Publications Search Results
1 - 10 of 149
-
PublicationR&D&I and Industry Examples: The CCU Project Carbon2Chem( 2024)In the reduction of CO2 emissions from industrial plants, an economic or technical limit has been reached in many places. A further noticeable reduction of emissions can only be achieved by cross-industrial cooperation between different industrial sectors. The coupling of processes allows the material use of process gases and residual gases containing CO and CO2. In addition to the reduction of CO2 emissions, a reduction in the use of fossil raw materials can thus also be achieved. In the Carbon2Chem® joint project, companies from various industries as well as scientists in the field of basic and applied research are working to realize this coupling. The central aim is for the targeted integrated system comprising chemicals, steel production and energy generation to ultimately be more economical and sustainable than the sum of the individual systems involved. In addition to the preparation of a synthesis gas suitable for material use from metallurgical gases, the integration of renewable energies and the dynamic accumulation of metallurgical gases represent a major challenge. The flexibilization of the overall system required for this purpose calls for new technical and organizational solutions, which are to be found on an industrial scale in the Carbon2Chem® project funded by the BMBF.
-
PublicationFault detection in automated production systems based on a long short-term memory autoencoder( 2024)In this paper, a hybrid model of regularized Long Short-Term Memory (LSTM) and autoencoder for fault detection in automated production systems is proposed. The presented LSTM autoencoder is used as a stochastic process model, which captures the normal behavior of a production system and allows to predict the probability distribution of sensor data. Discrepancies between the observed sensor data and the predicted probability density distribution are detected as potential faults. The approach combines the advantages of LSTMs and autoencoders: The correlations between individual sensor signals are exploited by an autoencoder, while the temporal dependencies are captured by LSTM neurons. A key challenge in training such a process model from historical data is to control the information passed through the latent space of the autoencoder. Different regularization methods are investigated for this purpose. Fault detection with the proposed LSTM autoencoder has been evaluated on the use case of an industrial penicillin production, achieving significantly improved results in comparison to the baseline LSTM.
-
PublicationICNAP Study Report 2023( 2023-11-27)
-
-
PublicationA data-driven approach for motion planning of industrial robots controlled by high-level motion commands( 2023)Most motion planners generate trajectories as low-level control inputs, such as joint torque or interpolation of joint angles, which cannot be deployed directly in most industrial robot control systems. Some industrial robot systems provide interfaces to execute planned trajectories by an additional control loop with low-level control inputs. However, there is a geometric and temporal deviation between the executed and the planned motions due to the inaccurate estimation of the inaccessible robot dynamic behavior and controller parameters in the planning phase. This deviation can lead to collisions or dangerous situations, especially in heavy-duty industrial robot applications where high-speed and long-distance motions are widely used. When deploying the planned robot motion, the actual robot motion needs to be iteratively checked and adjusted to avoid collisions caused by the deviation between the planned and the executed motions. This process takes a lot of time and engineering effort. Therefore, the state-of-the-art methods no longer meet the needs of today’s agile manufacturing for robotic systems that should rapidly plan and deploy new robot motions for different tasks. We present a data-driven motion planning approach using a neural network structure to simultaneously learn high-level motion commands and robot dynamics from acquired realistic collision-free trajectories. The trained neural network can generate trajectory in the form of high-level commands, such as Point-to-Point and Linear motion commands, which can be executed directly by the robot control system. The result carried out in various experimental scenarios has shown that the geometric and temporal deviation between the executed and the planned motions by the proposed approach has been significantly reduced, even if without access to the “black box” parameters of the robot. Furthermore, the proposed approach can generate new collision-free trajectories up to 10 times faster than benchmark motion planners.
-
PublicationF5G OpenLab: Enabling Twin Transition through Ubiquitous Fiber Connectivity( 2023)The paper introduces a new open laboratory, the F5G OpenLab, which aims at fostering the advancement of fiber-based solutions for everything. F5G OpenLab intends to aid in creating a sustainable and eco-friendly ICT industry and accelerate the digital transformation through autonomous networking solutions that are secure and trustworthy. In particular, the purpose of the F5G OpenLab is to establish an ecosystem for the validation of optical networking solutions that facilitate twin transition, provide a vendor-neutral platform for the assessment of vertical use cases, and enable the development of fiber-based solutions. It provides a platform to verify and unify next-gen networking solutions, access to early hardware and software releases, and unique testing and measurement facilities. Finally, the F5G OpenLab supports the development of blueprints for a green and digital transformation, capitalizing on the benefits of fiber technology for all industry sectors. We present its architecture as well as key features and capabilities. Moreover, we report several proof-of-concept demonstrations focused on industry 4.0 vertical use-cases.
-
PublicationTowards Self-learning Industrial Process Behaviour from Payload Bytes for Anomaly Detection( 2023)Network Intrusion Detection System (NIDS) for process-based anomaly detection have been developed as one of the cybersecurity solutions against industrial process targeted attacks such as Stuxnet. In practice, the real-world industrial plants could not complement the advancements in the industrial cybersecurity research as upgrading the infrastructure is an expensive and deterrent process for plant owners. In addition, the infrastructure information might be lost over the intended longer lifetime, hence, configuring a NIDS in the absence of such information is a challenge. Moreover, the existing NIDS solutions analyze the industrial process values/parameters with the knowledge of their semantics, and would fail when the semantics is not known or lost. As a solution to aforementioned problem, we propose an industrial communication paradigm aware Process Payload Profiling Framework (P3F), capable of self-learning process behavior from network traffic without the knowledge of underlying process parameters being exchanged. We also report P3F’s successful detection of an anomaly in the process of a miniaturized PROFINET-based industrial system, caused by a simulated process-targeted cyberattack
-
PublicationInvestigation and Improvements of Neural Field Efficiency and Quality( 2023)
-
PublicationThe Carbon2Chem® Laboratory in Oberhausen - A Workplace for Lab-Scale Setups within the Cross-Industrial Project( 2022-06-16)Within the Carbon2Chem® network, basic research is mandatory for a successful implementation and realization of sustainable technologies for CO2 emission reduction. For this purpose, the exchange of knowledge between the project partners in the individual subareas is as essential as obtaining precise data on the fundamental parameters on a laboratory scale in order to transfer them later to large-scale plants. Therefore, the Carbon2Chem® laboratory offers a platform to gain detailed insights into the individual sub-processes and to then apply these findings at the technical center in Duisburg.
-
Publication5G-Industry Campus EuropeThe 5G-Industry Campus Europe offers a unique ecosystem for research, development and testing of 5G technology for industrial applications. The Fraunhofer IPT and its research partners in Aachen are testing the first industrial 5G applications at the 5G-Industry Campus Europe. In seven sub-projects, various application scenarios are being researched, from 5G sensor technology for monitoring and controlling complex manufacturing processes, mobile robots cooperating on an assembly task, or AGVs (automated guided vehicles) enabling flexible and logical supply chains - for all these applications, 5G offers the possibility of reliable real-time communication and thus the possibility of networked, adaptive production. Furthermore, the research partners are also testing the use of modern edge cloud systems for fast data processing in order to exploit the potential of 5G in networked, adaptive production.
