Purpose of this Webpage

This webpage is the on-line version of the MBD4 Deliverable document. This Deliverable brings together articles and tutorials on Diagnosis that have been produced by or through the work of MONET and presents them in one place for ease of reference. This on-line version of the documents contains links to the full text of the materials listed below.

Tutorial Material – MONET Summer School 2003
Tutorial Material – Bridge Workshop’01
Tutorial Material – Bridge day SafeProcess / DX’03
Tutorial Material – Safeprocess’03

TECHNICAL ARTICLE MATERIAL

Technical Article Material – Bridge Workshop’01
Technical Article Material – DX’01
Technical Article Material – DX’02
Technical Article Material – Bridge Day at SafeProcess / DX’03
Technical Article Material – DX’03
Technical Article Material – SafeProcess’03
Index
References

TUTORIAL MATERIAL

Abstract: This chapter surveys theories that provide principled approaches to automating the task of diagnosing broken artefacts and presents systems that implement these approaches. The key idea of model-based diagnosis is to explicitly represent the knowledge about a device as a model of the device structure and of the behaviour of its constituents and to organise diagnosis as an inference process based on this model and the observed behaviour. This approach created the demand for and the possibility of developing a rigorous theoretical foundation for automated diagnosis. In particular, this comprises a formal characterisation of the goal and of the inferences that achieve the goal, given model-based predictions and the actual observations of the artifacts behaviour. We argue that diagnosis is becoming a major field of application and an important touchstone for the utility of logical theories and AI.

9. Fault Diagnosis Based on Analytical Models for Linear and Non-Linear Systems - A Tutorial. Michael Kinnaert

Abstract: The diagnosis systems considered in this paper rely on the inconsistency between the actual process behaviour and its expected behaviour as described by an analytical model. The inconsistency is exhibited in signals called residuals. Two methods for residual generation are presented in a tutorial way: the parity space and the observer based approaches. Linear and nonlinear models are successively considered as a basis for the design of the residual generators. Keywords: Fault detection and isolation, linear and nonlinear systems, observer, analytical redundancy, residual generation, parity space method.

10. Fundamentals of Model-based Diagnosis. Johan de Kleer and James Kurien

Abstract: Over the last 25 years, the Computer Science community and particularly the Artificial Intelligence community have developed a framework for system diagnosis, called Model-Based Diagnosis. This framework is extremely general and covers a broad range of capabilities including detecting malfunctions, isolating faulty components, handling multiple faults, identifying repair actions, and automatically generating embedded software. This field grew independently of the fault detection and isolation community (FDI) and has developed its own terminologies and conventions. This paper is an attempt to present the fundamental concepts of Model-Based Diagnosis (MBD) in one place and in one consistent terminology, and thus make the field much more accessible to the FDI community. Keywords: Model-based diagnosis, consistency-based diagnosis, constraint-propagation, qualitative models, probabilistic inference, analytical redundancy relations.

11. Model-based Statistical Signal Processing and Decision Theoretic Approaches to Monitoring. Michèle Basseville

Abstract: Model-based statistical signal processing and decision theoretic methods for FDI are presented. It is explained what the basic issues of averaging, sensitivity, correlation, projection, rejection, aggregation, signal-to-noise ratio, information, distance . . . . . bring into the investigation and the analysis of FDI problems. Some emphasis is put on the design of tools which perform the early warning task required for fatigue prevention, aided control and condition-based maintenance. The application of a statistical approach to the vibration monitoring problem is discussed. Keywords: FDI, likelihood, information, sensitivity, projection, rejection, component faults, small deviations, early detection, local approach, vibration monitoring.

Abstract: Two distinct and parallel research communities have been working along the lines of the Model-Based Diagnosis approach: the FDI community and the DX community that have evolved in the fields of Automatic Control and Artificial Intelligence, respectively. This paper clarifies and links the concepts that underlie the FDI analytical redundancy approach and the DX logical approach. The formal match of the two approaches is demonstrated and the theoretical proof of their equivalence is provided under various assumptions.

Abstract: Fault detection via parameter estimation relies in the principle that possible faults in the monitored system can be associated with specific parameters and states of the mathematical model of the system given in the form of an input-output relation:

where y(t) represents the output vector of the system, u(t) the input vector, x(t) the state variables which are partially measurable, q the non measurable parameters which are likely to change on the occurrence of a fault, and e(t) the modeling errors and/or noise terms affecting the process.

22. Mixing Chronicle and Petri Net Approaches in Evolution Monitoring Problems. Armen Aghasaryan and Christophe Dousson

Abstract: This paper is dedicated to a mixed approach of two formalisms used in telecommunication networks and system monitoring. The first one is based on the chronicle recognition that detects relevant pieces of evolution, whereas the second one, is based on the Petri nets and is take into account the complete behavior of the system. We propose a cooperative and complementary use of both approaches.

23. Application of Multiple Sliding Time Windows to Fault Detection Based on Interval Models.Joaquim Armengol, Josep Vehí, Louise Travé-Massuyès and Miguel Ángel Sainz

Abstract: Interval models may be used in many cases to express the imprecision and the uncertainty related to complex systems. The envelopes may be used to represent the results of the simulation of these models. One of the applications of the envelopes is as reference behaviour for Fault Detection (FD) based on analytical redundancy. In this case, the properties of the envelopes (completeness, soundness) have important consequences on the results of the FD, like missed or false alarms. This paper presents the Modal Interval Simulator (MIS), which approaches the FD problem by means of errorbounded envelopes, i.e. by the simultaneous computation of an overbounded envelope and an underbounded one. Modal Interval Analysis, which provides tools to compute interval extensions of real functions with the adequate semantics, is used for computing these envelopes. The MIS system uses multiple sliding time windows for performing FD. This allows the detection of faults of different kinds avoiding (provided that some assumptions are fulfilled) false alarms.

24. Principles of Distributed Diagnosis of Discrete-Event Systems. Gianfranco Lamperti and Marina Zanella

Abstract: This paper introduces the notion of distributed modelbased diagnosis (DMBD) of discrete-event systems (DESs), which is supported by six principles. The first principle establishes the class of systems which are relevant to DMBD. The second principle allows for a variety of observers, each watching the system under a different view. The third principle considers complex observations, that is, structures of (uncertain) observable events. According to the fourth principle, the diagnostic process is performed in a distributed, possibly parallel, way. The fifth principle requires the diagnostic process to be supported by some optimization criteria. The last principle supposes that candidate diagnoses be given incrementally, at different computational time points. Such abstract principles are then substantiated in the domain of active systems, which have been the focus of the authors? research over the last few years and whose diagnosis involves the generation of the system behavior, namely the active space, based on the given observation. In order to fulfill the fifth principle, an algebra is introduced for the manipulation of active spaces, so as to optimize the behavior reconstruction process. For meeting the sixth principle, a method is envisaged for progressively refining the set of candidate diagnoses at each processing step.

25. Diagnosis of Discrete-Event Systems: the Method and an Example. Jan Lunze, Jochen Schroder and Peerasan Supavatanakul

Abstract: This paper describes an approach for detecting and identifying faults that occur in dynamical systems with discrete-valued inputs and outputs. The models used in the diagnosis are described by stochastic automata. The diagnostic problem is to determine the smallest possible set of faults that is compatible with the measured input and output sequences. The diagnostic problem is shown to be, in principle, an observation problem, so that it can be solved by an extended observation method. The effectiveness of the diagnostic method is illustrated by diagnosing valve faults in a batch process.
Keywords : discrete-event system, stochastic automata, diagnosis, diagnosability, batch process.

