Now showing 1 - 10 of 13
  • Publication
    Open Access
    Root cause analysis using anomaly detection and temporal informed causal graphs
    (2024-03)
    Rehak, Josephine
    ;
    Youssef, Shahenda
    ;
    Beyerer, Jürgen
    In industrial processes, anomalies in the production equipment may lead to expensive failures. To avoid and avert such failures, the identification of the right root cause is crucial. Ideally, the search for a root cause is backed by causal information such as causal graphs. We have extended a framework that fuses causal graphs with anomaly detection to infer likely root causes. In this work, we add the use of temporal information to draw temporal valid conclusions about the potential propagation of anomalous information in causal graphs. The use of the framework is demonstrated on a robotic gripping process.
  • Publication
    Open Access
    Metrics for the evaluation of learned causal graphs based on ground truth
    (2024-03)
    Rehak, Josephine
    ;
    Falkenstein, Alexander
    ;
    Doehner, Frank
    ;
    Beyerer, Jürgen
    The self-guided learning of causal relations may contribute to the general maturity of artificial intelligence in the future. To develop such learning algorithms, powerful metrics are required to track advances. In contrast to learning algorithms, little has been done in regards to developing suitable metrics. In this work, we evaluate current state of the art metrics by inspecting their discovery properties and their considered graphs. We also introduce a new combination of graph notation and metric, which allows for benchmarking given a variety of learned causal graphs. It also allows the use of maximal ancestral graphs as ground truth.
  • Publication
    Open Access
    End-to-end MLOps integration: a case study with ISS telemetry data
    (2024-03) ;
    Geier, Christian
    ;
    ;
    Creutzenberg, Martin
    ;
    Pfeifer, Jann
    ;
    Turk, Samo
    ;
    Kubeflow integrates a suite of powerful tools for Machine Learning (ML) software development and deployment, typically showcased independently. In this study, we integrate these tools within an end- to-end workflow, a perspective not extensively explored previously. Our case study on anomaly detection using telemetry data from the International Space Station (ISS) investigates the integration of various tools—Dask, Katib, PyTorch Operator, and KServe—into a single Kubeflow Pipelines (KFP) workflow. This investigation reveals both the strengths and limitations of such integration in a real-world context. The insights gained from our study provide a comprehensive blueprint for practitioners and contribute valuable feedback for the open source community developing Kubeflow.
  • Publication
    Open Access
    Retrofitting cyber-physical production systems with radio-based sensors and ML
    (2024-03)
    Kühnert, Christian
    ;
    Wallner, Stefan
    ;
    Wessels, Lars
    ;
    Wunsch, Andreas
    ;
    Ziebarth, Mathias
    Manufacturing companies usually isolate their production networks from other networks to ensure security against external attacks and to guarantee a fail-safe 24/7 operational service. However, these measures make it technically and organizationally complex to install new sensors or deploy new software in the production process. As a result, machine learning is only used to a limited extent in manufacturing, as these models require regular adaptations. To tackle this challenge, one possible solution is to install an additional network that is not connected to the production network. This network can be utilized for rapid prototyping of new sensors, advanced data analysis, or the deployment of machine learning models. One possible solution is to install a radio-based low-power, long-range network, having the property to capture data over large distances with only little power consumption. This paper examines the potential of retrofitting cyberphysical systems with such a network in combination with machine learning methods. The results are evaluated through three practical use cases: monitoring a workspace with a molding machine, monitoring the cycles of a washing machine, and predicting the daily consumption profile of a main water pipeline.
  • Publication
    Open Access
    Hybrid online timed automaton learning algorithm for discrete manufacturing systems
    (2024-03)
    Martens, Simon
    ;
    Maier, Alexander
    ;
    Wunn, Alexander
    This paper presents the Hybrid Online Timed Automaton Learning Algorithm (HyOTALA), a novel approach for anomaly detection in cyber-physical production systems (CPPS). It addresses the challenge of using hybrid data, combining sparse discrete and continuous signals, by learning a hybrid timed automaton model in an online setting. This model captures the dynamics of discrete manufacturing processes and provides significant advances in model identification for CPPS. The effectiveness of HyOTALA is demonstrated through a practical application, highlighting its potential to improve anomaly detection capabilities in industrial settings.
