No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Diagnosing with a hybrid fuzzy–Bayesian inference approach
Abstract
A diagnosis based on Bayesian theory requires knowledge of the a priori and conditional probabilities of the states of the system being diagnosed. The a priori probabilities are frequently provided nowadays by the manufacturers of these systems. In turn, the probabilities of conditional observations are, as a rule, not available. The question arises as to whether and under what conditions it is possible to substitute conditional probabilities with some aggregate obtainable on the grounds of fuzzy logic. This article responds to this question by proposing a hybrid approach with novelty characteristics in both theoretical and practical terms. In the initial phase of the deliberations, it was concluded that the fundamental difference between Bayesian and fuzzy approaches is that the fuzzy approach considers the uncertainty and lack of precision of observations but overlooks the frequency of observations, and the opposite is true of the Bayesian approach. It therefore seems reasonable to seek the hybridization of both methods so that the Bayesian approach carrying the information regarding the subjective probabilities of faults can be applied in practice. To this end, it has been shown that the probability of a conditional observation can be estimated by calculating the degree of truth of the premise for that observation in the state-specific fuzzy rule. The reminder is devoted to presenting numerical and simulation examples illustrating and verifying the proposed approach.
... Those papers concerns fault detection and isolation methods based inter alia on analytical (Kościelny, Bartyś, & Sztyber, 2021) or datadriven models (Cai et al., 2021;Pulido et al., 2019) or signal analysis approaches (Jian et al., 2021). Model-based approaches are widely used in the process industry (Diedrich & Niggemann, 2022;Kościelny, Bartyś, & Sztyber, 2021;Kościelny et al., 2022;Pulido et al., 2019;Santos et al., 2022), while signal-based approaches are mainly used for diagnosing rotating machinery (Cai et al., 2021;Jose et al., 2021). ...
... Those papers concerns fault detection and isolation methods based inter alia on analytical (Kościelny, Bartyś, & Sztyber, 2021) or datadriven models (Cai et al., 2021;Pulido et al., 2019) or signal analysis approaches (Jian et al., 2021). Model-based approaches are widely used in the process industry (Diedrich & Niggemann, 2022;Kościelny, Bartyś, & Sztyber, 2021;Kościelny et al., 2022;Pulido et al., 2019;Santos et al., 2022), while signal-based approaches are mainly used for diagnosing rotating machinery (Cai et al., 2021;Jose et al., 2021). In terms of fault isolation, classification (Jose et al., 2021;Lee et 2021; Zhang et al., 2021) and automated reasoning (Kościelny, Syfert, & Wnuk, 2021;Kościelny et al., 2022) methods can be distinguished. ...
... Therefore, an important premise of the new approach was the adoption of a three-instead of bi-valued diagnostic signals. This promises significant increase in the fault distinguishability (Kościelny, Bartyś, & Sztyber, 2021;Kościelny et al., 2022). ...
The genesis of the proposed fault isolation approach was the belief that the added value may be achieved through the synthesis of various approaches. In this paper, we propose a new method of fault isolation, which promises the achievement of a favorable effect of synergy. Principally, the proposed approach merges a fault isolation system (FIS) with a modified hitting set tree (HS) inferring algorithm. As a result, a new hybrid approach based on three-valued residuals is presented. It manifests two highly desirable features of diagnostic systems, namely efficient isolability of multiple faults which is inherent for the hitting set tree approach, and a high distinguishability of faults provided by the fault isolation system. The beneficial synergy effect was indicated and confirmed in the example of the diagnosis of a two-tank system. The performed comparative study shows a significant increase in the fault distinguishability obtained with the proposed approach compared with others. The advantageous features of the proposed approach in terms of fault isolability, flexibility of applied models, and the usage of expert knowledge predestinate it for industrial implementations.
... Fault detection is carried out using a variety of models: analytical [1,2], neural [3,4], fuzzy [5,6], and statistical [7]. Only if the analytical models that take into account the impact of a fault on the residuals are known [2] is it possible to determine the relationship between the residuals and the faults. ...
... In practice, models that represent the process state without any fault are usually used. In such cases, there are two options for obtaining knowledge about the impact of faults on residuals: using expert knowledge [1] or learning [8]. The learning technique requires the acquisition of experimental data not only for the normal state, but also for all faulty states that are to This paper focuses on the analysis of reasoning methods based on columns. ...
... The aim was to calculate and compare the diagnostic quality indicators for all analysed algorithms. Using the results of the work [1], the accuracy of diagnosing for both groups of diagnostic reasoning methods was compared. ...
The paper concerns a significant problem in the diagnostics of industrial processes, which is the need to achieve high fault distinguishability. High distinguishability results in the generation of precise diagnoses that enable making appropriate security decisions. In the known approaches, the efforts to obtain high distinguishability are focused on the selection of an appropriate set of generated residuals. The paper presents a new method of diagnostic reasoning using the notation of faults/diagnostic signals’ relations in the form of a Fault Isolation System (FIS), which enables the use of multivalent diagnostic signals. In addition, the proposed method uses knowledge (usually incomplete) about the sequence of symptoms. Reasoning was carried out on the basis of simple, physically possible signatures, resulting from the FIS. Assumptions and a diagnostic algorithm are given. The reasoning algorithm works in a step-by-step manner, after observing further symptoms. In each reasoning step, two diagnoses are generated in parallel. A more accurate, but less certain diagnosis is formulated on the basis of the value of all diagnostic signals, and the diagnosis is less accurate, but more reliable, solely on the basis of symptoms. An example of using the method for diagnosing a set of connected liquid tanks is given. The method was compared with other reasoning methods based on columns (signatures) and, also, with row-based reasoning methods. It is shown that the proposed method allows the increase of the distinguishability of faults compared to other methods. The distinguishability grows with the knowledge of elementary symptom sequences. It is also noted that the proposed approach makes possible diagnosing not only faults, but also cyber attacks.
... To illustrate the properties of the proposed fault distinguishability metrics, we will refer to the relatively simple example of the diagnostic system presented in [12,10]. However, due to the limitations in the paper's volume, the calculation example will be limited solely to an analysis of the fault distinguishability in FSM and FIS excluding a case of diagnosis based on timed sequences of fault symptoms. ...
... A collection of all considered faults is shown in Table 1. By analogy with [10,12], it was assumed that phenomenological partial models would be used for fault detection. In these models, the influence of disturbances and measurement noise was neglected. ...
This paper addresses the problem of assessing fault distinguishability of faults. Fault distinguishability is understood as the ability of a diagnostic system to isolate faults. The objective of diagnosis is the early detection and accurate isolation of the faults that have occurred. The accuracy of diagnosis depends on the fault distinguishability achieved. The problem of assessing the fault distinguishability is therefore extremely important. This provides an opportunity to compare the effectiveness and assess the quality of different solutions of a diagnostic system. This paper proposes two new metrics for fault distinguishability assessment, providing the possibility to evaluate and compare the performance of fault isolation carried out based on both signatures or symptoms with both binary or trinary diagnostic signal values. The proposed metrics allow analysis of fault distinguishability for all diagnostic inference approaches based on Fault Signature Matrix as well on Fault Isolation System.
