Research project EN-AI-BLER
The goal of the EN-AI-BLER research project is to automatically connect non-standardized individual machines and production equipment to platforms. In this way, a consistent data pipeline is to be achieved, which represents a major challenge for manufacturing companies, especially in heterogeneous production landscapes consisting of diverse machines and equipment. The provision of production data makes the project an enabler of tomorrow's production, which is essentially based on Artificial Intelligence (AI) methods.
Artificial intelligence increases flexibility
As part of the collaborative project, a team from the wbk Institute of Production Engineering at the Karlsruhe Institute of Technology (KIT), together with iT Engineering Software Innovations GmbH and Braun Sondermaschinenbau GmbH, aims to develop an AI-based identification algorithm. This can be used to search through a wide variety of variables from different control architectures and consistently define the type and naming or assign them to a defined model. This allows existing machines and systems to be connected to the growing number of platforms in a simple and scalable manner.
For industrial applications such as condition monitoring systems, process monitoring or intelligent planning algorithms, the next step is to provide the data consistently defined by the identification algorithm in a uniform manner. Thus, signals can be assigned to the motor current or the axis speed, for example. These signals must then be subdivided into individual, granular segments using an AI-based segmentation function in order to later make the different raw data usable on many heterogeneous production machines. Trained models can thus be profitably applied across different plants.
The EN-AI-BLER project is funded by the Baden-Württemberg Ministry of Economics, Labor and Housing as part of the Baden-Württemberg AI Innovation Competition and is scheduled to be completed by Dec. 31, 2021. To enable AI solutions "made in Baden-Württemberg" to be developed more quickly, the state is funding a total of 44 selected projects as part of the "AI Innovation Competition". One of these is our "EN-AI-BLER" project.
Opportunities for industry
The wbk team brings to the project, above all, extensive expertise in the fields of Industry 4.0 and artificial intelligence. "We could observe that many companies implement AI pilot projects and also see the success of this technology. However, the hurdle to broad application in productions is the standardization and generalization of so-called data pipelines and data structures. With EN-AI-BLER, we are trying to create these structures consistently in order to transfer AI algorithms to new plants and products very quickly and easily," says Markus Netzer, academic staff member and project manager for "EnAIbler" at the wbk Institute for Production Technology at KIT.
In the project, the wbk team is responsible for the intelligent functions, which are integrated into solutions of the software provider iT Engineering and applied to Braun special machines in customer plants as well as the company's own production. "Particularly for the classic mechanical and plant engineering sector in Baden-Württemberg, digitization - exacerbated by the pandemic - represents a major technical and financial hurdle. With EN-AI-BLER, we are creating a cost-optimal basis for broad scaling of industrial AI systems, without individual programming effort," explains Philipp Gönnheimer, group leader and project manager for "EnAIbler" at the wbk Institute for Production Technology at KIT.
Further information:
https://www.wbk.kit.edu/wbkintern/Forschung/Projekte/EN-AI-BLER/index.php
Source: wbk
At the wbk, data structures are created for broad application in productions. Source: wbk