Manufacturing control system and method for manufacturing control

Abstract

The invention relates to a manufacturing control system (1) with an arrangement of manufacturer-specific modules, which arrangement comprises modules for controlling manufacturing units, modules for controlling logistics units and / or modules for providing and / or processing product data (6). The proprietary data models of the manufacturer-specific modules are linked by means of at least one ontology unit. The invention also relates to a method for manufacturing control by means of an arrangement of manufacturer-specific modules.

Description

[0001] This case is a divisional application of the patent application filed on October 16, 2020, with application number 202011108214.2 and invention title "Manufacturing Control System and Method for Manufacturing Control". Technical Field

[0002] This invention relates to a manufacturing control system. It also relates to a method for manufacturing control.

[0003] These types of manufacturing control systems are typically used to control manufacturing facilities that can be used to manufacture a variety of products. Here, the term "manufacturing facility" also includes the entire manufacturing plant (i.e., the factory). Background Technology

[0004] In modern manufacturing plants, especially factories, manufacturing processes and logistics are largely automated. Modern factories (so-called smart factories) are subdivided into flexible manufacturing units, where logistics processes (especially those for transporting materials between manufacturing units) are handled by autonomous vehicles, particularly FTS (driverless transport vehicles).

[0005] To achieve an optimized manufacturing process, it is known to control these distributed units through a higher-level control device. This allows for optimizations such as avoiding FTS downtime or improving the utilization rate of individual manufacturing units.

[0006] However, a major problem is that each unit is constructed with manufacturer-specific modules that have proprietary data models.

[0007] For example, in product development, product data for parts to be manufactured is generated in a data model that differs from the data model of manufacturing units that require these parts, such as those used to perform assembly steps. The data model used to operate logistics units (e.g., FTS) is also constructed in a different manner.

[0008] This makes it difficult to coordinate the various units within a manufacturing facility or plant. This, in particular, makes the automation and optimization of processes within the manufacturing facility challenging.

[0009] In principle, these problems can be mitigated by setting appropriate interfaces between the various units to perform data conversions, etc. However, this implies a considerable amount of additional structural overhead—especially since manufacturing facilities have a large number of manufacturer-specific modules with individual, different proprietary data models, which necessitates different interfaces at the system level. Summary of the Invention

[0010] The present invention is based on the following objective: to provide a manufacturing control system of the type mentioned at the beginning, which has a simple and flexible architecture and at the same time has high functionality.

[0011] To solve this task, features according to the present invention are provided. Advantageous embodiments and desired extensions of the invention are described in the extended technical solutions of the invention.

[0012] This invention relates to a manufacturing control system having a manufacturer-specific arrangement of modules, including modules for controlling manufacturing units, modules for controlling logistics units, and / or modules for providing and / or processing product data. The proprietary data models of the manufacturer-specific modules are linked through at least one ontology unit. The ontology unit establishes a class structure that forms the ontology. This class structure has hierarchically or grouped classes that enable semantic networking of the proprietary data model. The ontology unit has a pre-defined number of inference units that link the ontology's classes. These inference units constitute interface units for different proprietary data models, enabling continuous information exchange between all proprietary data models without requiring separate interfaces at the system level. The ontology unit is also structured as a resource unit, configured to generate manufacturing processes. Data and programs generated in the resource unit, process unit, and product unit are input to a verification layer and a programming layer. In the verification layer, the created program code is checked and verified, including debugging: checking whether the programmed process is feasible, and especially whether it is conflict-free.

[0013] The present invention also relates to a method for manufacturing control.

[0014] Using the ontological unit according to the invention, semantic networking is performed on all the different proprietary data models of the various units of a manufacturing apparatus (e.g., a complete factory).

[0015] This enables continuous information exchange between all proprietary data models without the need for separate interfaces at the system level. Instead, semantic networking, achieved through ontology units, allows for the mapping of data models into a comprehensive digital model—the so-called digital twin (Zwilling)—enabling bidirectional, coordinated data flow between all units of a manufacturing facility without additional hardware overhead at the system level. This is particularly relevant for networking proprietary data models from entirely different domains such as product development, manufacturing, and logistics.

[0016] According to an advantageous configuration of the invention, a class structure for forming the ontology is predefined in the ontology unit. This class structure has hierarchically or grouped classes that enable semantic networking of proprietary data models.

[0017] Advantageously, classification can be performed using structured meta-models within the ontology library.

[0018] Advantageously, semantic networking of proprietary data models can be configured with the help of these ontologies.

[0019] Therefore, data mapping can be flexibly performed on all proprietary data models.

[0020] One possible implementation of semantic networking for different proprietary data models is that the ontology unit has a pre-given number of reasoning units, by means of which the classes of the ontology can be linked.

[0021] Here, the inference unit constitutes an interface unit for different proprietary data models.