26. Efficient Diagnosis of Hybrid Systems Using Models of the Supervisory Controller. Sriram Narasimhan and Gautam Biswas

Abstract: This paper presents a model-based approach to diagnosis of hybrid systems. We have developed a combined qualitative/quantitative diagnosis scheme that uses hybrid models of the system and a model of the supervisory controller. By applying the supervisory controller model to diagnostic analysis we significantly cut down on the complexity in tracking behaviors, and in generating and refining hypotheses across discrete mode changes in the system behavior. We present the algorithms for hybrid diagnosis: hypotheses generation by back propagation, and hypotheses refinement by forward propagation and parameter estimation. Example scenarios demonstrate the effectiveness of this approach.

27. A General Framework for Fault Diagnosis Based on Statistical Hypothesis Testing. Mattias Nyberg

Abstract: A framework for fault diagnosis, called structured hypothesis tests, is presented. It has earlier been developed within the area of automatic control, but is in fact very much inspired by the ideas developed in the AI area. The motivation was originally to handle dynamic systems with noise. However, it is here shown that also the noise-free case can be perfectly handled. The system to be diagnosed, and also the different faults, are described by differential equations, algebraic equations, and probability distribution functions. By using the framework, it is in the isolation possible to utilize all such modeled knowledge about the faults. The diagnosis system is constructed by combining a set of different hypothesis tests. In this way, the task of diagnosis is transferred to the task of validating a set of
different models with respect to the measured data.

28. Consistency-Based Diagnosis of Dynamic Systems Using Quantitative Models and Off-Line Dependency-Recording. Belarmino Pulido, Carlos Alonso and Felipe Acebes

Abstract: For more than ten years different techniques have been proposed to perform model-based diagnosis of dynamic systems from the Artificial Intelligence community. Nevertheless, there is no general framework yet. Main part of the research effort has been devoted to modeling issues. Most approaches have relied upon qualitative models due to the lack of accuracy, certainty and precision in quantitative models. Hence, one question arises, is still possible to use quantitative models in the Artificial Intelligence approach to model-based diagnosis? Despite of mentioned drawbacks, quantitative models offer some advantages. If combined with pre-compiled dependency-recording, these systems avoid one of the traditional problems in the qualitative modeling approach, the feedback loop problem. These are the bases of MORDRED, a model-based diagnosis system that combines quantitative models and the possible conflict concept. This work presents results obtained in MORDRED verification and validation phases. Moreover, it analyses drawbacks found during the whole design and implementation cycle, and proposed solutions.

29. From Tracking Continuous Mode Hypotheses to Diagnosing Technical Systems. Bernhard Rinner

Abstract: As most technical systems such as plants, automobiles and robots are getting more and more complex, the need for automatic monitoring and diagnosis of such systems is steadily increasing. Technical systems provide many challenges for a monitoring and diagnosis systems, whereas the most important ones are: First, technical systems are typically dynamic, i.e., they change their state while the monitoring and diagnosis system is processing its input. Second, the supervised system is in most cases not completely known, and an exact model of that system may not be specified. Third, the observations can only provide an in complete view on the supervised system due to discrete sampling, limited observability and noise.
In this paper, we present our model-based approach to monitoring and diagnosing technical systems. We address the challenges by (i) modeling the supervised system as a hybrid system and (ii) tracking continuous mode hypotheses. We have used this approach to implement a self-calibrating monitoring system which starts with a coarse description of the supervised sys tem and exploits the observation to refine the behavior prediction and the underlying model. We discuss important issues to extend self-calibrating monitoring to diagnosing technical systems.
Keywords: semi-quantitative reasoning; hybrid systems; tracker; self-calibrating monitoring; diagnosis.

30 Diagnosis of a Class of Discrete Event Systems Based on Parameter Estimation of a Modular Algebraic Model. Gernot Schullerus and Volker Krebs

Abstract: This paper presents a diagnosis approach suitable for modular discrete event processes where the evolution of the event times can be represented by means of the max-plus-algebra. We first outline a modeling formalism for such processes that results in a hierarchical modular algebraic model. Based on this model, the hierarchical diagnosis algorithm first detects the faulty module and then the faulty component in this module using parameter estimation. The principle and performance of the algorithm is illustrated by the diagnosis of valve faults in a batch process.

31. Model-based Diagnosability and Sensor Placement. Application to a Frame 6 Gas Turbine Subsystem. Louise Travé-Massuyès, Teresa Escobet and Robert Milne

Abstract: It is commonly accepted that the requirements for maintenance and diagnosis should be considered at the earliest stages of design. For this reason, methods for analysing the diagnosability of a system and determining which instrumentation is needed to achieve the desired level of diagnosability, are highly valued. This paper presents a method for:

32. Particle Filters for Real Time Planetary Rovers. Richard Dearden and Daniel Clancy

Abstract: Planetary rovers provide a considerable challenge for artificial intelligence in that they must operate for long periods autonomously, or with relatively little intervention. To achieve this, they need to have on-board fault detection and diagnosis capabilities. Traditional model-based diagnosis techniques are not suitable for rovers due to the tight coupling between the vehicle's performance and its environment. Hybrid diagnosis using particle filters is presented as an alternative, and its strengths and weaknesses are examined. We also present some extensions to particle filters that are designed to make them more suitable for use in diagnosis problems.

Abstract: Applying model-based diagnosis techniques to systems that exhibit hybrid behaviour presents an interesting set of challenges that mostly revolve around interactions of the continuous and discrete components of the system. In many real world systems, the overall physical plant is inherently continuous, but system control is performed by a supervisory controller that imposes discrete switching behaviours by reconfiguring the system components, or switching controllers. In this paper, we present a case study of an aircraft fuel system, and discuss methodologies for building system models for online tracking of system behaviour and performing fault isolation and identification. Empirical studies are performed on detection and isolation for a set of pump and pipe failures.

Abstract: This paper presents the Probabilistic General Diagnostic Engine (PGDE), a novel method of offline consistency-based fault isolation. Many existing proposals require qualitative logic models for consistency-based diagnosis due to their ability to speed the search for conflict sets through the use of an ATMS. However, for many applications, quantitative dynamic models are preferred or already available. The key strength of the PGDE is that it allows the use of any modelling language for which an appropriate calculation engine can be written. It also offers graceful degradation in the presence of uncertainty, commonly caused by noise or modelling errors. Finally, given perfect knowledge, it can be shown that the PGDE computes the same result as existing consistency-based diagnosis methods. To demonstrate the performance of the algorithm, we have used a quantitative dynamic model of the fluid power circuit of a single degree of freedom hydraulic test bench and developed an appropriate calculation engine for computing consistency between measured values and predicted results. Various failures were generated on the physical test bench and the PGDE isolated the faults with approximately 85% accuracy.

Abstract: When designing model-based fault-diagnosis systems, the use of consistency relations (also called e.g. parity relations) is a common choice. Different subsets are sensitive to different subsets of faults, and thereby isolation can be achieved. This paper presents an algorithm for finding a small set of submodels that can be used to derive consistency relations with highest possible diagnosis capability.
The algorithm handles differential-algebraic models and is based on graph theoretical reasoning about the structure of the model. An important step, towards finding these submodels and therefore also towards finding consistency relations, is to find all minimal structurally singular (MSS) sets of equations. These sets characterize the fault diagnosability. The algorithm is applied to a large nonlinear industrial example, a part of a paper plant. In spite of the complexity of this process, a small set of consistency relations with high diagnosis capability is successfully derived.