Keywordsfault distinguishability metricsfault isolationprocess diagnosisdiagnostic accuracydiagnostic inference
... Different researches and studies have been proposed to deal with the diagnosis task of wind turbines using several approaches [3,[5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20]. These proposed methodologies adopt distinctive design of schemes, resulting in different properties according to the used techniques. ...
... 1,2 ∈ ℜ 1 * 1 = 1,2 ∈ ℜ 1 * 1 −̂1 ,2 ∈ ℜ 1 * 1 (16) Where: Y(k) is the output of the pitch system, Ŷ(k) is the observed output (KF or LO). ...
This paper aims to present a robust fault diagnosis structure-based observers for actuator faults in the pitch part system of the wind turbine benchmark. In this work, two linear estimators have been proposed and investigated: the Kalman filter and the Luenberger estimator for observing the output states of the pitch system in order to generate the appropriate residual between the measured positions of blades and the estimated values. An inference step as a decision block is employed to decide the existence of faults in the process, and to classify the detected faults using a predetermined threshold defined by upper and lower limits. All actuator faults in the pitch system of the horizontal wind turbine benchmark are studied and investigated. The obtained simulation results show the ability of the proposed diagnosis system to determine effectively the occurred faults in the pitch system. Estimation of the output variables is effectively realized in both situations: without and with the occurrence of faults in the studied process. A comparison between the two used observers is demonstrated.
... Trivalued signatures results from the G SF (16). They have been used, among others, by Kościelny et al. (2021a) or Kościelny and Bartyś (2021). ...
... To cope with the uncertainty of fault symptoms, a hybrid diagnostic inference method based on the fusion of the Bayesian approach and fuzzy inference was proposed Kościelny et al. (2021a). On the other hand, in the work of Sztyber and Kościelny (2016) the uncertainties of the symptoms were accounted by combining fuzzy logic and the Dempster-Shafer theory. ...
The paper proposes an original, comprehensive, and methodically consistent graph theory-based approach to the description of the diagnosed process and the diagnosing system. The main baseline of the presented approach is the dichotomous approach to diagnosing. It involves a separate description of both the process and the diagnostic system. This approach reflects the practice of designing implementable diagnostic systems. Thus, it can be seen as a proposal of a new, alternative, and, at the same time, flexible design procedure with great potential for applications. The primary motivation behind it was an attempt to circumvent the numerous limitations of well-known and well-established diagnosis approaches proposed by the communities working on fault detection and isolation (FDI) and artificial intelligence theories for diagnosis (DX). Accordingly, the paper identifies and provides an extensive discussion and a critical analysis of the existing limitations. Numerous examples and references to practical applications of the approach are indicated. Keywords: graph of the process, the graph of the diagnostic system, fault detection, and isolation, qualitative models, limitations of diagnostic approaches.
... Fault detection can be carried out using different types of models: analytical [1][2][3], neural [4][5][6], fuzzy [7], and statistical [8,9]. A among them, techniques based on partial parametric models identified in the fault-free state deserve special attention. ...
... A among them, techniques based on partial parametric models identified in the fault-free state deserve special attention. In such cases, there are two options for obtaining knowledge about the impact of faults on residuals: using expert knowledge [1][2][3] or learning [5]. The learning technique requires the acquisition of experimental data not only for the normal state but also for all faulty states that are to be recognized. ...
This paper is concerned with the issue of the diagnostics of process faults and the detection of cyber-attacks in industrial control systems. This problem is of significant importance to energy production and distribution, which, being part of critical infrastructure, is usually equipped with process diagnostics and, at the same time, is often subject to cyber-attacks. A commonly used approach would be to separate the two types of anomalies. The detection of process faults would be handled by a control team, often with a help of dedicated diagnostic tools, whereas the detection of cyber-attacks would be handled by an information technology team. In this article, it is postulated here that the two can be usefully merged together into one, comprehensive, anomaly detection system. For this purpose, firstly, the main types of cyber-attacks and the main methods of detecting cyber-attacks are being reviewed. Subsequently, in the analogy to “process fault”—a term well established in process diagnostics—the term “cyber-fault” is introduced. Within this context a cyber-attack is considered as a vector containing a number of cyber-faults. Next, it is explained how methods used in process diagnostics for fault detection and isolation can be applied to the detection of cyber-attacks and, in some cases, also to isolation of the components of such attacks, i.e., cyber-faults. A laboratory stand and a simulator have been developed to test the proposed approach. Some test results are presented, demonstrating that, similarly to equipment/process faults, residua can be established and cyber-faults can be identified based on the mismatch between the real data from the system and the outputs of the simulation model.
... To combine the fuzzy prior distribution and the likelihood function, fuzzy logic rules can be applied [38]. One common rule used in fuzzy Bayesian inference is the product rule. ...
... Recently, fuzzy logic and the Bayesian approach have been used in the fields such as risk analysis [26], diagnosing [27,28], reducing electricity consumption [29], or DR management considering incomplete information [30]. Moreover, fuzzy intelligence can be also applied to DR scheduling considering load behavior [31]. ...
... In this environment, the performance of sensors and components deteriorates rapidly, which can easily lead to diagnostic model failure and misdiagnosis. Another method is establishing a fuzzy relationship such as Fuzzy Bayesian network [15]. This method can reduce the model's dependence on data volume effectively. ...
Digital twin driven fault diagnosis shows good performance in fault diagnosis of subsea control systems. However, the relation between digital twin and fault diagnosis is not clear. This cannot bring substantial improvement to fault diagnosis. A digital twin driven fault diagnosis method for subsea control system is proposed. Simulink is used for building a digital twin model and a fault diagnosis model is established based on Bayesian networks. The diagnosis results are input into the digital twin model to verify them. The results of verification are feedback to fault diagnosis model. Through this method, a framework that improves fault diagnosis by digital twin is proposed and provide a reference for related research. The performance of this method is verified by a redundant control system. The results show that in this framework, the digital twin model can improve diagnostic performance effectively.
... However, this involves presuming some simplifications and undertaking some assumptions. For example, frequently, the assumption regarding the infallibility of measurement devices or the credibility of observations is adopted [10]. It is a case in both branches of the model-based diagnostics developed by the FDI and DX research communities [11]- [13]. ...
This paper discusses the origin and problem of the fault compensation effect. The fault compensation effect is an underrated common side effect of the fault isolation approaches developed within the Fault Detection and Isolation (FDI) community. In part, this is justified due to the relatively low probability of such an effect. On the other hand, there is a common belief that the inability to isolate faults due to this effect is the evident drawback of model-based diagnostics. This paper shows how, and under which conditions, the fault compensation effect can be identified. In this connection, the necessary and sufficient conditions for the fault compensation effect are
formulated and exemplified by diagnosing a single buffer tank system in open and closed-loop arrangements. In this regard, we also show the drawbacks of a bi-valued residual evaluation for fault isolation. In contrast, we outline the advantages of a three-valued
residual evaluation. This paper also brings a series of conclusions allowing for a better understanding of the fault compensation effect. In addition, we show the difference between fault compensation and fault-masking effects.
