the scheduling function of the dynamic multi-tasking scheduling engine.

[0017] In a second aspect, an embodiment of the present application provides a product processing method, which is applied to a server in the industrial Internet operating system of the first aspect, and the method includes:

[0018] Determining product information of the product to be processed based on the user's functional requirements and / or appearance requirements of the product to be processed;

[0019] Obtaining production schedule information of the product to be processed through a digital twin model engine based on flexible production line data used to produce the product to be processed and historical production schedule information of other products of the same category as the product to be processed;

[0020] When the product to be processed is produced on the flexible production line according to the product information and the production scheduling information, the initial production data of the product to be processed is processed according to a preset data format by a heterogeneous data integration engine to generate target production data;

[0021] According to the target production data, the flexible production line is monitored by the digital twin model engine to obtain status data of the flexible production line;

[0022] When the status data indicates that there is a problem with the flexible production line, a new flexible production line is constructed through a dynamic multi-task scheduling engine to enable continued production of the product to be processed on the new flexible production line.

[0023] In the preferred technical solution of the above product processing method, the step of determining product information of the product to be processed includes:

[0024] Obtain historical product information of the other products through an industrial big data knowledge engine;

[0025] Based on the historical product information and the demand information of the product to be processed, the product information of the product to be processed is determined through the industrial edge intelligent information-physical system CPS management shell, and the demand information is used to represent the user's functional requirements and / or appearance requirements for the product to be processed.

[0026] In the preferred technical solution of the processing method for the above-mentioned product, the state monitoring of the flexible production line by the digital twin model engine based on the target production data to obtain the state data of the flexible production line includes:

[0027] According to the target production data, the flexible production line is monitored by the digital twin model engine to obtain status monitoring data of the flexible production line;

[0028] According to the status monitoring data, the status of the flexible production line in the remaining production time is predicted by the industrial big data knowledge engine to obtain the status data.

[0029] In a preferred technical solution of the above-mentioned product processing method, before obtaining the production scheduling information of the product to be processed through the digital twin model engine based on the flexible production line data used to produce the product to be processed and the historical production scheduling information of other products of the same category as the product to be processed, the method further includes:

[0030] According to the business logic in the product information and the functions of each physical discrete manufacturing equipment, the flexible production line is constructed through the industrial edge intelligent CPS management shell to connect the business flow and data flow throughout the entire life cycle of the product.

[0031] In a preferred technical solution of the above product processing method, when the product to be processed is produced on the flexible production line according to the product information and the production scheduling information, the method further includes:

[0032] The initial production data of the product to be processed is collected through the industrial edge intelligent CPS management shell.

[0033] In the preferred technical solution of the above product processing method, when the status data indicates that there is a problem with the flexible production line, a new flexible production line is constructed by a dynamic multi-task scheduling engine, including:

[0034] Calling a scheduling algorithm from a scheduling algorithm library through the dynamic multi-tasking scheduling engine;

[0035] Based on the scheduling algorithm, judging whether there is a problem with the flexible production line according to the status data;

[0036] When it is determined that there is a problem with the flexible production line, a new flexible production line is constructed.

[0037] In the preferred technical solution of the above-mentioned product processing method, judging whether there is a problem with the flexible production line based on the status data based on the scheduling algorithm includes:

[0038] Based on the scheduling algorithm, it is determined according to the status data whether the flexible production line has abnormal working conditions and / or equipment failure.

[0039] It will be understood by those skilled in the art that the industrial Internet operating system and product processing method provided in the embodiments of the present application include a heterogeneous data integration engine, a digital twin model engine, and a dynamic multi-task scheduling engine. The heterogeneous data integration engine processes the initial production data of the product to be processed according to a preset data format to generate target production data. The digital twin model engine obtains the production scheduling information of the product to be processed based on the flexible production line data used to produce the product to be processed and the historical production scheduling information of other products of the same type as the product to be processed. Based on the product information, production scheduling information and target production data of the product to be processed, when the flexible production line produces the product to be processed, the digital twin model engine can also monitor the status of the flexible production line and obtain the status data of the flexible production line. When the status data indicates that there is a problem with the flexible production line, the dynamic multi-task scheduling engine constructs a new flexible production line to enable continued production of the product to be processed on the new flexible production line. In an embodiment of the present application, the heterogeneous data integration engine can define and express the data of the products to be processed in a unified format, thereby realizing the integration of heterogeneous data. The digital twin model engine and the dynamic multi-task scheduling engine can monitor the flexible production line so that when there is a problem with the flexible production line, a new flexible production line can be built in time. The flexible production line is used to continue the production of the products to be processed, thereby realizing the optimized scheduling of tasks and the coordination of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The following describes the industrial Internet operating system and product processing method of the present application with reference to the accompanying drawings, which are as follows:

[0041] Figure 1 A schematic diagram of the structure of the industrial Internet operating system provided in an embodiment of the present application;

[0042] Figure 2Another structural diagram of the industrial Internet operating system provided in an embodiment of the present application;

[0043] Figure 3 This is a flow chart of Example 1 of the method for processing a product provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] First, those skilled in the art should understand that these embodiments are merely used to explain the technical principles of this application and are not intended to limit the scope of protection of this application. Those skilled in the art may adjust them as needed to suit specific applications.

