SciTePress - Affiliation Profile

OSTWESTFALEN LIPPE, Germany

Context (please use complete words only)

Field

Sort by

6 Papers

Mapping Data Sets to Concepts using Machine Learning and a Knowledge based Approach

Topics: Knowledge Representation and Reasoning; Ontologies; Cognitive Systems; Industrial Applications of AI; Knowledge-Based Systems

Authors: Andreas Bunte ¹ , Peng Li ¹ and Oliver Niggemann ²

Affiliations: ¹ Institute for Industrial IT, Germany , ² Institute for Industrial IT and Fraunhofer IOSB-INA, Germany

Digital Object Identifier: 10.5220/0006590204300437

Proceedings of ICAART 2018, ISBN: 978-989-758-275-2

Quick Abstract

Machine learning techniques have a huge potential to take some tasks of humans, e.g. anomaly detection or predictive maintenance, and thus support operators of cyber physical systems (CPSs). One challenge is to communicate algorithms results to machines or humans, because they are on a sub-symbolical level and thus hard to interpret. To simplify the communication and thereby the usage of the results, they have to be transferred to a symbolic representation. Today, the transformation is typically static which does not satisfy the needs for fast changing CPSs and prohibit the usage of the full machine learning potential. This work introduces a knowledge based approach of an automatic mapping between the sub-symbolic results of algorithms and their symbolic representation. Clustering is used to detect groups of similar data points which are interpreted as concepts. The information of clusters are extracted and further classified with the help of an ontology which infers the current operational state. Data from wind turbines is used to evaluate the approach. The achieved results are promising, the system can identify its operational state without an explicit mapping.

Related Ontology Subjects/Areas/Topics (16)

Artificial Intelligence	Enterprise Ontologies	Knowledge-Based Systems
Cognitive Systems	Evolutionary Computing	Ontologies
Computational Intelligence	Formal Methods	Simulation and Modeling
e-Business	Industrial Applications of AI	Soft Computing
Enterprise Engineering	Knowledge Representation and Reasoning	Symbolic Systems
Enterprise Information Systems

Exposing Design Mistakes During Requirements Engineering by Solving Constraint Satisfaction Problems to Obtain Minimum Correction Subsets

Topics: Intelligent User Interfaces; Constraint Satisfaction; Industrial Applications of AI; Knowledge-Based Systems

Authors: Alexander Diedrich ¹ , Björn Böttcher ¹ and Oliver Niggemann ²

Affiliations: ¹ Fraunhofer IOSB-INA, Germany , ² Institute for Industrial IT, Germany

Digital Object Identifier: 10.5220/0005679102800287

Proceedings of ICAART 2016, ISBN: 978-989-758-172-4

Quick Abstract

In recent years, the complexity of production plants and therefore of the underlying automation systems has grown significantly. This makes the manual design of automation systems increasingly difficult. As a result, errors are found only during production, plant modifications are hindered by not maintainable automation solutions and criteria such as energy efficiency or cost are often not optimized. This work shows how utilizing Minimum Correction Subsets (MCS) of a Constraint Satisfaction Problem improves the collaboration of automation system designers and prevents inconsistent requirements and thus subsequent errors in the design. This opens up a new field of application for constraint satisfaction techniques. As a use case, an example from the field of automation system design is presented. To meet the automation industry’s requirement for standardised solutions that assure reliability, the calculation of MCS is formulated in such a way that most constraint solvers can be used without any extensions. Experimental results with typical problems demonstrate the practicalness concerning runtime and hardware resources.

Related Ontology Subjects/Areas/Topics (9)

Artificial Intelligence	Human-Computer Interaction	Knowledge-Based Systems
Constraint Satisfaction	Industrial Applications of AI	Soft Computing
Enterprise Information Systems	Intelligent User Interfaces	Symbolic Systems

Evaluating Learning Algorithms for Stochastic Finite Automata - Comparative Empirical Analyses on Learning Models for Technical Systems

Topics: Exact and Approximate Inference; Regression; Inductive Learning; Computational Learning Theory; Stochastic Methods

Authors: Asmir Vodenčarević ¹ , Alexander Maier ² and Oliver Niggemann ³

Affiliations: ¹ University of Paderborn, Germany , ² OWL Universitiy of Applied Sciences, Germany , ³ OWL Universitiy of Applied Sciences and Fraunhofer IOSB – Competence Center Industrial Automation, Germany

Digital Object Identifier: 10.5220/0004255702290238

Proceedings of ICPRAM 2013, ISBN: 978-989-8565-41-9

Quick Abstract

Finite automata are used to model a large variety of technical systems and form the basis of important tasks such as model-based development, early simulations and model-based diagnosis. However, such models are today still mostly derived manually, in an expensive and time-consuming manner. Therefore in the past twenty years, several successful algorithms have been developed for learning various types of finite automata. These algorithms use measurements of the technical systems to automatically derive the underlying automata models. However, today users face a serious problem when looking for such model learning algorithm: Which algorithm to choose for which problem and which technical system? This papers closes this gap by comparative empirical analyses of the most popular algorithms (i) using two real-world production facilities and (ii) using artificial datasets to analyze the algorithms’ convergence and scalability. Finally, based on these results, several observations for choosing an appropriate automaton learning algorithm for a specific problem are given.