Abstract: A set S that has a non-empty intersection with every set in a collection of sets C is called a hitting set of C. If no element can be removed from S without violating the hitting set property, S is considered to be minimal. Several interesting problems can be partly formulated as ones that a minimal hitting set or more ones have to be found. Many of these problems are required for proper solutions, but sometimes the approximate solutions are enough. A genetic algorithm and advantaged algorithms were devised for computing minimal hitting sets. An improvement makes them get most minimal hitting sets efficiently. Furthermore, they are smaller, i.e. fewer rules.

Abstract: A novel capability of discrete model-based diagnosis methods is the ability to handle unknown modes where no assumption is made about the behaviour of one or several components of the system. This paper incorporates this novel capability of model-based diagnosis into a hybrid estimation scheme by calculating partial filters. The filters are based on causal and structural analysis of the specified components and their interconnection within the hybrid automaton model. Incorporating unknown modes provides a robust estimation scheme that can cope, unlike other hybrid estimation and multi-model estimation schemes, with unmodelled situations and partial information.

Abstract: In this paper we propose a component-based hybrid formalism, that represents physical phenomena by combining concurrent automata with continuous uncertain dynamic models. The formalism eases the modelling of complex physical systems, and adds concurrency to the supervision of hybrid systems. Uncertainties in the model are integrated as probabilities at the discrete level and intervals at the continuous level. Our modelling framework is rather generic while focusing on the construction of intelligent autonomous supervisors by integrating a continuous/discrete interface able to reason on-line in any region of the physical system state-space, for behaviour simulation, diagnosis and system tracking.

Abstract: Consistency-based diagnosis is the most widely used approach to model-based diagnosis within the Artificial Intelligence community. It is usually carried out through an iterative cycle of behaviour prediction, conflict detection, and candidate generation and refinement. Many approaches to consistency-based diagnosis have relied on some kind of on-line dependency-recording mechanism for conflict calculation. These techniques have had different problems, especially when applied to dynamic systems. Recently, off-line compilation of dependencies has been established as a suitable alternative approach. In this work we compare one compilation technique, based on the possible conflict concept, with results obtained with the classical on-line dependency recording engine as in GDE. Moreover, we compare possible conflicts with another compilation technique coming from the FDI community, which is based on analytical redundancy relations. Finally, we study the relationship between possible conflicts, analytical redundancy relations, and conflicts.

Abstract: Monitoring gains importance for many technical systems such as robots, production lines or anti lock brakes. A monitoring system for technical systems must be able to deal with incomplete knowledge of the supervised system, to process noisy observations and to react within predefined time windows. This paper presents a new approach to monitoring technical systems based on imprecise models. Our approach repeatedly partitions the uncertainty space of an imprecise model and checks the derived model's state for consistency with the measurements. Inconsistent partitions are then refuted resulting in a smaller uncertainty space and a faster failure detection. This paper further focuses on the extension of our basic approach to monitoring systems that exhibit both continuous and discrete behaviours. Our monitoring system has been implemented using COTS components and has been demonstrated in online monitoring of a non-trivial heating system.

Abstract: This paper describes an approach based on supervised learning techniques for the diagnosis of dynamic systems. The methodology can start with real system data or with a model of the dynamic system. In the second case, a set of simulations of the system is required to obtain the necessary data. In both cases, obtained data will be labelled according to the running conditions of the system at the gathering data time. Label indicates the running state of system: correct working or abnormal functioning of any system component. After being labelled, data will be treated to add additional information about the running of system. The final goal is to obtain a set of decision rules by applying a classification tool to the set of labelled and treated data. This way, any observation on the system will be classified according to those decision rules, having a return label indicating the currently running state of system. Returned label will be the diagnostic. This entire learning task is carried out off-line, before the diagnosing.

42. Fault Diagnosis of Dynamical Systems Based on State-set Observers. Jan Lunze, T. Steffen and U. Riedel

Abstract: This paper shows how state-set observation can be used to detect and identify faults in a dynamical systems with bounded measurement uncertainties. A state-set observer uses the process model and the measured input and output sequence to determine the set of states the process can be in. For diagnosis, a bank of observers is used based on the process models for every fault case. If an observer arrives at an empty set of states, the measured behaviour is inconsistent with the model of the corresponding fault and this fault is known to be not present. The remaining faults form the set of fault candidates. To further distinguish these, fault probabilities are calculated. A method for determining conditional probabilities from the observation sets is developed. This allows to combine the diagnostic strength of state-set observation and stochastic observation. Keywords: state-set observer, stochastic observer, dedicated observer scheme, consistency-based diagnosis, stochastic diagnosis.

43. Model Refinement for Monitoring - Refutation vs. Traditional Parameter Estimation. Andreas Doblander, Bernhard Rinner and Ulrich Weiss

Abstract: Model-based monitoring and diagnosis systems must be able to express and reason with incomplete knowledge. However, it is desired to refine the imprecision in the underlying model when more measurements from the supervised system are avail able. Imprecision is often specified by intervals of model parameters. In this paper we compare the two refinement methods refutation and parameter estimation in the context of monitoring. Refutation removes parts of the parameter intervals that are provable inconsistent with the measurements. Parameter estimation, on the other hand, searches for exact paranieter values that best match the measurements. This comparison is supported by various experiments with the refutation-based refinement implemented in the MOSES monitoring system and the MATLAB system identification toolbox.
Keywords: model—based monitoring; parameter estimation; interval models; system identification.

44. Discrete Event System Diagnosis using Parameter Estimation Methods. Joachim Fox, Gernot Schullerus, Matthias Schwaiger and Volker Krebs

Abstract: This paper presents a new diagnosis method for discrete event systems represented by general timed Petri nets. Parameter estimation is used to determine estimates for the holding times of the Petri net's places. The residuals between these estimated values and given nominal values serve as fault indicators and can be used for fault isolation. It is shown that a method developed earlier for timed event graphs is a special case of this new estimation algorithm. Copyright © 2003 IFAC.
Keywords: Fault diagnosis, discrete event systems, Petri nets, parameter estimation.

Abstract: Fault diagnosis is a critical task for autonomous operation of systems such as spacecraft and planetary rovers, and must often be performed on-board. Unfortunately, these systems frequently also have relatively little computational power to devote to diagnosis. For this reason, algorithms for these applications must be extremely efficient, and preferably anytime. In this paper we introduce the Gaussian particle filter (GPF), a new variant on the particle filtering algorithm specifically for non-linear hybrid systems. Each particle samples a discrete mode and approximates the continuous variables by a multivariate Gaussian that is updated at each time-step using an unscented Kalman filter. The algorithm is closely related to Rao-Blackwellized Particle Filtering and equally efficient, but is more broadly applicable. We demonstrate the algorithm on a Mars rover problem and show that it is faster and more accurate than traditional particle filters.

46. Multi-Modal Particle Filtering for Hybrid Systems with Autonomous Mode Transitions. Stanislav Funiak and Brian Williams

Abstract: Model-based diagnosis of embedded systems relies on the ability to estimate their hybrid state from noisy observations. This task is especially challenging for systems with many state variables and autonomous transitions. We propose a fair sampling algorithm that combines Rao-Blackwellised particle filters with a multi-modal Gaussian representation. In order to handle autonomous transitions, we let the continuous state estimates contribute to the proposal distribution in the particle filter. The algorithm outperforms purely simulational particle filters and provides unification of particle filters with hybrid hidden Markov model (HMM) observers.
Keywords: state estimation, hybrid modes, Monte Carlo method, analytic approximations, Kalman filters, filtering techniques, diagnostic inference.