[0045] Secondly, it should be noted that in the description of the embodiments of the present application, terms such as "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and does not indicate or imply that the device or component must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present application.

[0046] In addition, it should be noted that in the description of the embodiments of this application, unless otherwise clearly specified or limited, the terms "connected" and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be internal communication between two components. For those skilled in the art, the specific meanings of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0047] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0048] Before introducing the embodiments of the present application, the application background of the embodiments of the present application is first explained:

[0049] In recent years, technologies such as the internet, cloud computing, big data, and artificial intelligence have developed rapidly. Companies across the globe have conducted research on industrial internet operating systems (IIoT), resulting in the development of numerous representative IIoT operating systems. For discrete industries, IIoT operating systems are crucial because they shield the heterogeneity of underlying industrial layers and, with data and industrial mechanism models at their core, provide digital, networked, and intelligent services for the entire lifecycle of discrete manufacturing applications.

[0050] Among them, the industrial Internet operating system of the existing technology mainly includes the platform layer and the application layer. The platform layer includes artificial intelligence, big data, and cloud computing. The application layer includes the user end and the developer end. The platform layer and the application layer are protected by a security protection system. The data information in the security protection system is stored in the database. The security protection system includes data protection and physical protection. Data protection includes a data screening module, a hazard judgment module, a hazard preprocessing module, a manual warning module, and an abnormal hazard processing module. Physical protection includes a visual recognition module and a voice recognition module.

[0051] However, the existing industrial Internet operating system's data mining and analysis application capabilities are insufficient, and there are the following problems:

[0052] (1) The core support of the industrial Internet operating system for discrete industries is the integration of industrial data. However, industrial data involves different stages of the entire life cycle of discrete manufacturing, different business activities, various heterogeneous systems, and various heterogeneous data types, which leads to the problem of heterogeneous data integration.

[0053] (2) The core application function of the industrial Internet operating system in discrete industries is the coordinated regulation of resources. However, due to the influence of factors such as the large-scale growth of industrial Internet resource scheduling tasks, cross-organizational collaboration, and dynamic and changing environments, there is a problem of difficulty in coordinated scheduling of dynamic multiple tasks.

[0054] To sum up, traditional automation system solutions still occupy the mainstream position in the market. The industrial Internet operating system that provides overall solutions for the intelligent upgrading needs of the discrete manufacturing industry is still in its early stages of development. There are not many mature industrial Internet operating systems that can be promoted and replicated in the industry, and their scale is limited.

[0055] To address the above issues, the present application provides an industrial Internet operating system, which includes a heterogeneous data integration engine, a digital twin model engine, and a dynamic multi-task scheduling engine. The heterogeneous data integration engine can define and express the data of the product to be processed in a unified format, thereby realizing the integration of heterogeneous data. The digital twin model engine can monitor the status of the flexible production line when the flexible production line is producing the product to be processed. At the same time, the dynamic multi-task scheduling engine can build a new flexible production line when the digital twin model engine monitors and finds problems with the flexible production line, so that the product to be processed can continue to be produced on the new flexible production line, thereby achieving optimized task scheduling and resource coordination.

[0056] The technical solution of the present application is described in detail below through specific embodiments.

[0057] It should be noted that the following specific embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0058] Figure 1 A structural diagram of the industrial Internet operating system provided in the embodiment of this application. Figure 1 As shown, the industrial Internet operating system may include: a heterogeneous data integration engine, a digital twin model engine, and a dynamic multi-task scheduling engine.

[0059] Among them, the heterogeneous data integration engine processes the initial production data of the product to be processed according to the preset data format to generate target production data.

[0060] Optionally, the above processing may be: automatically formulating a conversion rule based on the above initial production data, converting the format of the above initial production data into a preset data format through the conversion rule, thereby generating target production data.

[0061] It should be understood that the preset data format can be pre-set by relevant staff according to actual needs, and the embodiments of the present application do not impose specific restrictions on this.

[0062] Optionally, the heterogeneous data integration engine can also build a flexible production line in the discrete manufacturing industry during the design, production, service and other business processes of the products to be processed, based on the business logic in the product information of the products to be processed and the functions of each physical discrete manufacturing equipment, so as to connect the business flow and data flow of the products to be processed throughout their entire life cycle.