  • Publication
    Open Access
    XAI for anomaly analysis by power plant operators - a case and user study
    (2024-03)
    Dix, Marcel
    ;
    Koltermann, Jan
    ;
    Mieck, Sebastian
    ;
    Pastler, Boris
    ;
    Kloepper, Benjamin
  • Publication
    Open Access
    A Hybrid model for hot rolling pass schedule design using reinforcement learning
    (2024-03)
    Kemmerling, Marco
    ;
    Frahm, Nils
    ;
    Idzik, Christian
    ;
    Abdelrazeq, Anas
    ;
    Bailly, David
    ;
    Schmitt, Robert H.
    Hot rolling of metals requires the design of pass schedules to accommodate different products and arrive at the desired target properties. Pass schedule design is often still performed manually, although automated approaches are emerging in the literature. One such approach is the use of reinforcement learning, which requires extensive training on simulations. To allow for efficient development of suitable reinforcement learning agents, it is important to reduce the computational cost of each simulation step as much as possible. While some simulation models for hot rolling exist, they have not been developed explicitly for reinforcement learning purposes. Today, there is a lack of fast approximate models to allow for efficient (pre-)training of reinforcement learning agents. To address this gap, we propose a hybrid model consisting of a data-driven model augmented by physical approximations. Our hybrid model produces better predictions than either of its components and is efficient enough to allow for quick training of reinforcement learning agents. We demonstrate that reinforcement learning agents pre-trained on our hybrid model can be transferred to a more accurate simulation and achieve good results after a short additional training phase. The two-stage training approach significantly reduces the overall training while achieving similar agent performances. The reduction in training time enables faster iteration cycles in agent and environment design, thereby creating a foundation for the development of reinforcement learning approaches for a wider range of hot rolling scenarios in the future.
  • Publication
    Open Access
    Regression via causally informed Neural Networks
    (2024-03)
    Youssef, Shahenda
    ;
    Doehner, Frank
    ;
    Beyerer, Jürgen
    Neural Networks have been successful in solving complex problems across various fields. However, they require significant data to learn effectively, and their decision-making process is often not transparent. To overcome these limitations, causal prior knowledge can be incorporated into neural network models. This knowledge improves the learning process and enhances the robustness and generalizability of the models. We propose a novel framework RCINN that involves calculating the inverse probability of treatment weights given a causal graph model alongside the training dataset. These weights are then concatenated as additional features in the neural network model. Then incorporating the estimated conditional average treatment effect as a regularization term to the model loss function, the potential influence of confounding variables can be mitigated, leading to bias minimization and improving the neural network model. Experiments conducted on synthetic and benchmark datasets using the framework show promising results.
  • Publication
    Open Access
    Leveraging self-supervised learning for vibration data in industrial separators
    (2024-03)
    Heuwinkel, Tim
    ;
    Merkelbach, Silke
    ;
    Janssen, Nils
    ;
    Enzberg, Sebastian von
    ;
    Dumitrescu, Roman
    Industrial separators play a pivotal role in production processes of various sectors such as chemical, pharmaceutical, biotechnology, oil extraction and food industries, with over 3000 distinct applications. Operating these separators involves managing several process parameters as well as discharge and cleaning cycles, which are hard to control mainly due to deficiencies of current physical sensor technology. Recent studies have shown that machine learning can be utilized to detect faults and particle presence in separators via vibration data. However, traditional machine learning methods require domain expertise or vast amounts of labeled data. We propose the use of self-supervised learning to resolve this issue by learning useful representations from unlabeled data, which is significantly easier and cheaper to obtain. An empirical validation on data from a disk stack separator shows that self-supervised learning can improve upon manual feature engineering and supervised approaches in terms of cost, accuracy and data efficiency.
  • Publication
    Open Access
    Integrating continuous-time neural networks in engineering: bridging machine learning and dynamical system modeling
    This paper examines the integration of Continuous-Time Neural Networks (CTNNs), including Neural ODEs, CDEs, Neural Laplace, and Neural Flows, into engineering practices, particularly in dynamical system modeling. We provide a detailed introduction to CTNNs, highlighting their underutilization in engineering despite similarities with traditional Ordinary Differential Equation (ODE) models. Through a comparative analysis with conventional engineering approaches, using a spring-mass-damper system as an example, we demonstrate both theoretical and practical aspects of CTNNs in engineering contexts. Our work underscores the potential of CTNNs to harmonize with traditional engineering methods, exploring their applications in Cyber- Physical Systems (CPS). Additionally, we review key open-source software tools for implementing CTNNs, aiming to facilitate their broader integration into engineering practices.