47. A Robust Method for Hybrid Diagnosis of Complex Systems. Sriram Narasimhan, Gyula Simon, Gautam Biswas, Nagabhusan Mahadevan, John Ramirez and Gabor Karsai

Abstract: The AI model-based diagnosis community has developed qualitative reasoning mechanisms for fault isolation in dynamic systems. Their emphasis has been on the fault isolation algorithms, and little attention has been paid to robust online detection and symbol generation that are essential components of a complete diagnostic solution. This paper discusses a robust diagnosis methodology for hybrid systems that combines fault detection with a combined qualitative and quantitative fault isolation scheme. We focus on fault detection, symbol generation, and parameter estimation, and illustrate the effectiveness of this method by running experiments on the fuel transfer system of aircraft. © 2003 IFAC. Keywords: Fault detection, Fault isolation, Qualitative analysis, Parameter estimation, Robust performance.

48. Application of Causal Graph GP for Description of Diagnosed Process. Jan Koscielny and Andrzej Ostasz

Abstract: The paper considers application of causal graph to description of diagnosed process. Presented graph, called Graph of Process (GP), is a qualitative model of the diagnosed process with respect to faults. The graph is used for designing the model structures for fault detection and identifying of fault- symptom relations. Theoretic background of graph GP has been presented as well as an example based on three tank system.
Keywords: directed graphs, fault detection, fault isolation.

49. A Logical Framework for Isolation in Fault Diagnosis. Stéphane Ploix, Samir Touaf, and Jean-Marie Flaus

Abstract: After some remarks on terminology, this paper introduces a general method for fault diagnosis in complex dynamic systems, which takes advantage of the results on analytical redundancy methods from the Automatic Control community and on logical reasoning from the Artificial Intelligence community. The proposed method tackles both the problem of diagnosing complex dynamic systems with models of normal and abnormal behavior including differential equations, and the problem of providing logically sound diagnosis. Moreover, it is shown how diagnoses can be sorted.
Keywords: diagnostic reasoning, analytical redundancy relation, fault detection and isolation, dynamic systems, terminology.

50. Combining AI, FDI, and Statistical Hypothesis-testing in a Framework for Diagnosis. Mattias Nyberg and Mattias Krysander

Abstract: A new framework for model-based diagnosis is presented using ideas from AI, FDI, and statistical hypothesis testing. The isolation mechanism is based on AI methods, and the main advantage is that multiple faults are handled implicitly. Thus, no special care for isolation of multiple faults is needed. The methods for residual generation, developed in the field of control theory (FDI), can within the framework be fully utilized. Since the framework is also based upon statistical hypothesis testing, it is suitable for problems including noise. Keywords: fault diagnosis, AI-methods, fault isolation, FDI-methods, multiple faults, noise.

51. Diagnosability Analysis Based on Component Supported Analytical Redundancy Relations. Louise Travé-Massuyès, Teresa Escobet and S. Spanache

Abstract: It is commonly accepted that the requirements for maintenance and diagnosis should be considered at the earliest stages of design. For this reason, methods for analysing the diagnosability of a system and determining which instrumentation is needed to achieve the desired level of diagnosability, are highly valued. This paper enhances the model based method proposed in Travé-Massuyès, et al. (2001) based on the concept of component supported analytical redundancy relations, which considers recent results crossing over the FDI and DX communities. Copyright © 2003 IFAC.
Keywords: Diagnosability, Sensor Placement, Analytical redundancy, Model-based Diagnosis, Structural analysis.

52. Failure Diagnosis of Stochastic Automata. David Thorsley and Demosthenis Teneketzis

Abstract: This paper extends the methodology of (Sampath et at, 1995) initally developed for logical finite-state machines to stochastic automata. Using probabilistic information, two notions of diagnosability weaker than that of Sampath et al. are defined, and a stochastic analogue of the logical diagnoser is constructed. The stochastic diagnoser is used to (i) specify off-line conditions sufficient to guarantee our notions of diagnosabihty; and (ii) determine how to perform on-line diagnosis of failure events.
Keywords: discrete-event systems, stochastic automaton, fault diagnosis, failure detection, probabilistic models

53. Deriving Qualitative Deviations from MATLAB™ Models. Luca Console, Gianluca Correndo and Claudia Picardi

Abstract: Modelling is still the critical bottleneck for a wide application of model-based diagnosis. This is especially critical in domains where quantitative models of subsystems are developed during the design process (especially for simulating control strategies) and thus building a second (qualitative) model for diagnosis is not acceptable. In this paper we propose a semi-automatic approach for deriving a special type of qualitative model (namely, we derive models based on deviations, which proved to be very powerful when applied to the automotive domain, especially for on-line on-board diagnosis) from quantitative models developed within the Matlab toolset.
We also discuss a prototype implementation of the approach and some experiments we made with a model of the Common Rail fuel injection system, where we show that the model extracted in an automatic way can be effectively used for on-board diagnosis.
Keywords: diagnosis, model-based reasoning, qualitative reasoning.

54. Enhancing Consistency-based Diagnosis with Machine Learning Techniques in the Content of Supervision. Carlos Alonso, Juan Rodriguez and Belarmino Pulido

Abstract: This paper propose a diagnosis architecture that integrates consistency-based diagnosis with induced time series classifiers, trying to combine the advantages of both methods. Consistency-based diagnosis allows fault detection and localization without prior knowledge of the device fault modes. Machine learning techniques are able to induce time series classifiers that may be used to identify fault modes of a dynamic systems. The diagnostician performs fault detection and localization resorting to consistency-based diagnosis trough possible conflicts. Then, a time series classifier, induced from simulated examples, generates a sequence of faults modes, coherent with the result of the fault localization stage, and ordered by fault modes confidence.
Finally, to simplify the diagnosis task, it is considered as a subtask of a supervisory system, who is in charge of identifying the physical system working conditions.
Keywords: supervision, diagnosis of dynamic systems, machine learning, hybrid diagnosis system, time series classifiers, model-based diagnosis.

55. Continous Diagnosis of Discrete-Event Systems. Gianfranco Lamperti and Marina Zanella

Abstract: This paper presents a diagnostic technique that subsumes two complementary approaches to diagnosis of discrete-event systems (DESs), namely the diagnoser approach and the active system approach. The more significant shortcomings of such approaches are, on the one side, the need for the generation of the global system model and, on the other, the completion of the system observation before starting the diagnostic task. The former makes the application of the diagnoser approach prohibitive in real contexts, where the system model is too large to be generated, even off-line. The latter makes it
impossible to the active system approach to perform diagnosis while monitoring. The proposed technique, which overcomes such limitations, copes with a generalized class of DESs that integrate both synchronous and asynchronous behavior.
Keywords: Diagnosis, Monitoring, Discrete-event systems

56. Computation of Fault Detection Delay in Discrete-Event Systems. Tae-Sic Yoo and Humberto Garcia

Abstract: The notion of diagnosability based on failure-event specifications is revisited. We present a modified version of diagnosability in which terminating faulty traces are handled differently. We also introduce the notion of language-diagnosability based on failure-language specifications that generalizes diagnosability based on failure-event specifications. A polynomial-time algorithm for verifying language-diagnosability is developed. Building upon the verification algorithm, we introduce a polynomial-time algorithm for computing the worst case diagnosis delay of a given system. Despite of significant practical importance, this delay computation has not been previously considered in the literature. The computation of the worst case diagnosis delay involves the shortest path computation of a weighted, directed graph. We exploit a special weighting structure of the graph resulting from the verification algorithm, which enables an algorithm with a lower complexity than the commonly used Bellman-Ford shortest path algorithm.
Keywords: Discrete-Event Systems, Fault Diagnosis, Fault Detection Delay

57. Identification of Timed Discrete-Event Model for Diagnosis. C. Falkenberg, Peerasan Supavatanakul and Jan Lunze

Abstract: This paper deals with the discrete–event modelling of dynamical systems whose inputs and outputs can only be qualitatively measured. The timed automaton is used as the discrete–event representation of such systems. The diagnosis is based on the investigation whether the qualitative
input / output sequences obtained from the system are consistent with the timed automaton. It shows how the timed automaton can be set up by qualitative identification using a set of measurement data and how the information included in the timed automaton is used for fault detection and identification. The results are applied to an industrial actuator.
Keywords: Identification, model–based diagnosis, discrete–event system, timed automata.

