Gary Hildebrandt

Zusammenfassung

Ongoing digitalization, accelerated by concepts such as Industry 4.0 and the Internet of Things, is leading to increasingly interconnected production systems. A key enabler of this development is the Digital Twin, serving as a digital representation of an asset, using real-time data to support forecasting, monitoring, and optimization tasks. As connectivity increases, so does the diversity of available data sources for Digital Twins. Integrating this data requires transformation pipelines that convert raw data into formats suitable for Digital Twin models. While various tools exist for manually constructing such pipe- lines, limited research has been conducted on their automated discovery and generation. As a result, indus- trial connectivity is often hindered by the high engineering effort required for manual integration, making such efforts economically impractical. In this paper, we propose an approach that encapsulates data trans- formations into modular, reusable transformation units, which can be flexibly combined to form integration pipelines. This concept is further enhanced by leveraging Large Language Models (LLMs) to automatically generate appropriate transformation routes based on available data sources, algorithms, and model require- ments. A prototypical implementation within a robot-based gluing scenario demonstrates the feasibility. The presented system enables Digital Twins to evaluate and integrate external data sources dynamically, supporting a more scalable and efficient realization of interconnected production systems.

Zusammenfassung

Offshore Power-to-X platforms enable flexible conversion of renewable energy, but place high demands on adaptive process control due to volatile operating conditions. To face this challenge, using Digital Twins in Power-to-X platforms is a promising approach. Comprehensive knowledge integration in Digital Twins requires the combination of heterogeneous models and a structured representation of model information. The proposed approach uses a standardized description of behavior models, semantic technologies and a graph-based model understanding to enable automatic adaption and selection of suitable models. It is implemented using a graph-based knowledge representation with Neo4j, automatic data extraction from Asset Administration Shells and port matching to ensure compatible model configurations.

Zusammenfassung

In the automation industry, concepts such as Digital Twins, representing virtual counterparts of physical assets, enhance efficiency, sustainability, and economic viability by enabling e.g., optimizations, advanced monitoring, and predictive maintenance. To achieve this, Digital Twins rely on models (including simulation models), which in turn depend on data describing the current state of the physical asset. As connectivity increases, the available data extends beyond the data of the sensors of the asset itself. It extends to data from the surrounding environment and even beyond. This introduces new challenges for Digital Twins and their models in terms of data integration. In previous research it was shown that current implementations are not adequately prepared for this scenario. Currently, no analysis in the literature addresses how interaction within a connected environment alters the requirements for Digital Twins. Therefore, this paper explicitly identifies the emerging challenges associated with extending the information acquisition scope of Digital Twins beyond the immediate physical asset. This is accomplished through a systematic analysis of potential scenarios that consider the inclusion of external data. Our analysis reveals several new requirements, such as the need for an extended description of the model’s purpose and the specific requirements for the data input. Digital Twins must possess enhanced capabilities to identify appropriate data sources within a dynamic production environment, as well as methods to manage multiple, potentially redundant data streams effectively. Through this analysis, we aim to highlight potential research directions for Digital Twins in the automation industry and advocate for the integration of interconnectivity into Digital Twin systems.