[0063] The product information is determined based on the user's functional requirements and / or appearance requirements for the product to be processed.

[0064] Optionally, the heterogeneous data integration engine can also build a virtual flexible production line based on the above business logic through relevant models to achieve model semantic consistency conversion.

[0065] Optionally, the above-mentioned related models can be stored in a model resource library, and the heterogeneous data integration engine can call the above-mentioned related models through the industrial edge intelligent information-physical system (Cyber-Physical Systems, CPS) management shell.

[0066] Among them, the digital twin model engine obtains the production scheduling information of the product to be processed based on the flexible production line data used to produce the product to be processed and the historical production scheduling information of other products of the same type as the product to be processed.

[0067] Optionally, the digital twin model engine can also monitor the status of the flexible production line and obtain status data of the flexible production line when the flexible production line produces the product to be processed based on the product information, production scheduling information and target production data of the product to be processed.

[0068] Optionally, the digital twin model engine can also build a full-process digital twin model of the product to be processed based on the virtual flexible production line constructed by the above-mentioned heterogeneous data integration engine and the historical production scheduling information of other products of the same type as the product to be processed, as well as the digital twin mechanism model in the scenario-based mechanism model library, and realize virtual-reality mapping through the industrial edge intelligent CPS management shell, so that the digital twin model engine can predict the production planning, working conditions, equipment status, etc. of the product to be processed through the above-mentioned full-process digital twin model, thereby determining the production scheduling plan and performing status monitoring of the flexible production line.

[0069] When status data indicates a problem with a flexible production line, the dynamic multi-task scheduling engine can create a new flexible production line to continue production of the product being processed. This allows the dynamic multi-task scheduling engine to allocate production resources to meet changing production needs and the demands of multiple customers for large-scale, personalized product customization.

[0070] An embodiment of the present application provides an industrial Internet operating system, which includes a heterogeneous data integration engine, a digital twin model engine, and a dynamic multi-task scheduling engine. The heterogeneous data integration engine processes the initial production data of the product to be processed according to a preset data format to generate target production data. The digital twin model engine obtains the production scheduling information of the product to be processed based on the flexible production line data used to produce the product to be processed and the historical production scheduling information of other products of the same type as the product to be processed. When the flexible production line produces the product to be processed based on the product information, production scheduling information and target production data of the product to be processed, the digital twin model engine can also monitor the status of the flexible production line and obtain the status data of the flexible production line. When the status data indicates that there is a problem with the flexible production line, the dynamic multi-task scheduling engine constructs a new flexible production line to enable continued production of the product to be processed on the new flexible production line. In an embodiment of the present application, the heterogeneous data integration engine can define and express the data of the products to be processed in a unified format, thereby realizing the integration of heterogeneous data. The digital twin model engine and the dynamic multi-task scheduling engine can monitor the flexible production line so that when there is a problem with the flexible production line, a new flexible production line can be built in time. The flexible production line is used to continue the production of the products to be processed, thereby realizing the optimized scheduling of tasks and the coordination of resources.

[0071] based on Figure 1In the illustrated embodiment, in some embodiments, the industrial Internet operating system may also include: an industrial big data knowledge engine and an industrial edge intelligent CPS management shell.

[0072] Among them, the industrial big data knowledge engine obtains historical product information of other products.

[0073] Optionally, the industrial big data knowledge engine can also obtain historical production scheduling information of other products of the same category as the product to be processed.

[0074] Optionally, the industrial big data knowledge engine can also obtain the status monitoring data of the flexible production line based on the status monitoring of the flexible production line by the digital twin model engine, predict the status of the flexible production line in the remaining production time, and obtain status data.

[0075] Optionally, the industrial big data knowledge engine can also store relevant data processed by the heterogeneous data integration engine, where the relevant data can be target production data, as well as relevant data such as product design, production, and service processes.

[0076] Optionally, in order to predict the status of the flexible production line during the remaining production time, the industrial big data knowledge engine may include the following modules:

[0077] Target domain data enhancement knowledge representation module, target domain self-supervised learning knowledge representation module, and cross-domain deep transfer learning module based on multi-task meta-learning.

[0078] Among them, the target domain data enhancement knowledge representation module can use signal domain conversion, adversarial learning, automatic data augmentation and other technologies to perform data enhancement to address problems such as insufficient and low-quality historical industrial data samples of the products to be processed, generate enhanced samples that integrate cross-domain knowledge, and use them in the transfer learning model of cross-domain samples, which facilitates the classification (fault diagnosis, anomaly detection) and prediction (life prediction, inventory prediction) problems of complex industrial scenarios.

[0079] The target domain self-supervised learning knowledge representation module can perform self-supervised knowledge representation on the industrial heterogeneous data collected in real time during the product design and manufacturing process, combine the industrial multi-source data representations in the time domain and space domain, and use supervised learning to obtain deep decoupling of multi-source heterogeneous data with invariance, equivariance, and mobility to adapt to dynamically changing complex industrial tasks.