58. An Integration of FDI and DX Approaches to Polynomial Models using Gröbner Bases. Rafael Gasca, C. Del Valle, R. Ceballos and Miguel Toro

Abstract: In engineering applications, many models are a set of polynomial constraints. In order to automate and improve the diagnosis of these models, we propose a new approach for the integration of both FDI and DX approaches. It allows us to achieve a synergy that produces results that could not be obtained if each one was operating individually.
This paper uses Gröbner bases to generate a more single model of the systems. First, it eliminates the non-observable variables of the constraints of the model and afterwards, we construct a context network with the minimal possible conflicts and polynomial constraints in the nodes of this network. Our methodology proposes an algorithm that uses different aspects of FDI and DX approaches to obtain the minimal diagnosis. This novel approach may be very useful for on-boarding diagnosis. Copyright © 2002 IFAC.
Keywords: Diagnosis, Polynomial models, Model-based diagnosis, Gröbner bases.

59. Automatic Qualitative Model Abstraction from Numeric Simulation Model. Yuhong Yan

Abstract: This paper presents a simulation-based method to automatically transform quantitative model in CAD environment into qualitative model accepted by modelbased diagnosis tools. This approach is inspired by the discriminability analysis for multiple behavior modes (Struss et al., 2002). Unlike other methods that begin with fine domain partitioned model, our approach is a refine process. New landmarks are generated until the target modes are concluded to be discriminable or nondiscriminable. The condition of discrminability is given. This approach is applied to analyze complex CAD model for automobile industry with good results.
Keywords: Diagnosis, Model.

60. A Novel Framework for Integrating Discrete Event System Control and Diagnosis. Gregory M Provan

Abstract: This paper describes a framework that provides a clear semantics and well-developed algorithms for both the control and diagnostics of discrete event Systems. This framework integrates important characteristics of two discrete event system representations, finite state machines (used primarily for control) and causal networks (used primarily for diagnosis). We provide a clear semantics for system models that incorporates control and diagnosis, extending the framework of both finite state machines and causal networks. We show mappings between the two representations, and demonstrate the enhanced power of this integrated representation over the existing representations.
Keywords: Model-based diagnosis, discrete event systems, control.

62. Isolation Decision for a Multi-agent-based Diagnostic System. Stéphane Ploix, Sylvie Gentil and S. Lesecq

Abstract: The variety of diagnostic methods proves that none can pretend to be much better than the others. A significant improvement of industrial applications can only be achieved when FDI problems are dealt with and solved in a framework of integrated use of different FDI methods. This paper presents a study about the possible cooperation between detection methods. Two different isolation methods, which analyze symptoms provided by observer-based and signal-based detection algorithms are compared using a two water tank system: one method is based on signature tables and the other one is based on a logical approach. Strategic aspects are also stressed. Copyright © 2003 IFAC.
Keywords: Complex systems, Fault diagnosis, Fault Detection and Isolation, State Observer, Signal-based diagnostic.

63. Residual Generators for DAE Systems Utilizing Minimal Subsets of Model Equations. Jonas Biteus, and Mattias Nyberg

Abstract: A common approach to design diagnostic systems is to use residual generators. These generators are usually constructed considering all the model equations. However, there are several advantages of instead consider small subsets of model equations, so called minimal structurally singular (MSS) sets of equations. This paper presents a new method for finding residual generators for MSS sets. A special property of the MSS set, namely that it is minimally over determined, is utilized. Two approaches are considered, one which is based on the use of a dynamic numerical equation solver, and another which uses a static numerical equation solver. The approaches are demonstrated on a non-linear point-mass satellite system. Copyright © 2002 IFAC.
Keywords: Fault diagnosis; Residual generator; Fault detection; Fault isolation; Diagnosis; Structural analysis; Differential equations.

64. Robust Fault Detection Using Linear Interval Observers.Viçenc Puig, Joseba Quevedo, Teresa Escobet, and Alexandru Stancu

Abstract: The problem of robustness in fault detection using observers has been treated basically using the active approach, based on decoupling the effects of the uncertainty from the effects of the faults on the residual. On the other hand, the passive approach is based on propagating the effect of the uncertainty to the residuals and then using adaptive thresholds. In this paper, the passive approach based on adaptive thresholds produced using a model with uncertain parameters bounded in intervals, also known as an "interval model", will be presented in the context of linear observer methodology, deriving their corresponding interval version. Finally, an example based on an industrial actuator used as an FDI benchmark in the European project DAMADICS will be used for testing the proposed approach. Copyright © 2003 IFAC.
Keywords: Fault Detection, Fault Diagnosis, Robustness, Observers, Adaptive Threshold.

65. Complementing an Interval Based Diagnosis Method with Sign Reasoning in the Automotive Domain.Xavier Olive, Louise Travé-Massuyès and Jerome Thomas

Abstract: In the automotive field, the use of an ECU (Electronic Control Unit) to control several functions (such as engine injection or ABS) increases. To diagnose such systems, diagnosis trees are built. These trees allow the garage mechanics to find the faulty component(s) by performing a set of tests (measurements) which has the lowest global cost as possible. Two methods, interval based and sign consistency based, which compute diagnosis of electronic circuits are presented, their different features are outlined and it is shown how they can beneficially complement each other.
Keywords: diagnosis, sign and interval reasoning, electrical circuits, test sequencing problem.

66. Fault Detection and Isolation of Rain Gauges and Limnimeters of Barcelona's Sewer System using Interval Models. Viçenc Puig, Joseba Quevedo, J. Figueras, S. Riera, G. Cembrano, M. Salamero and G. Wilhelmi

Abstract: In this paper, fault detection and isolation of rain gauges and limnimeters (water level sensors in the sewers) of Barcelona's urban sewer system is presented. The Barcelona urban drainage network has a telemetry network containing 22 rain gauges and more than 100 limnimeters used for the control system. In order to detect and isolate faulty instruments and to reconstruct faulty measurements from data fusion, a fault diagnosis system is necessary. The proposed fault diagnosis strategy is based on building an interval linear model for every instrument (Puig, 2002). Then, while the real measure is inside the interval of predicted behaviour (or envelope) generated using its interval model no fault can be indicated. However, when the measure is outside its envelope a fault can be indicated. Fault isolation is based on the matching of the real fault signature with the theoretical fault signature using structured residuals (Gertler, 1998). Copyright © 2003 IFAC.
Keywords: Fault Detection, Fault Diagnosis, Robustness, Fault-tolerance, Adaptive Threshold.