الملخص
الهدف من هذا العمل هو اقتراح هيكل تشخيصي وتنبؤي فعال لتور بينات الرياح. للقيام بذلك، يجب تطوير نظام موثوق لتحسين إنتاج طاقة الرياح، وجعله أكثر كفاءة وفعالية. أيضا ألسباب اقتصادية، يجب أن تكون هذه االستراتيجيات التشخيصية والنذير قوية وفعالة في وجود عيوب أثناء التشغيل. سيتم استخدام نموذج تنبؤي باستخدام المراقبين واألنظمة الذكية لتوليد المخلفات على أساس التكرار، ويجب أن يأخذ في االعتبار عيوب الماكينة المحتملة. يستخدم هذا الهيكل لنمذجة التشغيل العادي للعملية، وفي حالة الفشل، للكشف عن العيوب وتصنيفها. سيتم التحقق من صحة هذه االستراتيجية التشخيصية عن طريق المحاكاة على نموذج لتوربينات الرياح األفقية. الكلمات المفتاحية :توربينات الرياح، التشخيص، التنبؤ، المراقب، مرشح كالمان، األعطال، الجهاز العصبي الغموض، السيطرة المتسامحة، اإلشراف
Abstract
The objective of this work is to propose an efficient diagnostic and prognostic structure for wind turbines. To do this, the developed system must be a reliable system to improve the wind energy production, make it more efficient and more effective. Also for economic reasons, these diagnostic and prognostic strategies must be robust and efficient in the presence of defects during operation. A predictive model using observers and intelligent systems will be used for the generation of residues based on redundancy, and must take into account possible machine defects. This structure is used for modeling the normal operation of the process and, in the case of failure, for the detection and classification of defects. A validation of this diagnostic strategy will be done by simulation on a model of a horizontal wind turbine.
Key words: Wind turbine, diagnosis, prognosis, observer, Kalman filter, defects, neuro-fuzzy, tolerant control, supervision.
Résumé
L'objectif de ce travail est de proposer une structure de diagnostic et de pronostic efficace pour les turbines éolienne. Pour cela, le système développé doit ètre fiable pour améliorer la production de l’énergie éolienne, la rendre plus efficace et plus performante. Aussi pour des raisons économiques, ces stratégies de diagnostic et de pronostic doivent être robustes et performantes en présence des défauts au cours de fonctionnement. Un modèle à caractère prédictif en utilisant les observateurs et les systèmes intelligents est utilisé pour la génération des résidus en se basant sur la redondance, et doit prendre en compte les éventuels défauts de la machine. Cette structure est utilisée pour la modélisation du fonctionnement normal du processus et dans le cas défaillant pour la détetion et la classification des défauts. Une validation de cette stratégie de diagnostic sera faite par simulation sur un modèle d’une turbine éolienne de type horizontale.
Mots clé : Turbine Eolienne, diagnostic, pronostic, observateur, Filtre de Kalman, défauts, système neuro-flou, commande tolérante, supervisio
A wind turbine (WT) is an electromechanical system that often operates under a wide range of production conditions. These electrical systems are nowadays expanding rapidly, and they have considerable importance due to their efficiency as renewable energy sources. This led to proposing an innovative and efficient solution with intelligent systems to maintain and ensure the safe and stable operation of these dynamic systems. Maintenance tasks are based on the development of high-performance diagnostic tools, which consist in detecting and locating correctly and upstream the various failures affecting this wind machine. Where, the condition monitoring and supervision systems must rely on reliable fault diagnosis techniques in order to: avoid breakdowns and unscheduled shutdowns, improve their operation, and increase their energetic performances. In order to ensure adequate maintenance actions for the wind system, the purpose of this article is to propose and develop a robust and intelligent fault diagnosis structure. In this work, Kalman filters (KF) as state estimators are used to observe the output states of the sub-systems in order to generate the appropriate residuals evaluated by predetermined thresholds. Adaptive and hybrid network-based fuzzy inference systems (ANFIS) have been employed for the evaluation and classification stages of the detected faults to minimize the degradation of the wind turbine. All possible faults of wind turbine systems, sensors, and actuators are tested and investigated in all parts: pitch angle systems, drive, and generator with converter. The developed fault detection and identification structure are tested on a horizontal WT benchmark model using different scenarios and faults. The simulation results show the ability of the proposed and developed diagnostic methodology to detect the faults occurring efficiently and correctly in the machine. Thus, by using this robust diagnostic strategy, the condition monitoring system can maintain and ensure stable and safe operation to generate sufficient electrical power.
In practical applications, there are usually complex stochastic and vague uncertainties caused by inherent randomness and unknown dynamics. How to model uncertainty and interpret the modeling result is of significance for practical applications. To address these issues, a new probabilistic fuzzy logic system (PFLS) is proposed based on two-stage Bayesian learning. On the basis of the structure of traditional FLS, the proposed PFLS can be used as probabilistic fuzzy interpreter for complex uncertainty modeling. Specifically, in the training stage, expectation–maximization-based Gaussian mixture model is used to classify given data and learn the corresponding probability distributions, which are further used to generate samples to compensate for the limited given data. Based on Bayesian learning, fuzzy rules are learned by maximizing joint probability distributions of given data and rules. In the later inference stage, Bayesian learning-based probabilistic fuzzy inference is developed to make the activated fuzzy rules best match the input and output data pairs. Through integration with the two-stage Bayesian learning, PFLS can properly handle both vague and stochastic uncertainties, and interpret modeling results with fuzzy rules under maximum likelihood. Experiments based on temperature predictions in complex industrial processes demonstrate the effectiveness of the proposed method.
The paper is aimed to discuss the problem of fault isolation robustness. According to the authors, this problem has been so far underestimated and poorly discussed in the community of process diagnostics. It was pointed out that the lack of robustness of fault isolation maybe one of the main reasons limiting the broader application of advanced diagnostic systems. This was a strong motivation to undertake the research and evaluation of the robustness of fault isolation approaches. Therefore, a significant part of the paper was devoted to identifying and discussing the main reasons for the lack of fault isolation robustness. Also, for research and comparative purposes, the term robustness of fault isolation was defined together with a preliminary proposal for its estimation which supposedly may find industrial acceptance. Finally, a heuristic qualitative metric of fault isolation robustness has been proposed. It allows for a quick and straightforward assessment of the fault isolation robustness, which promises its practical usability.