[0080] The cross-domain deep transfer learning module based on multi-task meta-learning can perform single-task and multi-task migration and multiple auxiliary task selection mechanism design based on the transfer learning mechanism of the meta-learning mechanism according to the product to be processed and the relevant historical data of the product to be processed, so as to facilitate classification and accurate prediction in complex industrial scenarios, and ultimately realize the deep cross-domain knowledge transfer of new tasks.

[0081] Among them, the industrial edge intelligent CPS management shell determines the product information of the products to be processed based on historical product information and demand information of the products to be processed.

[0082] The above-mentioned demand information may be the user's demand for the product to be processed, including functional requirements and / or appearance requirements.

[0083] Optionally, the Industrial Edge Intelligent CPS management shell can also collect initial production data of the products to be processed. The above-mentioned collection and processing can include: accessing physical discrete manufacturing heterogeneous resources and collecting all-round data of discrete manufacturing equipment in real time.

[0084] The intelligent CPS management shell enables semantic understanding, operation, and scheduling of discrete manufacturing physical equipment, products, and service resources. Intelligent perception networks enable automated resource perception and intelligent adaptive matching of heterogeneous device protocols. Furthermore, adaptive mapping between entities and models is achieved through automatic labeling, identification, classification, retrieval, and compilation optimization of models in the model resource library.

[0085] In the above embodiment, the product information of the product to be processed is determined through the intelligent CPS management shell and the industrial big data knowledge engine, thereby obtaining a design solution suitable for the user, laying the foundation for subsequent product processing and improving the accuracy of processing.

[0086] based on Figure 1 In some embodiments, the industrial Internet operating system may further include:

[0087] The scheduling algorithm library stores scheduling algorithms, which are used to implement the scheduling function of the dynamic multi-tasking scheduling engine.

[0088] In this embodiment, the scheduling algorithm library can store scheduling algorithms for easy calling by the dynamic multi-task scheduling engine. After the dynamic multi-task scheduling engine calls the scheduling algorithm, it can optimize the design of the flexible production line and build a new flexible production line suitable for the business goals, thereby achieving efficient and accurate scheduling under multiple tasks.

[0089] Figure 2 Another structural diagram of the industrial Internet operating system provided in the embodiment of this application. Figure 2 As shown, the industrial Internet operating system includes:

[0090] Discrete industry applications, industrial application mobile software (application, APP), core components, basic common components, discrete manufacturing resources, safety protection system and standard identification system.

[0091] Among them, discrete industry applications include user interaction, R&D innovation, precision sales, collaborative procurement, intelligent manufacturing, smart logistics, and intelligent services, among other applications throughout the entire life cycle of discrete manufacturing products.

[0092] Among them, industrial application APPs include interactive customization industrial APPs, development and design industrial APPs, precision sales industrial APPs, modular procurement industrial APPs, intelligent manufacturing industrial APPs, intelligent logistics industrial APPs, and intelligent service industrial APPs, thereby realizing personalized customization, networked collaboration, intelligent production and service extension.

[0093] Among them, the core components include the industrial cloud layer and the industrial edge layer. The industrial cloud layer includes low-code rapid construction tools for industrial Internet applications, scenario-based mechanism model library, industrial engine, scheduling algorithm library and big data lake. The industrial edge layer includes model resource library, industrial edge intelligent CPS management shell and interface protocol library.

[0094] Among them, the low-code rapid construction tool for industrial Internet applications is based on other core component libraries. It is aimed at the personalized customization and networked collaborative application needs of discrete industries and integrates core components such as industrial edge intelligence, model resource library, interface protocol library, big data lake, industrial engine, and scenario-based mechanism model library to build a rapid development tool for industrial Internet application code. This tool covers the model resource library and compiler of cloud-native applications such as software user interface (UI), services, control, entities, processes, and rules. Based on graphical low-code rapid development technology, it realizes the rapid construction of model-driven cloud-native applications, reduces the difficulty of building industrial apps, and provides underlying architectural support for the developer community and the formation of a large-scale customized industrial chain ecosystem.

[0095] Among them, the scenario-based mechanism model library stores digital twin mechanism models for different application scenarios of discrete manufacturing, which facilitates scheduling applications.

[0096] Among them, the industrial engine includes a digital twin model engine and a dynamic multi-task scheduling engine.

[0097] Among them, the big data lake includes a heterogeneous data integration engine, a big data lake governance tool, and an industrial big data knowledge engine.

[0098] Among them, the industrial edge layer includes the model resource library, the industrial edge intelligent CPS management shell and the interface protocol library.