67. FCC Diagnosis using Several Causal and Knowledge Based Models. Bruno Heim, Sylviane Gentil, Benoit Celse, Sylvie Cauvin, and Louise Travé-Massuyès

Abstract: In order to deal with the complexity of the diagnosis of FCC pilot plants, several modelling approaches were developed, combined and tested on-line. Two causal modelling approaches were investigated based on control loop analysis and on the detailed equations describing the behaviour of the process. These models are used online to detect faults on process variables. Information on the components of the system allows faults on physical components to be isolated. Then using expert knowledge, information is given to the operator. This paper details the different kinds of models, their use in the diagnosis module and a case study on the IFP FCC pilot plant. This work is conducted as a part of the Chem project 1. Copyright © 2003 IFAC.
Keywords: Model based diagnosis, causal graphs, knowledge based system.

68. A Qualitative Case-based Approach for Situation Assessment in Dynamic Systems. Application in a Two Tank System. Joan Colomer, Joaquim Melendez and F. Ignacio Gamero

Abstract: This work is focused on evaluation of symptoms for situation assessment. Symptoms are described as episodes representing qualitative trends. The aim is to reproduce reasoning and evaluation performed by experienced operators observing signal evolution. The use of qualitative representations of signal trends is proposed to manage previous experiences as cases under a Case Based Reasoning (CBR) approach. The paper discusses how to adapt CBR methodology in performing supervisory tasks taking advantage of previous experiences. Special interest is focused on proposing a case definition and a matching method based on qualitative representations. The usefulness of the approach is shown in an application example.
Keywords: Case Based Reasoning, Qualitative Analysis, Supervision.

69. Fault Detection in a Pilot Plant using Interval Models and Multiple Sliding Time Windows. Joaquim Armengol, Josep Vehí, Miguel Ángel Sainz and Pau Herrero

Abstract: Analytical redundancy is one of the techniques that can be used for Fault Detection. An important problem in this case is how the uncertainty associated to the systems and the measurements is taken into account. This paper proposes to consider them by means of interval models and interval measurements. The consistency between them is checked and a fault is detected when there is an inconsistency thus avoiding false alarms. The used technique is also based on Modal Interval Analysis which provides tools to compute interval extensions of real functions with the adequate semantics and saves much computational effort compared to other techniques based on global optimization algorithms. Time windows of different lengths are used in order to improve the Fault Detection results. This method is being applied to several real processes within the European project CHEM.
Keywords: Fault detection, Intervals, Uncertain dynamic systems, Redundancy, Processes.

70. Empirical Validation of a Decision Procedure Based on Temporal Band Sequences Analysis for Thermo-fluid Processes. R. Mrani Alaoui, B. Ould Bouamama, and Patrick Taillibert

Abstract: The safe operation of process engineering plants rests on Fault Detection and Isolation procedures which mainly consist in evaluating residuals. In the decision step (comparison to 0) these procedures lead to false alarms because of the presence of noise in the measurements which have to be derivated. A decision procedure based on Temporal Band Sequences was proposed in order to overcome this difficulty. This paper evaluates the procedure in comparing its results to those obtained with classical approaches on an analytical redundancy relation extracted from a steam generator.
Copyright © 2003 IFAC.
Keywords: Diagnosis, Analytical Redundancy Relations, Temporal Band Sequences, Process engineering.

71. Fault Diagnosis of Stochastic Automata Networks. Jan Lunze and Jochen Schröder

Abstract: Discrete event systems can be represented as a composite system where the subsystems and the interconnection structure are explicitly described. Such structured models, like the stochastic automata networks considered in this paper, have a much lower complexity than the overall system model. However, in most of the analysis and control methods, such a composite model is first transformed into a single model of very high complexity and methods that have been developed for unstructured models are applied. This paper shows that diagnostic tasks can be decomposed in such a way that the calculation steps to be performed concern merely the subsystem automata and no complete composition of the overall automaton is necessary. In this way, the complexity of the diagnosis of stochastic automata networks is considerably reduced. The results are illustrated by an example. Copyright © 2003 IFAC.
Keywords: fault diagnosis, discrete-event systems, automata networks, component-oriented modelling.

72. Online Monitoring of Hybrid Systems using Imprecise Models. Bernhard Rinner and Ulrich Weiss

Abstract: This paper presents a model-based monitoring system which is based on imprecise models where the structure is known and the parameters may be imprecisely specified by numerical intervals. This monitoring approach is applied to hybrid systems and is now able (i) to follow a known sequence of imprecisely modeled modes, (ii) to detect unknown transitions and (iii) to refine the time uncertainty of the transitions as well as the imprecision of mode models. The implemented system is demonstrated by online monitoring of a non-trivial heating system.
Keywords: model-based monitoring; hybrid systems; imprecise models; parameter Estimation.

73. Robust Diagnosis of Switching Systems. Sherif Abdelwahed, Gabor Karsai and Gautam Biswas

Abstract: This paper presents an approach for robust diagnosis of switching systems based on an extended version of the timed failure propagation graph model. The extended failure propagation graph model is a labelled graph used for the representation of failure conditions and their propagation modelled as causal relations with timing properties for a general class of systems with both timebased and event-driven dynamics such as hybrid and discrete event systems. We introduce the extended model and describe the structure and main components of the failure detection and isolation system based on the proposed model.
Keywords: Failure diagnosis, Switching Hybrid Systems, Robust Diagnosis.

74. Comparison of Interval Models and Quantised Systems in Fault Detection with Application to the DAMADICS Actuator Benchmark Problem. Viçenc Puig, Joseba Quevedo, Alexandru Stancu, Jan Lunze, J. Neidig, P. Planchon and Peerasan Supavatanakul

Abstract: Model-based fault detection relies on checking the discrepancy between the measurements obtained from a monitored process and its model. In general, the model representing the process behaviour has substantial uncertainties. In this contribution, two fault detection methods that explicitly take the modelling uncertainties into account are presented and compared. The first approach uses the interval model while the second tackles the process to be monitored as a quantised system. The main ideas of both approaches are discussed and the results of the fault detection using each approach will be analysed and compared with respect to the DAMADICS actuator benchmark. Copyright © 2003 IFAC.
Keywords: Fault Detection, Qualitative Approaches, Robustness, Interval Models, Quantised Systems.

75. Passive Robust Fault Detection using Nonlinear Interval Observers: Application to the DAMADICS Benchmark Problem. Alexandru Stancu, Viçenc Puig, Joseba Quevedo and Ron Patton

Abstract: In this paper, the passive approach to robust fault-detection based on adaptive thresholds produced using a model with uncertain parameters bounded in intervals, also known as an "interval model", will be presented in the context of observer methodology, deriving their corresponding interval version. Moreover, some non-linearities present in the real system will be included in the structure of the model in order to improve the accuracy of the predicted behaviours. This will lead to the design of non-linear interval observers. These observers provide at every time instant an interval bounding the uncertainty on system states. Finally, this approach will be applied to detect some of the faults proposed in an industrial actuator used as an FDI benchmark in the European project DAMADICS. Copyright © 2003 IFAC.
Keywords: Fault Detection, Robustness, Non-linear Models, Observers, Intervals.