For a control system with bus network topology, it is significant while difficult to accurately recognize and locate the fault of the transmission medium. The reason is that the occurrence of one fault can lead to the abnormality of multiple terminal nodes. How to use these abnormal signals to realize effective fault recognition and fault location is the key to solving the problem. Inspired by the manual inference in engineering practice, this article proposes a belief rule-based fault recognition and location model (BR-FRL) for transmission medium in bus network systems. The core idea is fault location based on the results of fault recognition. To adapt to the flexible expansion of bus topology, the designed BR-FRL model has self-organization characteristics. The highlights of this article include three aspects: 1) the progressive reasoning of fault recognition and location based on belief rules is realized in one model; 2) a new model architecture is proposed, which integrates the belief rules, the group method of data handling (GMDH), and the cautious conjunctive rule (CCR); and 3) the parameter learning based on stochastic gradient descent (SGD) is applied to BR-FRL, and the sensitivity analysis of evidence weight in CCR of BR-FRL is performed. The effectiveness of BR-FRL is verified by an experiment on a gigabit-capable passive optical network (G-PON).
In this work, we outlined the application of the Bayesian technique for integrating the results of multiple tests while treating any disease. We provided an overview of the fundamental concept of Bayesian analysis in making an inference about a phenomenon using a concordance between available information and prior knowledge. An attempt is being made to demonstrate the applicability using core aspects, viz., nature of probability, parameters, and inferential procedure used to draw inference about the population characteristic under both paradigms. Here, we tried to sketch the underlying steps and assumptions regarding use of the Bayes theorem and the analytical techniques that can be used to analyze and interpret medical data while treating any disease.
Reliable and timely fault detection and isolation are necessary tasks to guarantee continuous performance in complex industrial systems, avoiding failure propagation in the system and helping to minimize downtime. Model-based diagnosis fulfils those requirements, and has the additional advantage of using reusable models. However, reusing existing complex non-linear models for diagnosis in large industrial systems is not straightforward. Most of the times, the models have been created for other purposes different from diagnosis, and many times the required analytical redundancy is small. The approach proposed in this work combines techniques from two different research communities within Artificial Intelligence: Model-based Reasoning and Neural Networks. In particular, in this work we propose to use Possible Conflicts, which is a model decomposition technique from the Artificial Intelligence community to provide the structure (equations, inputs, outputs, and state variables) of minimal models able to perform fault detection and isolation. Such structural information is then used to design a grey box model by means of state space neural networks. In this work we prove that the structure of the Minimal Evaluable Model for a Possible Conflict can be used in real-world industrial systems to guide the design of the state space model of the neural network, reducing its complexity and avoiding the process of multiple unknown parameter estimation in the first principles models. We demonstrate the feasibility of the approach in an evaporator for a beet sugar factory using real data.
Under the trend of increasing installed capacity of wind power, the intelligent fault diagnosis of wind turbine is of great significance to the safe and efficient operation of wind farms. Based on the knowledge of fault diagnosis of wind turbines, this paper builds expert system diagnostic knowledge base by using confidence production rules and expert system self-learning method. In Visual Studio 2013 platform, C # language is selected and ADO.NET technology is used to access the database. Development of Fault Diagnosis Expert System for Wind Turbine. The purpose of this paper is to realize on-line diagnosis of wind turbine fault through human-computer interaction, and to improve the diagnostic capability of the system through the continuous improvement of the knowledge base.
The definitions and conditions for fault isolability of single faults for various forms of the diagnostic relation are reviewed. Fault isolability and unisolability on the basis of a binary diagnostic matrix are analyzed. Definitions for conditional and unconditional isolability and unisolability on the basis of a fault information system (FIS), symptom sequences and directional residuals are formulated. General definitions for conditional and unconditional isolability and unisolability in the cases of simultaneous evaluation of diagnostic signal values and a sequence of symptoms are provided. A comprehensive example is discussed.
Failure prediction is essential for predictive maintenance due to its ability to prevent failure occurrences and maintenance costs. At present, mathematical and statistical modeling are the prominent approaches used for failure predictions. These are based on equipment degradation physical models and machine learning methods, respectively. None of these approaches ensures failure predictions well before their occurrence to provide sufficient time to treat potential causes pro actively. Therefore, in this paper, we present a Bayesian based methodology to learn and associate failure signatures with potential failure occurrences. In this approach, event driven maintenance data is used as symptoms which is aggregated on discretized intervals. The failures probabilities as predicted by the Bayesian network are plotted as temporal evolution. This is further exploited to extract either rules or patterns as failure signatures and critical regions. These are then used to monitor and predict the potential failure occurrences. The proposed methodology is tested on the data collected from a well reputed semiconductor manufacturer with promising results.
This paper introduces a set of comprehensive general reasoning rules about single faults based on diagnostic matrix. Rea-soning scheme unifies inference about faults based on conventional binary diagnostic matrix, bi-and three-valued Fault Isolation System as well as on its fuzzy counterparts. They were introduced and defined notions of alternative and domi-nant fault signatures, fuzzy fault signatures as well as a matrix of alternative signatures. This matrix is supposed to be used instead of classic diagnostic matrix. It was also shown that dominant fault signatures are transformable into alternative ones. Finally, three variants of concise general reasoning rules of faults are given. Three examples illustrate key point issues of the paper. First example refers to medical diagnostic case. It shows an example of dominant fault signatures, and in fact proposes a rational approach for planning of diagnostic tests. Second and third examples describe fuzzy reasoning approach making the use of matrix of fuzzy alternative signatures applicable for use with p valued fuzzy diagnostic signals. Future works are outlined in summary section.
Influenza is a viral disease that usually affects the nose, throat, bronchi, and seldom lungs. This disease spreads as seasonal epidemics around the world, with an annual attack rate of estimated at 5%–10% in adults and 20%–30% in children. Thus, influenza is regarded as one of the critical health hazards of the world. Early diagnosis (consisting of determination of signs and symptoms) of this disease can lessen its severity significantly. Examples of signs and symptoms of this disease consist of cough, fever, headache, bireme, nasal congestion, nasal polyps and sinusitis. These signs and symptoms cannot be measured with near-100% certainty due to varying degrees of uncertainties such as vagueness, imprecision, randomness, ignorance, and incompleteness. Consequently, traditional diagnosis, carried out by a physician, is unable to deliver desired accuracy. Hence, this paper presents the design, development and application of an expert system to diagnose influenza under uncertainty. The recently developed generic belief rule-based inference methodology by using the evidential reasoning (RIMER) approach is employed to develop this expert system, termed as Belief Rule Based Expert System (BRBES). The RIMER approach can handle different types of uncertainties, both in knowledge representation, and in inference procedures. The knowledge-base of this system was constructed by using records of the real patient data along with in consultation with the Influenza specialists of Bangladesh. Practical case studies were used to validate the BRBES. The system generated results are effective and reliable than from manual system in terms of accuracy.
