Peter Struss

ECAI Conference 2014 Conference Paper

Combining Engineering and Qualitative Models to Fault Diagnosis in Air Handling Units

• Peter Struss
• Raymond Sterling
• Jesús Febres
• Umbreen Sabir
• Marcus M. Keane

This paper presents a methodology for model-based fault localization and identification that exploits both numerical (Modelica) models and a qualitative model-based approach to diagnosis. It has been applied to diagnosis of an air handling unit based on data recorded by a building management system. The main steps from model development to diagnosis based on the recorded data are discussed.

ECAI Conference 2010 Conference Paper

A Fault-model-based Debugging Aid for Data Warehouse Applications

• Peter Struss
• Vikas Shivashankar
• Mohamed Zahoor

The paper describes a model-based approach to developing a general tool for localizing faults in applications of data warehouse technology. A model of the application is configured from a library of generic models of standard (types of) modules and exploited by a consistency-based diagnosis algorithm, originally used for diagnosing physical devices. Observing intermediate results can require high efforts or even be impossible, which limits the discriminability between different faults in a sequence of data processing steps. To compensate for this, fault models are used. This becomes a feasible solution for standard modules of a data warehouse application along with a stratification of the data. Fault models capture the potential impact of faults of process steps and data transfer on the data strata as well as on sets of data. Reflecting the nature of the initial symptoms and of the potential checks, these descriptions are stated at a qualitative level. The solution has been validated in customer report generation of a provider of mobile phone services.

ECAI Conference 2008 Conference Paper

A Compositional Mathematical Model of Machines Transporting Rigid Objects

• Peter Struss
• Axel Kather
• Dominik Schneider
• Tobias Voigt

We present models of various elements of a plant that involves the transportation of lumped material. An application context is provided by a project on diagnosing disturbances in food packaging plants and, more specifically, bottling plants. While there exist models of flow of homogeneous matters, such as liquid material in a hydraulic system, based on simultaneous equations of Kirchhoff/Ohm type, in our project we need to cope with non-negligible transportation time of objects and capture phenomena like the tailback of units (if transportation is blocked) or the propagation of gaps in the flow of units. Because the application context requires compositionality of the model, i. e. local, context-free models of the individual transportation elements, we are also facing the problem that whether or not a single element produces an output flow (or accepts an input flow) cannot be determined solely by the model of this element, but only through modeling the interaction with the subsequent element, which may block the output (or the previous one not providing the input). This issue is addressed by modeling the potential of an existing flow distinctly from the actual occurrence of a flow, an idea which also can enhance models of continuous flow.

IJCAI Conference 2007 Conference Paper

• Michael Esser
• Peter Struss

Testing embedded software systems on the control units of vehicles is a safety-relevant task, and developing the test suites for performing the tests on test benches is time-consuming. We present the foundations and results of a case study to automate the generation of tests for control software of vehicle control units based on a specification of requirements in terms of finite state machines. This case study builds upon our previous work on generation of tests for physical systems based on relational behavior models. In order to apply the respective algorithms, the finite state machine representation is transformed into a relational model. We present the transformation, the application of the test generation algorithm to a real example, and discuss the results and some specific challenges regarding software testing.

IJCAI Conference 2007 Conference Paper

• Peter Struss

The paper presents the theoretical foundations and an algorithm to reduce the efforts of testing physical systems. A test is formally described as a set of stimuli (inputs to the system) to shift the system into a particular situation or state, and a set of variables whose observation or measurement refutes hypotheses about the behavior mode the system is operating in. Tests (either generated automatically or by humans) may contain redundancy in the sense that some of its stimuli and/or observables maybe irrelevant for achieving the result of the test. Identifying and dropping them contributes to redu-cing the cost of set-up actions and measurements. We define different kinds of irrelevant stimuli, discuss their practical importance, and present criteria and algorithms for computing reduced tests.