A
Advantaged Algorithms … 36
Aircraft Fuel System … 33, 47
Algebra / Algebraic Models … 24, 27, 30, 35
Analytical Redundancy … 9, 10, 12, 17, 23, 39, 51, 69, 70
Automata … 25, 38, 52, 57, 61
Stochastic… see Stochastic Automata
Timed … see Timed Automata
Automated Diagnosis… see Automatic Control / Monitoring
Automatic Control / Monitoring … 1, 12, 27, 29
Autonomous Systems … 32, 38, 45, 46

C
Case-based Reasoning … 68
Causal Graph … 48, 67
Chronicle Recognition … 22
Control, Automatic … see Automatic Control / Monitoring
Consistency-based Diagnosis … 10, 34, 39, 42, 54, 65
Consistency Relations … 35

D
Dedicated Observer Scheme … 42
Dependency-Recording … 28
Directed Graph … 56
Discrete Events / Systems … 24, 25, 30, 37, 38, 40, 44, 52, 55, 56, 71, 73
Distributed Model-based Diagnosis (DMBD) … 24
Dynamic Models / Systems … 25, 27, 28, 29, 34, 38, 39, 40, 41, 42, 47, 49, 54, 57, 60, 68

G
Gas Turbines … 31
Gaussian Particle Filter … see Particle Filter, Gaussian
General Diagnosis Engine (GDE) … 39
          Also see … Probabilistic General Diagnosis Engine (PGDE)
Genetic Algorithms … 36
Graph … 48, 56, 67, 73
          Causal … see Causal Graph
          Directed … see Directed Graph
          Propagation … see Propagation Graph

L
Learning Techniques … 41
Logical Approach Theories … 1, 12

M
Machine Learning … 54
MATLAB … 43, 53
Minimal Structural Singular (MSS) Sets … 35
Minimal Hitting Sets … 36
Modal Interval Simulator (MIS) … 23
Model
          Algebraic … see Algebraic Models
          Behavioural … see Behavioural Models
          Hybrid … see Hybrid Models / Systems
          Interval … see Interval Models
          Knowledge-based … see Knowledge-based Models
          Polynomial … see Polynomial Models
          Probabilistic … see Probabilistic Model
Model Abstraction … 59
Monitoring Systems … 40, 72
Monte Carlo Method … 46
MOSES Monitoring … 43

O
On-Board Diagnosis Systems … 32, 45, 53, 58
On-Line … 33, 38, 40, 47, 52, 53

P
Paper Plant … 35
Parameter Estimation … 14, 30, 43, 44, 47
Parity Space … 5, 9
Particle Filters … 32, 45
Particle Filter, Gaussian… 45
Petri Nets … 22
Planetary Rovers … 32, 45
Polynomial Models … 56, 58
Probabilistic General Diagnosis Engine (PGDE) … 34
Probabilistic Model … 52
Propagation Graph … 73

R
Rail Fuel Injection System … 53
Redundancy, Analytical … see Analytical Redundancy
Refutation … 43

S
Semi-Qualitative Reasoning … 29
Self-Calibrated Monitoring … 29
Sensor Placement … 31, 51
Sliding Time Windows … 23, 69
State Estimation … 46
State-set Observers … 42
Statistical Hypothesis Testing … 27, 50
Stochastic Automata … 25, 52
Stochastic Diagnosis … 42
Stochastic Observer … 42
Structural Analysis … 51, 63
Supervisory Controller system… 26, 33, 54
System Behaviour … 24, 33
System Monitoring … 14, 22, 43

1. Dressler O. and Struss P. The Consistency-based Approach to Automated Diagnosis of Devices. Proceedings of MONET Summer School, Hersonissus, Crete, September 8 - 12, 2003.

2. Struss P. Model-based Diagnosis. Proceedings of MONET Summer School, Hersonissus, Crete, September 8 - 12, 2003.

3. Travé-Massuyés L. Bridging AI and Control Engineering Model-based Diagnosis Approaches. Proceedings of MONET Summer School, Hersonissus, Crete, September 8 - 12, 2003.

4. Staroswiecki M. Diagnosis Terminology. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

5. Staroswiecki M. Tutorial on FDI Approaches. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

6. Struss P. Tutorial on DX Approaches. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

7. Escobet T. and Travé-Massuyès L. Parameter Estimation for Fault Detection on the Tank Benchmark. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

8. Lunze J. and Schroeder J. Diagnosis of Stochastic Automata. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

9. Kinnaert M. Fault Diagnosis Based on Analytical Models for Linear and Non-Linear Systems - A Tutorial. Presented at the Bridge Day at SafeProcess/DX'03. In Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 13 - 26. And in Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 37 - 50.

10. de Kleer J. and Kurien, J. Fundamentals of Model-based Diagnosis. Presented at the Bridge Day at SafeProcess/DX'03. In Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 1 - 12. And in Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 25 - 36.

11. Basseville M. Model-based Statistical Signal Processing and Decision Theoretic Approaches to Monitoring. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes (SAFEPROCESS 2003). Washington, D.C., U.S.A., June 2003. p. 1 - 12.

12. Cordier M-O., Dague P., Dumas M., Levy F., Montmain J., Staroswiecki M. and Trave-Massuyes L. A comparative Analysis of AI and Control Theory Approaches to Model-Based Diagnosis. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

13. Nyberg M. Structured Hypothesis Tests: Aspects bridging DX and FDI. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

14. Escobet T. and Travé-Massuyès L. Parameter Estimation for Fault Detection on the Tank Benchmark. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

15. Theseider Dupré D. Qualitative Deviations and Causal Simulation. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

16. Biswas G., Mosterman P., Manders E., Barnett J., Narasimhan S., Feenstra P. and Yu L. Qualitative Diagnosis with Temporal Causal Graphs. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

17. Armengol J. and Travé-Massuyès L. Fault Detection with Modal Interval Models on the Tank Benchmark. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

18. Mrani Alaoui R., Taillibert P., Staroswiecki M. and Bouamana B. Parity Space and Temporal Bound Sequence Models. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

19. Zanella M. Diagnosis of Active Systems: Concepts and Tools. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

20. Cordier M.-O. A Decentralized Approach for Diagnosing DES Modelled with Communicating Automata. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

21. Picardi C. and Ribaudo M. System Diagnosis with process algebras. Bridge Workshop held at Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps.

22. Aghasaryan A. and Dousson C. Mixing Chronicle and Petri Net Approaches in Evolution Monitoring Problems. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p.1-8.

23. Armengol J., Vehí J., Travé-Massuyès L. and Ángel Sainz M. Application of Multiple Sliding Time Windows to Fault Detection Based on Interval Models. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p.9-16.

24. Lamperti G. and Zanella M. Principles of Distributed Diagnosis of Discrete-Event Systems. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p. 95-102.

25. Lunze J., Schroder J. and Supavatanakul P. Diagnosis of Discrete-Event Systems: the Method and an Example. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p. 111-118.

26. Narasimhan S. and Biswas G. Efficient Diagnosis of Hybrid Systems Using Models of the Supervisory Controller. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p. 127-134.

27. Nyberg M. A General Framework for Fault Diagnosis Based on Statistical Hypothesis Testing. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p. 135-142.

28. Pulido B., Alonso C. and Acebes F. Consistency-Based Diagnosis of Dynamic Systems Using Quantitative Models and Off-Line Dependency-Recording. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p. 175-182.

29. Rinner B. From Tracking Continuous Mode Hypotheses to Diagnosing Technical Systems. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p. 183-188.

30. Schullerus G. and Krebs V. Diagnosis of a Class of Discrete Event Systems Based on Parameter Estimation of a Modular Algebraic Model. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p. 189-197.

31. Travé-Massuyès L., Escobet T. and Milne R. Model Based Diagnosability and Sensor Placement. Application to a Frame 6 Gas Turbine Subsystem. Proceedings of Twelfth International Workshop on Principles of Diagnosis, March 7-9, 2001, Via Lattea, Italian Alps. p. 205-212.

32. Dearden R. and Clancy D. Particle Filters for Real Time Planetary Rovers. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 1 - 6.

33. Narasimhan S., Biswas G., Karsai G. and Szemethy T. Hybrid Modelling and Diagnosis in the Real World: A Case Study. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 7 - 15.