In this paper, problems of the online diagnostics of multiple faults in large scale systems are presented. Possibilities of recognition of multiple faults based on the knowledge of signatures of the single faults are discussed. The distinction between simultaneous faults and the series of faults is introduced and defined. The necessary condition for diagnostic reasoning assuming an occurrence of multiple faults is formulated. This paper will show that multiple faults appearing simultaneously or within short time intervals are correctly isolated under the assumption of single faults, if the subsets of faults suitable for isolation are disjunctive. A novel, practical and easy to implement algorithm designed for reasoning multiple faults is presented. An industrial example of the online diagnostics of a steam-water line of the power boiler is given.
Fault-tolerant control aims at a graceful degradation of the behaviour of automated systems in case of faults. It satisfies the industrial demand for enhanced availability and safety, in contrast to traditional reactions to faults that bring about sudden shutdowns and loss of availability. The book presents effective model-based analysis and design methods for fault diagnosis and fault-tolerant control. Architectural and structural models are used to analyse the propagation of the fault throughout the process, to test the fault detectability and to find the redundancies in the process that can be used to ensure fault tolerance. Design methods for diagnostic systems and fault-tolerant controllers are presented for processes that are described by analytical models, by discrete-event models or that can be dealt with as quantised systems. Five case studies on pilot processes show the applicability of the presented methods. The theoretical results are illustrated by two running examples used throughout the book. The book addresses engineering students, engineers in industry and researchers who wish to get a survey over the variety of approaches to process diagnosis and fault-tolerant control. The authors have extensive teaching experience with graduates and PhD students as well as industrial experts. Parts of this book have been used in courses for this audience. The authors give a thorough introduction to the main ideas of diagnosis and fault-tolerant control and present some of their most recent research achievements that they have obtained together with their research groups in a close cooperation within European research projects. The second edition includes new material about reconfigurable control, diagnosis of nonlinear systems, and remote diagnosis. The application examples are extended by a steering-by-wire system and the air path of a diesel engine, both of which include experimental results. The bibliographical notes at the end of all chapters have been up-dated. The chapters end with exercises to be used in lectures.
Full-text available online at http://www.shortliffe.net/Buchanan-Shortliffe-1984/MYCIN Book.htm or at http://aitopics.org/publication/rule-based-expert-systems-mycin-experiments-stanford-heuristic-programming-project
Artificial intelligence, or AI, is largely an experimental science—at least as much progress has been made by building and analyzing programs as by examining theoretical questions. MYCIN is one of several well-known programs that embody some intelligence and provide data on the extent to which intelligent behavior can be programmed. As with other AI programs, its development was slow and not always in a forward direction. But we feel we learned some useful lessons in the course of nearly a decade of work on MYCIN and related programs. In this book we share the results of many experiments performed in that time, and we try to paint a coherent picture of the work. The book is intended to be a critical analysis of several pieces of related research, performed by a large number of scientists. We believe that the whole field of AI will benefit from such attempts to take a detailed retrospective look at experiments, for in this way the scientific foundations of the field will gradually be defined. It is for all these reasons that we have prepared this analysis of the MYCIN experiments.
This paper explains the role of Bayes Theorem and Bayesian networks arising in a medical negligence case brought by a patient who suffered a stroke as a result of an invasive diagnostic test. The claim of negligence was based on the premise that an alternative (non-invasive) test should have been used because it carried a lower risk. The case raises a number of general and widely applicable concerns about the decision-making process within the medical profession, including the ethics of informed consent, patient care liabilities when errors are made, and the research problem of focusing on 'true positives' while ignoring 'false positives'. An immediate concern is how best to present Bayesian arguments in such a way that they can be understood by people who would normally balk at mathematical equations. We feel it is possible to present purely visual representations of a non-trivial Bayesian argument in such a way that no mathematical knowledge or understanding is needed. The approach supports a wide range of alternative scenarios, makes all assumptions easily understandable and offers significant potential benefits to many areas of medical decision-making.
Two distinct and parallel research communities have been working along the lines of the model-based diagnosis approach: the fault detection and isolation (FDI) community and the diagnostic (DX) community that have evolved in the fields of automatic control and artificial intelligence, respectively. This paper clarifies and links the concepts and assumptions that underlie the FDI analytical redundancy approach and the DX consistency-based logical approach. A formal framework is proposed in order to compare the two approaches and the theoretical proof of their equivalence together with the necessary and sufficient conditions is provided.
The fault compensation effect sets an important and up-to-date unsolved theoretical and practical problem for diagnostics of processes. Compensation of influences of faults on residual values has a significant impact on the diagnosis. On the one hand, there is a view that model-based diagnostics does not allow for the fault isolation for which a fault compensation effect occurs. On the other hand, there is a view that application of approaches based on Reiter’s theory allows for the isolation of such faults. The paper shows that these views are not necessarily true. Based on the analysis of a simple system composed of two interconnected tanks, it has been shown that the use of a tri-valued evaluation of residuals allows not only for increasing values of fault distinguishability metrics, but also to take into account the problem of their mutual compensation. In addition, it has been shown that consistency based diagnostics methods developed on the basis of Reiter’s theory may lead to the generation of false diagnoses indicating seemingly compensating faults, even though it is, in fact, impossible due to specific physical limitations of the diagnosed system.
This paper presents an effective method for crack identification to improve the training of Artificial Neural Networks (ANN) parameters using Jaya algorithm. Dynamic and static datasets are introduced using eXtended IsoGeometric Analysis (XIGA) to improve the accuracy of the proposed application based on the frequency and strain measurements. Based on the concept used in our previous works, XIGA provided more accurate results for fracture mechanics applications than other modelling techniques. Therefore, XIGA datasets of cracked plate are used to improve ANN technique for static and dynamic analyses. Model updating of the cracked plate is considered by introducing the mass of accelerometers and identifying Young’s modulus of the plate and stiffness of springs using Jaya algorithm. The difference between measured and calculated frequencies is used as an objective function to calibrate the XIGA model. The crack length is predicted using an adaptive approach without any previous knowledge based on the data provided from a numerical model. Jaya algorithm is used to optimize the most important parameters of ANN. Several numerical examples with different crack scenarios and different boundary conditions are studied in order to evaluate the proposed approach. The results show that the proposed application is able to predict all considered scenarios and accurately identify the crack length. Experimental data of cracked plates are used to validate the numerical predictions. Hence, this application is found to be robust and accurate for crack identification in plates.
The process of learning candidate causal relationships involving diseases and symptoms from electronic medical records (EMRs) is the first step towards learning models that perform diagnostic inference directly from real healthcare data. However, the existing diagnostic inference systems rely on knowledge bases such as ontology that are manually compiled through a labour-intensive process or automatically derived using simple pairwise statistics. We explore CBN, a Clinical Bayesian Network construction for medical ontology probabilistic inference, to learn high-quality Bayesian topology and complete ontology directly from EMRs. Specifically, we first extract medical entity relationships from over 10,000 deidentified patient records and adopt the odds ratio (OR value) calculation and the K2 greedy algorithm to automatically construct a Bayesian topology. Then, Bayesian estimation is used for the probability distribution. Finally, we employ a Bayesian network to complete the causal relationship and probability distribution of ontology to enhance the ontology inference capability. By evaluating the learned topology versus the expert opinions of physicians and entropy calculations and by calculating the ontology-based diagnosis classification, our study demonstrates that the direct and automated construction of a high-quality health topology and ontology from medical records is feasible. Our results are reproducible, and we will release the source code and CN-Stroke knowledge graph of this work after publication.
