Link to Publications
Link to Activities
Link to Projects
Machine learning for the discretization of hybrid, time-dependent state spaces:
Signals of complex plants today contain both discrete and continuous information. By converting them to value-discrete and discrete-time states, problems such as predictive maintenance, optimization and diagnostics can be addressed.
Niggemann, Oliver; Stein, Benno; Vodenčarević, Asmir; Maier, Alexander; Kleine Büning, Hans: Learning Behavior Models for Hybrid Timed Systems. In: Twenty-Sixth Conference on Artificial Intelligence (AAAI-12) Jul 2012.
von Birgelen, Alexander; Niggemann, Oliver: Anomaly Detection and Localization for Cyber-Physical Production Systems with Self-Organizing Maps. S.: 55-71, Springer Vieweg, Aug 2018.
Anomaly Detection for High-Dimensional, Time-Prone Signals of Technical Systems:
By learning models of normal behavior, non normal operating conditions can be detected early (condition monitoring) and downtimes predicted (predictive maintenance).
Niggemann, Oliver; Frey, Christian: Data-driven anomaly detection in cyber-physical production systems. In: at – Automatisierungstechnik(63) S.: 821–832, Oct 2015.
Niggemann, Oliver; Lohweg, Volker: On the Diagnosis of Cyber-Physical Production Systems – State-of-the-Art and Research Agenda. In: Twenty-Ninth Conference on Artificial Intelligence (AAAI-15) Austin, Texas, USA, Jan 2015.
Machine-learning-based diagnosis algorithms for complex production facilities:
Unlike the detection of anomalies, the use of machine learning methods in diagnosis, i.e. the detection of the causes of the errors, is more difficult. Enhancements to the established model-based diagnostics will enable future adaptive, complex systems to be diagnosed based on data.
Diedrich, Alexander; Niggemann, Oliver: Model-based Diagnosis of Hybrid Systems using Satisfiability Modulo Theory. Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19), Hawaii, USA, Jan 2019.
Bunte, Andreas; Stein, Benno; Niggemann, Oliver: Model-Based Diagnosis for Cyber-Physical Production Systems Based on Machine Learning and Residual-Based Diagnosis Models. Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19), Hawaii, USA, Jan 2019.
Use of A-Priori Knowledge to Improve Machine Learning and Artificial Intelligence:
Technical systems are characterized by the abundance of a-priori knowledge such as system structures and physical laws. As a result, methods of machine learning require less data and become extrapolatable. Furthermore, methods of artificial intelligence such as diagnosis can use causal relationships and calculate optimizations.
Otto, Jens; Vogel-Heuser, Birgit; Niggemann, Oliver: Automatic Parameter Estimation for Reusable Software Components of Modular and Reconfigurable Cyber Physical Production Systems in the Domain of Discrete Manufacturing. In: IEEE Transactions on Industrial Informatics IEEE, Jan 2018.
von Birgelen, Alexander; Buratti, Davide; Mager, Jens; Niggemann, Oliver: Self-Organizing Maps for Anomaly Localization and Predictive Maintenance in Cyber-Physical Production Systems. In: 51st CIRP Conference on Manufacturing Systems (CIRP CMS 2018) CIRP-CMS, May 2018.
Letzte Änderung: 28. August 2020