ECAI Conference 2004 Conference Paper

AUTAS: A Tool for Supporting FMECA Generation in Aeronautic Systems

• Claudia Picardi
• Luca Console
• Frederic Berger
• Jan Breeman
• Tony Kanakis
• Jeroen Moelands
• Stephan Collas
• Emmanuel Arbaretier

ECAI Conference 2004 Conference Paper

Models of Behavior Deviations in Model-Based Systems

• Peter Struss

IJCAI Conference 2003 Conference Paper

Automated Qualitative Domain Abstraction

• Martin Sachenbacher
• Peter Struss

Automated problem-solving for engineered devices is based on models that capture the essential aspects of their behavior. In this paper, we deal with the problem of automatically abstracting behavior models such that their level of granularity is as coarse as possible, but still sufficiently detailed to carry out a given behavioral prediction or diagnostic task. A task is described by a behavior model, as composed from a library, a specified granularity of the possible observations, and a specified granularity of the desired results. The goal is to determine partitions for the domains of the variables (termed qualitative values) that are both necessary and sufficient for the task at hand. We present a formalization of the problem within a common relational (constraint-based) framework, present results regarding solutions to task-dependent qualitative domain abstraction, and devise methods for automatically determining qualitative values for a device model.

ECAI Conference 2002 Conference Paper

IDD: Integrating Diagnosis in the Design of automotive systems

• Claudia Picardi
• Rosanna Bray
• Fulvio Cascio
• Luca Console
• Philippe Dague
• David Millet
• Bernd Rehfus
• Peter Struss

ECAI Conference 1994 Conference Paper

Model-Based Diagnosis with the Default-Based Diagnosis Engine: Effective Control Strategies that Work in Practice

• Oskar Dressler
• Peter Struss

AAAI Conference 1994 Conference Paper

Testing Physical Systems

• Peter Struss

We present a formal theory of model-based testing, an algorithm for test generation based on it, and outline how testing is implemented by a diagnostic engine. The key to making the complex task of test generation feasible for systems with continuous domains is the use of model abstraction. Tests can be generated using manageable finite models and then mapped back to a detailed level. We state conditions for the correctness of this approach and discuss the preconditions and scope of applicability of the theory.

ECAI Conference 1992 Conference Paper

Back to Defatilts: Characterizing and Computing Diagnoses as Coherent Assumption Sets

• Oskar Dressler
• Peter Struss

ECAI Conference 1992 Conference Paper

Knowledge-based Diagnosis - An Important Challenge and Touchstone for AI

• Peter Struss

IJCAI Conference 1989 Conference Paper

"Physical Negation"—Integrating Fault Models into the General Diagnostic Engine

• Peter Struss
• Oskar Dressier

The General Diagnostic Engine (GDE) provides an elegant and general framework for model-based diagnosis. However, like many other diagnostic systems, GDE's device models capture only the correct, or intended, behavior of its components. It is lacking an important part of diagnostic reasoning: knowledge about how components may behave when they are faulty. This fact can limit the performance of GDE considerably. We present a solution for integrating the use of fault models into GDE in a very homogeneous way, a system called GDE +. Unlike the basic GDE, it can not only exploit contradictions between the assumed correct behavior of components and the observations, but also analyze whether the faultiness of components would really explain the observations. Based on an extended version of the ATMS, GDE + is able to prove the correctness of components and to rule out implausible diagnostic hypotheses.

AAAI Conference 1988 Conference Paper

Global Filters for Qualitative Behaviors

• Peter Struss

Current methods in qualitative physic sometimes predict behaviors of physical systems that do not correspond to any realvalued solution. One reason is that the merging of distinct behaviors cannot be avoided by local criteria. It is necessary to determine the possible continuations of a qualitative behavior taking into account its complete history. Such global criteria for the partial elimination of spurious solutions are developed for 2nd order differential equations. The application of these filters is shown to reduce the set of behaviors for the mass-spring system predicted by other qualitative physics systems.