In this paper, the diagnosis system of power plant gas turbine has been developed to detect the deterioration of engine performance. This system can be analyzed the gas path measurement to predict the deterioration of engine main component by using artificial neural network. The deterioration performance data of gas turbine was generated by using the thermodynamic model. So, the artificial neural network model was built to predict the deteriorated characteristics of gas turbine. Thermodynamic model was used to simulate gas turbine performance as well as the deterioration of engine components (compressor, combustion chamber and turbine) which were represented by changing component characteristic parameters (efficiency and flow capacity). On one hand, the probability of these deteriorated components was simulated to generate deteriorated data (measurement parameters and deterioration degree of each component). On the other hand, the neural network was trained with deterioration data and the best structure of neural network (number of hidden layers, number of neurons in hidden layer and transfer function) was selected based on the minimum value of the mean square error. The different deterioration data (testing data) was generated in thermodynamic model to test the effectiveness of the neural network. The comparison between the mean square error value of single and multi-neural network output parameters at training and testing data were achieved. In final, the testing with the real engine data were achieved.
The best known residual generation methods in model based fault detection and isolation, including parity equations, diagnostic observers and Kalman filtering, are presented in a consistent framework. The discussion is organized along two residual enhancement concepts, namely structured and fixed direction residual sets. It is shown that, once the design objectives are selected, parity equation and observer based designs lead to equivalent residual generators. Robustness in the face of modelling errors is addressed and partially robust residual generator algorithms based on multiple model variants and on partial parameter insensitivity are reviewed.
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.
Bayes' theorem provides a method of inverting conditional probabilities in probability calculus and statistics. Subjective logic generalises probability calculus whereby arguments are represented as opinions that can contain degrees of uncertainty. This paper presents Bayes' theorem in the formalism of subjective logic.
The book that launched the Dempster–Shafer theory of belief functions appeared 40 years ago. This intellectual autobiography looks back on how I came to write the book and how its ideas played out in my later work.
The importance of fault diagnosis activity for the maintenance and continued functioning of system, especially complex electro-mechnical systems such as flight control systems, nuclear reactor control systems, complex machineries and process control systems is well recognized. Automated diagnosis systems are necessary for large systems. But for large complex systems characterized by a large number of subsystems and components. The paper presents the development of 'fault-art' a domain independent expert system for fault diagnosis and reliability studies.
Knowledge about the relation betwen faults and the observed symptoms is necessary for fault isolation. Such a relation can be expressed in various forms, including binary diagnostic matrices or information systems. The paper presents the use of fuzzy logic for diagnostic reasoning. This method enables us to take into account various kinds of uncertainties connected with diagnostic reasoning, including the uncertainty of the faults-symptoms relation. The presented methods allow us to determine the fault certainty factor as well as certainty factors of the normal and unknown process states. The unknown process state factor groups all the states with unknown and multiple faults with the states with improper residual values, while the normal state factor indicates similarity between the observed state and the pattern fault-free state.
This paper addresses the problem of fault detection and isolation of wind turbines using a mixed Bayesian/Set-membership approach. Modeling errors are assumed to be unknown but bounded, following the set-membership approach. On the other hand, measurement noise is also assumed to be bounded, but following a statistical distribution inside the bounds. To avoid false alarms, the fault detection problem is formulated in a set-membership context. Regarding fault isolation, a new fault isolation scheme that is inspired on the Bayesian fault isolation framework is developed. Faults are isolated by matching the fault detection test results, enhanced by a complementary consistency index that measures the certainty of not being in a fault situation, with the structural information about the faults stored in the theoretical fault signature matrix. The main difference with respect to the classical Bayesian approach is that only models of fault-free behavior are used. Finally, the proposed FDI method is assessed against the wind turbine FDI benchmark proposed in the literature, where a set of realistic fault scenarios in wind turbines are proposed.
Fed-batch fermentation process is an effective method for production. Due to the various feeding streams andoperational conditions in different fed-batches, usually it is difficult to formulate a kinetics-based ordinary differential equations model for industrial fed-batch fermentation process. On the other hand, there are plenty of historical data collected during the fermentation process. In this paper, we firstly applied the graphical modeling method to model the fed-batch fermentation process. In this proposed method, the missing data within records are imputed, and then, the correlations between variables are determined by the low order conditional independence method, after that, the parameters of these related variables are learned by the multivariate auto regressive method. The calculation of L-lysine fed-batch fermentation process demonstrates the effectiveness of the proposed approximate model method.
Failure prediction is essential for predictive maintenance due to its ability to prevent failure occurrences and maintenance costs. At present, mathematical and statistical modeling are the prominent approaches used for failure predictions. These are based on equipment degradation physical models and machine learning methods, respectively. None of these approaches ensures failure predictions well before their occurrence to provide sufficient time to treat potential causes pro actively. Therefore, in this paper, we present a Bayesian based methodology to learn and associate failure signatures with potential failure occurrences. In this approach, event driven maintenance data is used as symptoms which is aggregated on discretized intervals. The failures probabilities as predicted by the Bayesian network are plotted as temporal evolution. This is further exploited to extract either rules or patterns as failure signatures and critical regions. These are then used to monitor and predict the potential failure occurrences. The proposed methodology is tested on the data collected from a well reputed semiconductor manufacturer with promising results.
Process disturbances can propagate over entire plants and it can be difficult to locate their root causes from observed effects. Bayesian Networks offer a way to represent unit operations, processes and whole plants as probabilistic models which can be used to infer and rank likely causes from observed effects. This paper presents a methodology to use deterministic steady-state process models to derive Bayesian Networks based on alarm event detection. An example heat recovery network is used to illustrate the model building and inferential procedures.
This paper presents the underlying theory of diagnosing multiple faults with the
binary dynamic diagnostic matrix. The fundamental statements regarding inconsistency and
multiple fault isolation with the dynamic binary diagnostic matrices are formulated and proved.
Finally, an algorithm of multiple fault isolation is presented.