[0022] This system operates as follows: data from a proprietary data model is structured within multiple object models. These object models forming the ontology have a class structure with hierarchically or grouped classes, wherein each class is preferably assigned multiple attributes, which are passed down through inheritance within the class structure. To access the stored data, the ontology has a pre-defined number of rules, which are assigned to the inference unit. These rules link the attributes of at least one class structure and / or the classes of at least one class structure, along with the stored data, to each other. Here, the rules are logical linking rules that correlate the aforementioned elements in a pre-defined manner. Analysis is performed in the inference unit by assigning specific values ​​about attributes, classes, and / or data to these rules, thereby generating definite output parameters.

[0023] The inference unit constitutes the interface unit between different proprietary data models.

[0024] The rules assigned to the inference unit compile the structure of the object model of the first data model into the object model of the second data model. Thus, when data is transferred from one data model to another, the defined allocation of the transferred data is always guaranteed.

[0025] According to a favorable configuration, the ontology is described using standard protocols.

[0026] In particular, RDF (Resource Description Framework) is used as a standard protocol.

[0027] By using these industry standards, ontology units can be extended to open libraries.

[0028] An advantageous ontology structure is characterized by setting ontology units in the form of resource units, which are configured to generate data that defines the manufacturing process.

[0029] In addition, a body unit in the form of a process unit is provided, which is configured for programming manufacturing units and / or logistics units.

[0030] In addition, a body unit in the form of a product unit is provided, which is configured to process product data.

[0031] These individual body units are assigned a higher-level body unit, which forms the control logic for the individual body units.

[0032] By using resource units, process units, and product units, all relevant processes of a manufacturing unit can be mapped, and all relevant processes are coordinated through a higher-level entity unit.

[0033] The entire physical and digital components of the manufacturing apparatus are fully virtualized using ontology units. These ontology units are advantageously equipped with normalization layers to generate executable instructions for these components.

[0034] According to an advantageous embodiment of the invention, a set of rules (Regelwerk) for self-learning is set.

[0035] Therefore, known machine learning methods can be used, particularly to draw conclusions about changes in the entire manufacturing facility, and to optimize existing processes, especially manufacturing and logistics processes. Thus, the entire manufacturing control system forms a closed, adjustable system, achieving continuous matching of data flows throughout the entire product lifecycle of the product to be manufactured.

[0036] According to another advantageous configuration, the manufacturing control system according to the invention is integrated into a cloud computer network.

[0037] Therefore, the optimization of manufacturing control systems is not limited to local machine learning methods, but can be extended to networked systems worldwide when necessary. Attached Figure Description

[0038] The invention is described below with reference to the accompanying drawings. The drawings show: Figure 1 A block diagram illustrating an embodiment of a manufacturing control system according to the present invention is shown. Detailed Implementation

[0039] Figure 1An embodiment of a manufacturing control system 1 according to the present invention is shown. This manufacturing control system 1 is used to control the manufacturing process and logistics process in a manufacturing facility. In the present case, the manufacturing facility is constructed as a so-called smart factory, in which manufacturing cells are arranged and multiple autonomous vehicles, particularly FTS (Free-Traffic System). Different manufacturing or assembly processes are performed in each manufacturing cell. Logistics processes are performed using autonomous vehicles. In particular, FTS is used to transport materials, parts to be manufactured, or assembled units.

[0040] Manufacturing units can be used, in particular, to manufacture motor vehicles.

[0041] According to the present invention, the manufacturing control system 1 has an arrangement of body units by means of which proprietary data models of manufacturer-specific manufacturing units, logistics units, and product development-specific data and processes are semantically networked.

[0042] Here, the class structure that forms the ontology is typically predefined in the ontology unit. The class structure has hierarchically or grouped classes that enable semantic networking of proprietary data models.

[0043] Semantic networking with proprietary data models is configurable and tailored to the destination.

[0044] Advantageously, the ontology is described using standard protocols.

[0045] In particular, RDF (Resource Description Framework) is used as a standard protocol.

[0046] like Figure 1 As shown, resource unit 2, process unit 3, and product unit 4 are set as body units, which are controlled by control logic 11, which is a higher-level body unit. These body units form a virtual model of the process running in the manufacturing apparatus.

[0047] Each ontology unit contains an object model that forms an ontology. These object models are linked to each other through appropriate ontology links 5 (e.g., reasoning units), thereby enabling semantic networking of proprietary data models existing in resource unit 2, process unit 3, and product unit 4.

[0048] In resource unit 2, ontology makes data describing and defining the manufacturing process available in a structured and digital manner. This data constitutes a manufacturing-specific proprietary data model, which is incorporated into a coordinated data flow that can flow through all units of the manufacturing facility via ontology.

[0049] The manufacturing process typically includes not only the manufacturing process but also the logistics process.

[0050] In process unit 3, a defined manufacturing process for a manufacturing device or working device (e.g., a robot) is programmed using executable program code (e.g., SPS code) based on data from the proprietary data models of resource unit 2 and product unit 4. Coordinated data transfer between process unit 3 and resource unit 2 and product unit 4 can be achieved through the ontology without the need to establish corresponding interfaces at the system level.