Related Ontology Subjects/Areas/Topics (7)

Computational Learning Theory	Pattern Recognition	Stochastic Methods
Exact and Approximate Inference	Regression	Theory and Methods
Inductive Learning

Visual Anomaly Detection in Production Plants

Topics: Knowledge-Based Systems Applications; Intelligent Fault Detection and Identification; Engineering Applications

Authors: Alexander Maier ¹ , Tim Tack ¹ and Oliver Niggemann ²

Affiliations: ¹ OWL University of Applied Sciences, Germany , ² OWL University of Applied Sciences and Fraunhofer IOSB-INA, Germany

Digital Object Identifier: 10.5220/0004039600670075

Proceedings of ICINCO 2012, ISBN: 978-989-8565-21-1

Quick Abstract

This paper presents a novel method for visual anomaly detection in production plants. Since the complexity of the plants and the number of signals that have to be monitored by the operator grows, there is a need of tools to overcome the information overflow. The human is highly able to recognize irregularities in figures. More than 80% of the perceived information is captured visually. The approach proposed in this paper exploits this fact and subjects data to make the operator able to find anomalies in the displayed figures. In three steps the operator is lead from the visualization of the normal behavior over the anomaly detection and the localization of the faulty module to the anomalous signal.

Related Ontology Subjects/Areas/Topics (9)

Artificial Intelligence and Decision Support Systems	Informatics in Control, Automation and Robotics	Knowledge-Based Systems Applications
Engineering Applications	Intelligent Control Systems and Optimization	Robotics and Automation
Enterprise Information Systems	Intelligent Fault Detection and Identification	Signal Processing, Sensors, Systems Modeling and Control

ANOMALY DETECTION IN PRODUCTION PLANTS USING TIMED AUTOMATA - Automated Learning of Models from Observations

Topics: Discrete Event Systems; Instrumentation Networks and Software

Authors: Alexander Maier ¹ , Oliver Niggemann ¹ , Roman Just ¹ , Michael Jäger ¹ and Asmir Vodenčarević ²

Affiliations: ¹ OWL Universitiy of Applied Sciences, Germany , ² University of Paderborn, Germany

Digital Object Identifier: 10.5220/0003538903630369

Proceedings of ICINCO 2011, ISBN: 978-989-8425-74-4

Quick Abstract

Model-based approaches are used for testing and diagnosis of automation systems (e.g. (Struss and Ertl, 2009)). Usually the models are created manually by experts. This is a troublesome and protracted procedure. In this paper we present an approach to overcome these problems: Models are not created manually but learned automatically by observing the plant behavior. This approach is divided into two steps: First we learn the topology of automation components, the signals and logical submodules and the knowledge about parallel components. In a second step, a behavior model is learned for each component. Later on, anomalies are detected by comparing the observed system behavior with the behavior predicted by the learned model.

Related Ontology Subjects/Areas/Topics (4)

Discrete Event Systems		Instrumentation Networks and Software		Signal Processing, Sensors, Systems Modeling and Control
Informatics in Control, Automation and Robotics

AutomationML AS A BASIS FOR OFFLINE - AND REALTIME-SIMULATION - Planning, Simulation and Diagnosis of Automation Systems

Topics: Modeling, Simulation and Architectures; Systems Modeling and Simulation

Authors: Olaf Graeser , Barath Kumar , Oliver Niggemann , Natalia Moriz and Alexander Maier

Affiliation: Hochschule Ostwestfalen-Lippe, Germany

Digital Object Identifier: 10.5220/0003537403590368

Proceedings of ICINCO 2011, ISBN: 978-989-8425-75-1

Quick Abstract

The growing complexity of production plants leads to a growing complexity of the corresponding automation systems. Developers of such complex automation systems are faced with two significant challenges: (i) The control devices have to be tested before they are used in the plant. For this, offline- and hardware–in–the loop (HIL) simulations can be used. (ii) The diagnosis functions within the automation systems become more and more difficult to implement; this entails the risk of undetected errors. Both challenges may be solved using a system model, i.e. a joint model of the plant and the automation system: (i) Offline simulations and HIL tests use such models as an environment model and (ii) diagnosis functions use such models to define the normal system behaviour—allowing them to detect discrepancies between normal and observed behavior. System models cannot be modelled by one person in a single development step. Instead, such models must mirror the modularity of modern plants and automation systems. Here, the new standard AutomationML is used as basis for such a modular system model. But a modular system model is only a first step: Both testing and diagnosis require the simulation of such models. Therefore, a corresponding modular simulation system for AutomationML models is presented here; for this, the Functional Mock–Up Unit (FMU) standard is used. A prototypical tool chain and a model factory (MF) is used to show results for this modular testing and diagnosis approach.

Related Ontology Subjects/Areas/Topics (5)

Industrial Engineering		Modeling, Simulation and Architectures		Systems Modeling and Simulation
Informatics in Control, Automation and Robotics		Robotics and Automation

Topic(s):

Title:
ISBN:
Year:
Acronym:
Subject:

Title:
Author:
Affiliation:
Subject:

Name:
Affiliation:
Country:
Conference:
Subject:

Name:
Country:
Subject:

Affiliation

Filter listed results, by