Tunneling excavation is bound to produce significant disturbances to surrounding environments, and the tunnel-induced damage to adjacent underground buried pipelines is of considerable importance for geotechnical practice. A fuzzy Bayesian networks (FBNs) based approach for safety risk analysis is developed in this article with detailed step-by-step procedures, consisting of risk mechanism analysis, the FBN model establishment, fuzzification, FBN-based inference, defuzzification, and decision making. In accordance with the failure mechanism analysis, a tunnel-induced pipeline damage model is proposed to reveal the cause-effect relationships between the pipeline damage and its influential variables. In terms of the fuzzification process, an expert confidence indicator is proposed to reveal the reliability of the data when determining the fuzzy probability of occurrence of basic events, with both the judgment ability level and the subjectivity reliability level taken into account. By means of the fuzzy Bayesian inference, the approach proposed in this article is capable of calculating the probability distribution of potential safety risks and identifying the most likely potential causes of accidents under both prior knowledge and given evidence circumstances. A case concerning the safety analysis of underground buried pipelines adjacent to the construction of the Wuhan Yangtze River Tunnel is presented. The results demonstrate the feasibility of the proposed FBN approach and its application potential. The proposed approach can be used as a decision tool to provide support for safety assurance and management in tunnel construction, and thus increase the likelihood of a successful project in a complex project environment.
© 2015 Society for Risk Analysis.
Diagnosis is the process of identifying or determining the nature and root cause of a failure, problem, or disease from the symptoms arising from selected measurements, checks or tests. The different facets of the diagnosis problem and the wide spectrum of classes of systems make this problem interesting to several communities and call for bridging theories. This paper presents diagnosis theories proposed by the Control and the AI communities and exemplifies how they can be synergically integrated to provide better diagnostic solutions and to interactively contribute in fault management architectures.
Principal component analysis has been widely used in the process industries for the purpose of monitoring abnormal behaviour. The process of reducing dimension is obtained through PCA, while T-tests are used to test for abnormality. Some of the main contributions to the success of PCA is its ability to not only detect problems, but to also give some indication as to where these problems are located. However, PCA and the T-test make use of Gaussian assumptions which may not be suitable in process fault detection. A previous modification of this method is the use of independent component analysis (ICA) for dimension reduction combined with kernel density estimation for detecting abnormality; like PCA, this method points out location of the problems based on linear data-driven methods, but without the Gaussian assumptions. Both ICA and PCA, however, suffer from challenges in interpreting results, which can make it difficult to quickly act once a fault has been detected online. This paper proposes the use of Bayesian networks for dimension reduction which allows the use of process knowledge enabling more intelligent dimension reduction and easier interpretation of results. The dimension reduction technique is combined with multivariate kernel density estimation, making this technique effective for non-linear relationships with non-Gaussian variables. The performance of PCA, ICA and Bayesian networks are compared on data from an industrial scale plant.
Copyright © 2015 ISA. Published by Elsevier Ltd. All rights reserved.
There is an increasing demand for dynamic systems to become more safe and
reliable. This requirement extends beyond the normally accepted
safety-critical systems of nuclear reactors and aircraft where safety is
paramount important, to systems such as autonomous vehicles and fast
railways where the system availability is vital. It is clear that fault
diagnosis (including fault detection and isolation, FDI) has been becoming
an important subject in modern control theory and practice. For example,
the number of papers on FDI presented in many control-related conferences
has been increasing steadily. The subject of fault detection and isolation
continues to mature to an established field of research in control
engineering.
A large amount of knowledge on model-based fault diagnosis has been
accumulated through the literature since the beginning of the 1970s. However,
publications are scattered over many papers and a few edited books. Up to
the end of 1997, there is no any book which presents the subject in an
unified framework. The consequence of this is the lack of ``common
language'', different researchers use different terminology. This problem
has obstructed the progress of model-based FDI techniques and has been
causing great concern in research community. Many survey papers have been
published to tackle this problem. However, a book which presents the
materials in a unified format and provides a comprehensive foundation of
model-based FDI is urgently needed. Such a book would promote the subject
of model-based FDI and make the techniques more accessible for engineers
and research students. This view has been shared by many researchers in
this field. Such a book is also possible, because many important
definitions have been made and the correspondences between different
model-based FDI methods have been established.
This new book presents the subject of model-based FDI in a unified
framework. It contains many important topics and methods, however perfect
coverage and completeness is not the primary concern. The book focuses on
fundamental issues such as basic definitions and the importance of
robustness in FDI approaches. In this book, FDI concepts and methods are
illustrated by either simple academic examples or practical applications.
The first two chapters are of tutorial value and provide a starting point
for new comers to this field. The rest of the book presents the
state-of-the-art in model-based FDI by discussing many important robust
approaches and their applications. This will certainly appeal to experts
in this field.
The book targets both new comers, who want to get into this subject, and
experts, who are concerned with fundamental issues and are also looking
for inspiration for future research. The book is useful for both
researchers in academia and professional engineers in industry because
both theory and applications have been discussed. Although this is a
research monograph, it will be an important text for MSc \& PhD research
students world-wide. The largest market, however will be academics,
libraries and practising engineers and scientists throughout the world.
Fuzziness is explored as an alternative to randomness for describing uncertainty. The new sets-as-points geometric view of fuzzy sets is developed. This view identifies a fuzzy set with a point in a unit hypercube and a nonfuzzy set with a vertex of the cube. Paradoxes of two-valued logic and set theory, such as Russell's paradox, correspond to the midpoint of the fuzzy cube. The fundamental questions of fuzzy theory—How fuzzy is a fuzzy set? How much is one fuzzy set a subset of another?—are answered geometrically with the Fuzzy Entropy Theorem, the Fuzzy Subsethood Theorem, and the Entropy-Subsethood Theorem. A new geometric proof of the Subsethood Theorem is given, a corollary of which is that the apparently probabilistic relative frequency nA/N turns out to be the deterministic subsethood S(X, A), the degree to which the sample space X is contained in its subset A. So the frequency of successful trials is viewed as the degree to which all trials are successful. Recent Bayesian polemics against fuzzy theory are examined in light of the new sets-as-points theorems.
This paper considers a Bayesian approach to fault isolation. Given a set of measurements from the system, and a set of possible faults, the task is to calculate the probability that the faults are present. This probability can then be used to rank the faults, or for decisions on fault sccomodation. The method requires the conditional probability distribution de- scribing how the measurements react to the faults. In particular, the structure of dependencies between the tests is important. Knowing the structure facil- itates efficient computation methods and makes it possible to reduce the memory capacity needed. In this paper, the structure is estimated from training data using Bayesian methods. The method is ap- plied to diagnosis of the gas flow in a diesel engine.
A method of fault isolation in industrial processes is presented: the dynamic table of states (DTS) method. This method allows one to apply various known ways of fault isolation with the help of an object-independent universal algorithm of diagnostic reasoning. It makes it possible to detect and isolate both single and multiple faults, and enables one to formulate a diagnosis even in cases of changes in the set of available signals. An example of the application of the DTS method is also presented.