[0051] Provide and make product data 6 available in product unit 4. For example... Figure 1 As shown, based on external, customer-specific pre-given product specifications 7, product data 6 from a proprietary data model is provided to product unit 4, for example as CAD data.

[0052] Product data 6 is processed and provided in product unit 4. In particular, product data 6 is permanently and preferably non-volatilely stored as persistent data 8 in working memory as an immutable storage medium.

[0053] Furthermore, a digital model is generated from product data 6 using the entity of product unit 4, and the digital model is stored as a so-called digital twin 9.

[0054] Persistent data 8 and digital twin 9 can be analyzed through analysis unit 15.

[0055] Resource unit 2, process unit 3, and product unit 4, which have their own entities, are controlled and coordinated by the superior entity unit, which is the control logic 11.

[0056] Overall, resource unit 2, process unit 3, product unit 4 and the superior ontology unit form an ontology model. With the help of this ontology model, all proprietary data models of the manufacturing unit are semantically networked, thereby obtaining a standardized and coordinated data flow of all data in the entire system among all units of the manufacturing unit, without having to match the data to be transmitted at the system level through physical interfaces.

[0057] Here, resource unit 2 is used to prepare proprietary data for machine manufacturers who deliver and provide working or manufacturing devices (e.g., processing machines) and logistics units (e.g., FTS) with defined functionalities.

[0058] Product Unit 4 is used to prepare and provide customer-specific proprietary product data 6.

[0059] Finally, in process unit 3, the process designer also generates proprietary data by generating executable program code for manufacturing or operating devices.

[0060] The data and programs generated in resource unit 2, process unit 3 and product unit 4 will be input to the verification layer 12 and programming layer 13.

[0061] In verification layer 12, the created program code is inspected and verified. Debugging is particularly important: checking whether the programmed process is feasible, and especially whether it is conflict-free.

[0062] After successful verification, machine code for manufacturing and operating devices is generated using programming layer 13.

[0063] Normalization layer 14 is set as another component of the ontology model according to the present invention. In the normalization layer, program code written in a high-level language is compiled into user software (e.g., SPS software).

[0064] Finally, an adapter layer 15 is configured to establish a connection to the communication unit 16 (e.g., email, internet, etc.). Furthermore, the adapter layer 15 is also used to establish connections with external units 17 from suppliers, partner companies, etc.

[0065] List of reference numerals 1 Manufacturing Control System 2 Resource Units 3 Process Units 4 Product Units 5 links 6 Product Data 7 Product Specifications 8. Persistent Data 9. Digital Twins 10 Analysis Units 11 Control Logic 12 Verification Layer 13 Programming Layer 14 Normalization layer 15 Adapter Layer 16 Communication Units 17. External Units.

Claims

1. A manufacturing control system (1) having a manufacturer-specific arrangement of modules, the arrangement of modules including modules for controlling manufacturing units, modules for controlling logistics units, and / or modules for providing and / or processing product data (6), wherein, The manufacturer-specific modules are each constructed with proprietary data models, wherein the proprietary data models of the manufacturer-specific modules are linked through at least one ontology unit, wherein a class structure forming the ontology is set in the ontology unit, the class structure having hierarchically or grouped structured classes, by means of which semantic networking of the proprietary data model can be performed, wherein the ontology unit has a pre-given number of reasoning units, by means of which the classes of the ontology can be linked, wherein the reasoning units constitute interface units for different proprietary data models, enabling continuous information exchange between all proprietary data models without having to set up separate interfaces at the system level, wherein ontology units in the form of resource units (2) are set, the resource units being constructed for generating manufacturing processes, wherein the data and programs generated in the resource units (2), process units and product units are input to the verification layer and the programming layer, wherein in the verification layer, the created program code is checked and verified, wherein debugging is performed: checking whether the programmed process is feasible, especially whether it is conflict-free.

2. The manufacturing control system (1) according to claim 1, characterized in that, The semantic networking of the proprietary data model is configurable.

3. The manufacturing control system (1) according to claim 1, characterized in that, The ontology is described using a standard protocol.

4. The manufacturing control system (1) according to claim 3, characterized in that, RDF is used as the standard protocol.

5. The manufacturing control system (1) according to claim 1, characterized in that, The body unit is configured as a process unit (3) for programming manufacturing and / or logistics units.

6. The manufacturing control system (1) according to claim 1, characterized in that, The body unit is configured in the form of a product unit (4), which is configured to process the product data (6).

7. The device control system (1) according to claim 1, characterized in that, The upper-level body unit is set as the body unit, and the upper-level body unit constitutes the control logic (11).

8. The manufacturing control system (1) according to claim 1, characterized in that, Set up a set of rules for self-learning.

9. The manufacturing control system (1) according to claim 1, characterized in that, The manufacturing control system is integrated into a cloud computer network.

Images