[0099] Optionally, the model resource library extracts a subset of common information model elements from the classification structure of production equipment, products, and services to establish a unified multidimensional semantic ontology model for all types of equipment, products, and services. Furthermore, the model resource library uses semantic analysis technology to establish associations between various resource models and further instantiate these associations, providing a retrieval foundation for full-factor resource perception and adaptive intelligent matching within the industrial edge intelligent CPS management shell.

[0100] Optionally, the interface protocol library provides corresponding interface protocols for discrete manufacturing equipment, product and service resource semantic models to facilitate intelligent perception of all-factor resources within the factory network.

[0101] In this embodiment, in view of the polymorphic, heterogeneous and mixed characteristics of components such as resource access, data integration and task scheduling in the industrial Internet platform, the core components formulate a field-oriented component service assembly mechanism, adopt standardized industrial microservice technology, and establish unified interface standards and gateways between the logical execution modules of the industrial Internet operating system components. Through the component assembly and orchestration model, a core component library based on microservices and container technology is formed.

[0102] Among them, basic common components include basic component libraries such as container services, load balancing, cloud storage, and content delivery networks (CDN); development component libraries such as the general term for processes, methods, and systems (a combination of Development and Operations, DevOps), microservice governance, function services, open application programming interfaces (OpenAPI), etc.; middleware components such as databases, message queues, caches, and search engines; operation and maintenance component libraries such as monitoring and early warning, log services, cloud backup, and off-site disaster recovery, etc., and also include related components such as servers, storage devices, network devices, security devices, network bandwidth, and encryption machines.

[0103] Among them, discrete manufacturing resources include CNC bed, industrial robots, automated guided vehicles (AGV), sensors, industrial switches, cameras, augmented reality (AR) glasses, testing equipment and other resources.

[0104] In the embodiments of this application, in response to the demand for access to all-factor resources of equipment, products, and services in the discrete manufacturing industry, and addressing the problem of self-adaptive modeling of the semantics of all-factor resources, a unified semantic model of all-factor resources for discrete manufacturing equipment, products, and services is constructed, along with a model resource library. Equipment is divided into industrial robots, CNC machine tools, AGVs, and testing equipment according to basic types; products are divided into terminal intelligent consumer products, intelligent electromechanical products, and others; and services are divided into R&D and design services, production and manufacturing services, business management services, and after-sales maintenance services according to their full life cycle. Based on classification, common information is extracted, and a multi-feature semantic information annotation method and a semantic information cross-classification retrieval and matching method are integrated. By extracting a subset of common information model elements from the equipment, product, and service classification structure, a unified multidimensional semantic ontology model of various types of equipment, products, and service resources is established, forming a model resource library. An efficient, intelligent, and adaptive adaptation method for industrial heterogeneous protocols is constructed, and core components such as the industrial edge intelligent CPS management shell, model resource library, and interface protocol library are created. This effectively addresses the difficulty in accessing heterogeneous factor resources in existing technologies due to the wide variety, huge scope, and huge differences of discrete manufacturing resources.

[0105] Furthermore, in response to the needs of spatial sharing and integration of heterogeneous big data in the discrete manufacturing industry, and to the problem of cross-domain knowledge migration of strongly heterogeneous data, we conduct spatiotemporal multi-scale business process data integration, multi-dimensional semantically associated data space implicit knowledge fusion, and deep migration of industrial heterogeneous cross-domain knowledge; in response to the needs of multi-task scheduling optimization of discrete manufacturing processes, and to the problem of precise control of high-dynamic task uncertainty, we propose digital twin modeling of discrete manufacturing scheduling processes to achieve precise optimization scheduling of multi-tasks based on digital twin predictions.

[0106] The embodiments of the present application propose a unified semantic scalable model of all-factor resources of discrete manufacturing equipment, products and services based on multi-dimensional semantic modeling, establish an intelligent CPS management shell for adaptive adaptation of industrial heterogeneous protocols, and break through the bottleneck of the difficulty in unified standardized modeling and adaptive access of discrete manufacturing resources. At the same time, it breaks through the bottleneck of the difficulty in representing and migrating deep-level knowledge in industrial heterogeneous data space, proposes a multi-dimensional associated heterogeneous data space implicit knowledge representation fusion method based on high-order tensor space modeling, and utilizes the industrial knowledge cross-domain transfer learning method based on multi-task meta-learning to form core components such as business process heterogeneous data integration engine and industrial big data knowledge engine. Furthermore, it proposes multi-task intelligent optimization scheduling based on virtual-reality fusion prediction, breaks through the bottleneck limitation of the difficulty in accurately controlling dynamic multi-tasks in the complex and uncertain environment of discrete manufacturing, and forms core components such as industrial Internet digital twin model engine and dynamic multi-task scheduling engine.