34. Jones C., Bond G. and Lawrence P. Consistency-Based Fault Isolation for Uncertain Systems with Applications to Quantitative Dynamic Models. Proceedings of Thirteenth International Workshop on Principles of Diagnosis Semmering, Austria, May 2-4, 2002. p. 36 - 42.

35. Krysander M. and Nyberg M. Structural Analysis Utilizing MSS Sets with Application to a Paper Plant. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 51 - 57.

36. Li L. and Yunfei J. Computing Minimal Hitting Sets with Genetic Algorithm. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 65 - 69.

37. Hofbaur M. and Williams B. Hybrid Diagnosis with Unknown Behavioural Modes. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 97 - 105.

38. Benazera E., Trave-Massuyes L. and Dague P. State Tracking of Uncertain Hybrid Concurrent Systems. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 106 - 114.

39. Pulido B. and Alonso C. Possible Conflicts, ARRs, and Conflicts. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 122 - 128.

40. Rinner B. and Weiss U. Model-based Monitoring of Piecewise Continuous Behaviours using Dynamic Uncertainty Space Partitioning. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 146 - 150.

41. Abad P., Suarez A., Gasca R. and Ortega J. Using Supervised Learning Techniques for Diagnosis of Dynamic Systems. Proceedings of Thirteenth International Workshop on Principles of Diagnosis, Semmering, Austria, May 2-4, 2002. p. 165 - 171.

42. Lunze J., Steffen T. and Riedel U. Fault Diagnosis of Dynamical Systems Based on State-set Observers. Presented at Bridge Day at SafeProcess/DX'03. In Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 71 - 78.

43. Doblander A., Rinner B and Weiss U. Model Refinement for Monitoring - Refutation vs. Traditional Parameter Estimation. Presented at Bridge Day at SafeProcess/DX'03. In Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 47 - 52.

44. Fox J., Schullerus G., Schwaiger M. and Krebs V. Discrete Event System Diagnosis using Parameter Estimation Methods. Presented at Bridge Day at SafeProcess/DX'03. In Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 53 - 58.

45. Dearden R. and Hutter F. The Gaussian Particle Filter for Diagnosis of Non-Linear Systems. Presented at Bridge Day at SafeProcess/DX'03. In Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 65 - 70.

46. Funiak S. and Williams B. Multi-Modal Particle Filtering for Hybrid Systems with Autonomous Mode Transitions. Presented at the Bridge Day at SafeProcess/DX'03. In Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 59 - 64.

47. Narasimhan S., Simon G., Biswas G., Mahadevan N., Ramirez J. and Karsai G. A Robust Method for Hybrid Diagnosis of Complex Systems. Presented at the Bridge Day at SafeProcess/DX'03. In Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 41 - 46.

48. KoÊcielny J. and Ostasz A. Application of Causal Graph GP for Description of Diagnosed Process. Presented at the Bridge Day at SafeProcess/DX'03. In Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 879 - 884.

49. Ploix S., Touaf S. and Flaus J-M. A Logical Framework for Isolation in Fault Diagnosis. Presented at the Bridge Day at SafeProcess/DX'03. In Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 885 - 890.

50. Nyberg M. and Krysander M. Combining AI, FDI, and Statistical Hypothesis-testing in a Framework for Diagnosis. Presented at the Bridge Day at SafeProcess/DX'03. In Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 891 - 896.

51. Travé-Massuyès L., Escobet T. and Spanache S. Diagnosability Analysis Based on Component Supported Analytical Redundancy Relations. Presented at the Bridge Day at SafeProcess/DX'03. In Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 897 - 902.

52. Thorsley D and Teneketzis D. Failure Diagnosis of Stochastic Automata. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 131 - 136.

53. Console L., Correndo G. and Picardi C. Deriving Qualitative Deviations from MATLAB™ Models. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 87 - 92.

54. Alonso C., Rodriguez J. and Pulido B. Enhancing Consistency Based Diagnosis with Machine Learning Techniques in the Content of Supervision. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 79 - 86.

55. Lamperti G. and Zanella M. Continuous Diagnosis of Discrete-Event Systems. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003.
p. 105 - 112

56. Yoo T. and Garcia H. Computation of Fault Detection Delay in Discrete-Event Systems. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 207 - 212.

57. Falkenberg C., Supavatanakul P. and Lunze J. Identification of Timed Discrete-Event Model for Diagnosis. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 193 - 198.

58. Gasca R., Del Valle C., Ceballos R. and Toro M. An Integration of FDI and DX Approaches to Polynomial Models using Gröbner Bases. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 153 - 158.

59. Yan Y. Automatic Qualitative Model Abstraction from Numeric Simulation Model. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 199 - 206.

60. Provan G. A Novel Framework for Integrating Discrete Event System Control and Diagnosis. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 187 - 192.

61. Cordier et al. Efficient trajectories computing exploiting inversibility properties. Proceedings of the Fourteenth International Workshop on Principles of Diagnosis, DX'03, Washington D.C., June 2003. p. 93-98.

62. Ploix S., Gentil S. and Lesecq S. Isolation Decision for a Multi-agent-based Diagnostic System. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 489 - 494.

63. Biteus J. and Nyberg M. Residual Generators for DAE Systems Utilizing Minimal Subsets of Model Equations. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 603 - 608.

64. Puig V., Quevedo J., Escobet T. and Stancu A. Robust Fault Detection Using Linear Interval Observers. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 609 - 614.

65. Olive X., Travé-Massuyès L. and Thomas J. Complementing an Interval Based Diagnosis Method with Sign Reasoning in the Automotive Domain. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 615 - 620.

66. Puig V., Quevedo J., Figueras J., Riera S., Cembrano G., Salamero M. and Wilhelmi G. Fault Detection and Isolation of Rain Gauges and Limnimeters of Barcelonaís Sewer System using Interval Models. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 621 - 626.

67. Heim B., Gentil S., Celse B., Cauvin S. and Travé-Massuyès, L. FCC Diagnosis using Several Causal and Knowledge Based Models. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 711 - 716.

68. Colomer J., Melendez J. and Gamero F. A Qualitative Case-based Approach for Situation Assessment in Dynamic Systems. Application in a Two Tank System. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 717 - 722.

69. Armengol J., Vehí J., Sainz M. and Herrero P. Fault Detection in a Pilot Plant using Interval Models and Multiple Sliding Time Windows. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 729 - 734.

70. Mrani Alaoui R., Ould Bouamama B. and Taillibert P. Empirical Validation of a Decision Procedure Based on Temporal Band Sequences Analysis for Thermo-fluid Processes. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 741 - 746.

71. Lunze J. and Schröder, J. Fault Diagnosis of Stochastic Automata Networks. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 825 - 830.

72. Rinner B. and Weiss U. Online Monitoring of Hybrid Systems using Imprecise Models. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, pages 837 - 840, Washington, D.C., U.S.A., June 2003.

73. Abdelwahed S. Karsai G. and Biswas G. Robust Diagnosis of Switching Systems. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes (SAFEPROCESS 2003), Washington, D.C., U.S.A., June 2003. p. 843 - 848.

74. Puig V., Quevedo J., Stancu A., Lunze J., Neidig J., Planchon P. and Supavatanakul P. Comparison of Interval Models and Quantised Systems in Fault Detection with Application to the DAMADICS Actuator Benchmark Problem. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 1191 - 1196.

75. Stancu A., Puig V., Quevedo J. and Patton R. Passive Robust Fault Detection using Nonlinear Interval Observers: Application to the DAMADICS Benchmark Problem. Proceedings of the 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2003, Washington, D.C., U.S.A., June 2003. p. 1197 - 1202.