[0107] By constructing an industrial Internet operating system for the discrete manufacturing industry, the embodiments of the present application can assist enterprises in establishing a unified business operation platform based on existing factor resources. The functions and data of all software and equipment are scheduled through this unified platform, which will break through the core bottleneck problems of the industrial Internet operating system, form new systems, new components, and new applications, realize continuous improvement and iterative optimization of factories, and enable high-quality development of the discrete manufacturing industry.

[0108] The embodiments of this application focus on the three bottleneck challenges of "incomplete connection" of industrial Internet factor resources, "insufficient integration" of heterogeneous data, and "inaccurate control" of collaborative scheduling. Combined with the development trend of the new generation of information technology, the edge intelligence, big data space, deep transfer learning, digital twins, reinforcement learning scheduling decision-making and other technologies are integrated and innovated with the characteristics of the discrete industry industrial Internet, breaking through the three key problems of self-adaptation of semantic modeling of all factor resources, deep cross-domain knowledge transfer of strong heterogeneous data, and precise control of high dynamic task uncertainty, forming intelligent, standardized, and autonomous core technology components, and creating a discrete industry industrial Internet operating system with comprehensive intelligent connection, deep intelligent integration, and precise intelligent control. The present invention has important scientific value and application value for the breakthrough and development of new theories and technologies in the frontier field of industrial Internet, especially discrete industry industrial Internet operating systems. The industrial Internet operating system will become the core common support platform for industrial Internet industry applications, and will provide core technology and system support for the transformation and upgrading of discrete manufacturing enterprises, and increase costs, efficiency and profitability.

[0109] Figure 3 This is a flow chart of the first embodiment of the method for processing the product provided in the embodiment of this application. Figure 3 As shown, the processing method of the product is applied to the server in the industrial Internet operating system described in any of the above embodiments, and the processing method of the product may include the following steps:

[0110] S301: Determine product information of the product to be processed based on the user's functional requirements and / or appearance requirements of the product to be processed.

[0111] In actual applications, the server needs to obtain the customer's personalized needs for the product to be processed, and then determine the product information of the product to be processed according to the personalized needs.

[0112] Optionally, the user may input personalized requirements through the front-end device of the industrial Internet operating system, and the above-mentioned server responds to the user's input operation to obtain the personalized requirements of the product to be processed input by the user.

[0113] Optionally, the above-mentioned input operation can be a voice input operation, a text input operation, and a click input operation of a related control, etc., which can be determined according to actual conditions. The embodiment of the present application does not limit the specific input operation method.

[0114] S302. Obtain production scheduling information of the product to be processed through a digital twin model engine based on the flexible production line data used to produce the product to be processed and the historical production scheduling information of other products of the same category as the product to be processed.

[0115] In practical applications, after determining the product information of the pending product, the product will enter the production process. However, the large number of pending products places higher demands on flexible production lines and collaborative optimization. Therefore, it is also necessary to determine the production schedule of the pending products to ensure efficient production of the pending products.

[0116] Optionally, the full-process digital twin model in the digital twin model engine can be used to predict the production planning, working conditions, equipment status, etc. of the products to be processed based on flexible production line data and historical production scheduling information, thereby determining the production scheduling plan.

[0117] The flexible production line data may be all-around data of the flexible production line.

[0118] Optionally, the above-mentioned flexible production line data can be obtained in advance through the industrial edge intelligent CPS management shell, and the above-mentioned historical production scheduling information can be mined and migrated from historical production scheduling cases of other equipment through the industrial big data knowledge engine in advance.

[0119] S303. When the product to be processed is produced on the flexible production line according to the product information and the production scheduling information, the initial production data of the product to be processed is processed according to the preset data format through the heterogeneous data integration engine to generate target production data.

[0120] The initial production data may be the production data of each discrete manufacturing device in the flexible production line.

[0121] Optionally, the above processing may be: automatically formulating conversion rules according to the above initial production data through a heterogeneous data integration engine, converting the format of the above initial production data into a preset data format through the conversion rules, thereby generating target production data.

[0122] It should be understood that the preset data format can be pre-set by relevant staff according to actual needs, and the embodiments of the present application do not impose specific restrictions on this.

[0123] Optionally, the initial production data of the above-mentioned products to be processed can be collected through the industrial edge intelligent CPS management shell.

[0124] S304: Based on the target production data, the flexible production line is monitored through the digital twin model engine to obtain the status data of the flexible production line.

[0125] Among them, condition monitoring includes monitoring of working conditions and / or equipment.

[0126] S305. When the status data indicates that there is a problem with the flexible production line, a new flexible production line is constructed through the dynamic multi-task scheduling engine to enable continued production of the product to be processed on the new flexible production line.

[0127] In actual applications, when a problem with the status of a flexible production line is detected through the digital twin model engine, the product to be processed needs to be transferred to other flexible production lines for continued production to avoid affecting the production progress of the product to be processed.

[0128] An embodiment of the present application provides a product processing method, which determines product information of the product to be processed based on the user's functional requirements and / or appearance requirements for the product to be processed, obtains the production scheduling information of the product to be processed through a digital twin model engine based on the flexible production line data used to produce the product to be processed and the historical production scheduling information of other products of the same type as the product to be processed, and when the product to be processed is produced on the flexible production line based on the product information and the production scheduling information, the initial production data of the product to be processed is processed according to a preset data format through a heterogeneous data integration engine to generate target production data, and monitors the status of the flexible production line based on the target production data through the digital twin model engine to obtain the status data of the flexible production line, and when the status data indicates that there is a problem with the flexible production line, constructs a new flexible production line through a dynamic multi-task scheduling engine to enable continued production of the product to be processed on the new flexible production line. This technical solution can be applied to application scenarios such as flexible planning and scheduling, flexible production line reconstruction, working condition monitoring and prediction, and dynamic coordinated scheduling. Through the heterogeneous data integration engine, the data of the products to be processed are defined and expressed in a unified format. When problems occur in the flexible production line, a new flexible production line is constructed through the digital twin model engine and the dynamic multi-task scheduling engine to enable continued production of the products to be processed on the new flexible production line, thereby achieving optimized task scheduling and resource coordination.

[0129] Optional, based on Figure 3 In the embodiment shown, S301 can be implemented by the following steps:

[0130] Through the industrial big data knowledge engine, historical product information of other products is obtained. Based on the historical product information and the user's demand for the products to be processed, the product information of the products to be processed is determined through the industrial edge CPS management shell.

[0131] In an embodiment of the present application, the industrial big data knowledge engine is called to provide knowledge support, and then the industrial edge CPS management shell is called according to the user's needs for the product to be processed to optimize and improve the historical product information, thereby determining the product information of the product to be processed and obtaining the design plan of the product to be processed, thereby improving the matching degree between the produced product to be processed and the user needs.

[0132] Optional, based on Figure 3 In the embodiment shown, S304 can be implemented by the following steps:

[0133] Based on the target production data, the flexible production line is monitored through the digital twin model engine to obtain the status monitoring data of the flexible production line. Based on the status monitoring data, the status of the flexible production line in the remaining production time is predicted through the industrial big data knowledge engine to obtain the status data.

[0134] In an embodiment of the present application, the digital twin model engine can be called during the production execution process to perform real-time status monitoring of the flexible production line, and the industrial big data knowledge engine can be called to predict abnormal working conditions, equipment failures, timing trends, etc., and obtain status data, which can effectively prevent the flexible production line from having abnormalities during the remaining production time and affecting the production work of the products to be processed.

[0135] Optional, based on Figure 3 In the illustrated embodiment, before S302, the product processing method may further include the following steps:

[0136] Based on the business logic in product information and the functions of each physical discrete manufacturing equipment, a flexible production line is built through a heterogeneous data integration engine to connect the business flow and data flow throughout the entire product life cycle.

[0137] In an embodiment of the present application, the heterogeneous data integration engine constructs a flexible production line based on the business logic in the product information and the functions of each physical discrete manufacturing equipment, laying the foundation for the subsequent production of the processed products on the flexible production line and improving production efficiency.

[0138] Optional, based on Figure 3 In the illustrated embodiment, when the product to be processed is produced on a flexible production line based on the product information and the production scheduling information, the product processing method may further include the following steps:

[0139] The initial production data of the products to be processed is collected through the industrial edge intelligent CPS management shell.

[0140] In an embodiment of the present application, the industrial edge intelligent CPS management shell can access physical discrete manufacturing heterogeneous resources, and collect production data of each discrete manufacturing equipment in the flexible production line in real time, thereby obtaining the initial production data of the product to be processed, so that the flexible production line can be monitored subsequently to ensure the smooth production of the product to be processed.

[0141] Optional, based on Figure 3 In the embodiment shown, S305 can be implemented by the following steps:

[0142] Through the dynamic multi-task scheduling engine, the scheduling algorithm is called from the scheduling algorithm library. Based on the scheduling algorithm, it is judged according to the status data whether there is a problem with the flexible production line. When it is determined that there is a problem with the flexible production line, a new flexible production line is constructed.

[0143] In the above embodiment, the scheduling algorithm library stores scheduling algorithms, and the dynamic multi-task scheduling engine can call corresponding scheduling algorithms to schedule resources according to different scenarios, thereby solving the problem of inaccurate collaborative scheduling control in the prior art.

[0144] Optionally, based on the above embodiment, the above-mentioned scheduling algorithm is based on the status data to determine whether there is a problem with the flexible production line, which can be achieved through the following steps.

[0145] Based on the scheduling algorithm, it is determined whether the flexible production line has abnormal working conditions and / or equipment failure according to the status data.

[0146] In the above embodiment, since working conditions and equipment are the two major factors affecting the production of the product to be processed, it is possible to determine whether the flexible production line has abnormal working conditions and / or equipment failures through status data, and then determine whether the flexible production line affects the production process of the product to be processed based on the judgment results, so that subsequent scheduling and processing can be carried out in a timely manner in the event of an impact.

[0147] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. An industrial Internet operating system, characterized in that: include: Heterogeneous data integration engine, digital twin model engine, and dynamic multi-task scheduling engine; The heterogeneous data integration engine processes the initial production data of the product to be processed according to a preset data format to generate target production data; The digital twin model engine obtains the production scheduling information of the product to be processed based on the flexible production line data used to produce the product to be processed and the historical production scheduling information of other products of the same category as the product to be processed. The digital twin model engine may also monitor the status of the flexible production line when the flexible production line produces the product to be processed based on the product information of the product to be processed, the production scheduling information, and the target production data, and obtain the status data of the flexible production line, wherein the product information is determined based on the user's functional requirements and / or appearance requirements for the product to be processed. The dynamic multi-task scheduling engine constructs a new flexible production line when the status data indicates that there is a problem with the flexible production line, so as to enable continued production of the product to be processed on the new flexible production line; The system further comprises: Industrial big data knowledge engine and industrial edge intelligent CPS management shell; The industrial big data knowledge engine obtains historical product information of the other products; The industrial edge intelligent CPS management shell determines the product information of the product to be processed based on the historical product information and the user's functional requirements and / or appearance requirements for the product to be processed.

2. The system according to claim 1, wherein: The system further comprises: The scheduling algorithm library stores scheduling algorithms, and the scheduling algorithms are used to implement the scheduling function of the dynamic multi-tasking scheduling engine.

3. A product processing method, applied to a server in the industrial Internet operating system according to claim 1 or 2, the method comprising: Determining product information of the product to be processed based on the user's functional requirements and / or appearance requirements of the product to be processed; Obtaining production schedule information of the product to be processed through a digital twin model engine based on flexible production line data used to produce the product to be processed and historical production schedule information of other products of the same category as the product to be processed; When the product to be processed is produced on the flexible production line according to the product information and the production scheduling information, the initial production data of the product to be processed is processed according to a preset data format by a heterogeneous data integration engine to generate target production data; According to the target production data, the flexible production line is monitored by the digital twin model engine to obtain status data of the flexible production line; When the status data indicates that there is a problem with the flexible production line, a new flexible production line is constructed through a dynamic multi-task scheduling engine to enable continued production of the product to be processed on the new flexible production line; The step of determining product information of the product to be processed includes: Obtain historical product information of the other products through an industrial big data knowledge engine; Based on the historical product information and the user's functional requirements and / or appearance requirements for the product to be processed, the product information of the product to be processed is determined through the industrial edge intelligent CPS management shell.

4. The method according to claim 3, characterized in that The step of monitoring the status of the flexible production line by using the digital twin model engine according to the target production data to obtain the status data of the flexible production line includes: According to the target production data, the flexible production line is monitored by the digital twin model engine to obtain status monitoring data of the flexible production line; According to the status monitoring data, the status of the flexible production line in the remaining production time is predicted by the industrial big data knowledge engine to obtain the status data.

5. The method according to claim 3 or 4, characterized in that Before obtaining the production scheduling information of the product to be processed through the digital twin model engine based on the flexible production line data for producing the product to be processed and the historical production scheduling information of other products of the same category as the product to be processed, the method further includes: According to the business logic in the product information and the functions of each physical discrete manufacturing equipment, the flexible production line is constructed through the heterogeneous data integration engine to connect the business flow and data flow throughout the entire life cycle of the product.

6. The method according to claim 5, characterized in that When the flexible production line produces the product to be processed according to the product information and the production scheduling information, the method further includes: The initial production data of the product to be processed is collected through the industrial edge intelligent CPS management shell.

7. The method according to claim 6, characterized in that When the status data indicates that there is a problem with the flexible production line, a new flexible production line is constructed by a dynamic multi-task scheduling engine, including: Calling a scheduling algorithm from a scheduling algorithm library through the dynamic multi-tasking scheduling engine; Based on the scheduling algorithm, judging whether there is a problem with the flexible production line according to the status data; When it is determined that there is a problem with the flexible production line, a new flexible production line is constructed.

8. The method according to claim 7, characterized in that The determining whether there is a problem with the flexible production line based on the status data based on the scheduling algorithm includes: Based on the scheduling algorithm, it is determined according to the status data whether the flexible production line has abnormal working conditions and / or equipment failure.

Images