There is an increasing demand for dynamic systems to become safer, more reliable and more economical in operation. This requirement extends beyond the normally accepted safety-critical systems e.g., nuclear reactors, aircraft and many chemical processes, to systems such as autonomous vehicles and some process control systems where the system availability is vital. The field of fault diagnosis for dynamic systems (including fault detection and isolation) has become an important topic of research. Many applications of qualitative and quantitative modelling, statistical processing and neural networks are now being planned and developed in complex engineering systems. Issues of Fault Diagnosis for Dynamic Systems has been prepared by experts in fault detection and isolation (FDI) and fault diagnosis with wide ranging experience.Subjects featured include: - Real plant application studies; - Non-linear observer methods; - Robust approaches to FDI; - The use of parity equations; - Statistical process monitoring; - Qualitative modelling for diagnosis; - Parameter estimation approaches to FDI; - Fault diagnosis for descriptor systems; - FDI in inertial navigation; - Stuctured approaches to FDI; - Change detection methods; - Bio-medical studies. Researchers and industrial experts will appreciate the combination of practical issues and mathematical theory with many examples. Control engineers will profit from the application studies.
Safe, reliable, and efficient operation of complex dynamical systems requires the ability to detect, isolate, and identify degradation in system compo-nents. Degradations are typically modeled as in-cipient faults, which are slow drifts in system para-meters over time. This paper presents an efficient approach for the detection, isolation, and identifi-cation of incipient faults under uncertainty using a Dynamic Bayesian Network (DBN) approach. Ini-tially a DBN is used as an observer to track nominal system behavior. Once a fault is detected, incipi-ent fault hypotheses are generated using a variation of our qualitative TRANSCEND approach for abrupt fault isolation. A modified DBN that includes the active fault hypotheses is then used to isolate the true fault and estimate the rate of change in its pa-rameter value.
Featuring a model-based approach to fault detection and diagnosis in engineering systems, this book contains up-to-date, practical information on preventing product deterioration, performance degradation and major machinery damage.;College or university bookstores may order five or more copies at a special student price. Price is available upon request.
This paper surveys expert systems (ES) development using a literature review and classification of articles from 1995 to 2004 with a keyword index and article abstract in order to explore how ES methodologies and applications have developed during this period. Based on the scope of 166 articles from 78 academic journals (retrieved from five online database) of ES applications, this paper surveys and classifies ES methodologies using the following eleven categories: rule-based systems, knowledge-based systems, neural networks, fuzzy ESs, object-oriented methodology, case-based reasoning, system architecture, intelligent agent systems, database methodology, modeling, and ontology together with their applications for different research and problem domains. Discussion is presented, indicating the followings future development directions for ES methodologies and applications: (1) ES methodologies are tending to develop towards expertise orientation and ES applications development is a problem-oriented domain. (2) It is suggested that different social science methodologies, such as psychology, cognitive science, and human behavior could implement ES as another kind of methodology. (3) The ability to continually change and obtain new understanding is the driving power of ES methodologies, and should be the ES application of future works.
Bayesian belief networks provide a natural, efficient method for representing probabilistic dependencies among a set of variables. For these reasons, numerous researchers are exploring the use of belief networks as a knowledge representation in artificial intelligence. Algorithms have been developed previously for efficient probabilistic inference using special classes of belief networks. More general classes of belief networks, however, have eluded efforts to develop efficient inference algorithms. We show that probabilistic inference using belief networks is NP-hard. Therefore, it seems unlikely that an exact algorithm can be developed to perform probabilistic inference efficiently over all classes of belief networks. This result suggests that research should be directed away from the search for a general, efficient probabilistic inference algorithm, and toward the design of efficient special-case, average-case, and approximation algorithms.
Suppose one is given a description of a system, together with an observation of the system's behaviour which conflicts with the way the system is meant to behave. The diagnostic problem is to determine those components of the system which, when assumed to be functioning abnormally, will explain the discrepancy between the observed and correct system behaviour.We propose a general theory for this problem. The theory requires only that the system be described in a suitable logic. Moreover, there are many such suitable logics, e.g. first-order, temporal, dynamic, etc. As a result, the theory accommodates diagnostic reasoning in a wide variety of practical settings, including digital and analogue circuits, medicine, and database updates. The theory leads to an algorithm for computing all diagnoses, and to various results concerning principles of measurement for discriminating among competing diagnoses. Finally, the theory reveals close connections between diagnostic reasoning and nonmonotonic reasoning.
A fuzzy set is a class of objects with a continuum of grades of membership. Such a set is characterized by a membership (characteristic) function which assigns to each object a grade of membership ranging between zero and one. The notions of inclusion, union, intersection, complement, relation, convexity, etc., are extended to such sets, and various properties of these notions in the context of fuzzy sets are established. In particular, a separation theorem for convex fuzzy sets is proved without requiring that the fuzzy sets be disjoint.
In the paper, the indeterminacy phenomenon is discussed, that is, a phenomenon having two facets: uncertainty and vagueness. We argue that fuzzy sets are a reasonable mathematical tool for modeling of the latter. The necessary sound foundations of their theory can now be more easily established because of significant progress reached in the formal theory of fuzzy logic. Further direction in the development of fuzzy set theory is also discussed.
Increasing integration densities and high operating speeds are leading to subtle manifestations of defects at the board level. Board-level functional test is therefore necessary for product qualification. The diagnosis of functional failures is especially challenging, and the cost associated with board-level diagnosis is escalating rapidly. An effective and cost-efficient board-level diagnosis strategy is needed to reduce manufacturing cost and time-to-market, as well as to improve product quality. In this paper, we use Bayesian inference to develop a new board-level diagnosis framework that allows us to identify faulty devices or faulty modules within a device on a failing board with high confidence. Bayesian inference offers a powerful probabilistic method for pattern analysis, classification, and decision making under uncertainty. We apply this inference technique by first generating a database of fault syndromes obtained using fault-insertion test at the module pin level on a fault-free board, and then use this database along with the observed erroneous behavior of a failing board to infer the most likely faulty device. Results on a case study using an open-source RISC system-on-chip highlight the effectiveness of the proposed framework in terms of fault-localization accuracy and correctness of diagnosis.
pdf contains a preliminary version of the book
A multivalued mapping from a space X to a space S carries a probability measure defined over subsets of X into a system of upper and lower probabilities over subsets of S. Some basic properties of such systems are explored in Sections 1 and 2. Other approaches to upper and lower probabilities are possible and some of these are related to the present approach in Section 3. A distinctive feature of the present approach is a rule for conditioning, or more generally, a rule for combining sources of information, as discussed in Sections 4 and 5. Finally, the context in statistical inference from which the present theory arose is sketched briefly in Section 6.
In the literature, several fault diagnosis methods, qualitative as well quantitative, are proposed. The main objective of these methods is in one hand, to allow detection, isolation and identification of faults ; and in the other hand to insure safety, reliability and availability of systems. This paper presents a diagnosis method based on the use of a new and suitable mathematical tool : bayesian networks. Their learning and inference capabilities allow to model complex processes by taking into account the uncertainty and the incompleteness of the provided knowledge. Furthermore, the graphical representation of causal relations existing between variables, events or physical phenomena makes bayesian networks easy to use and leads to models which can be understandable by even a non specialist of the modeled domain.
This paper presents a probabilistic method for processing and analyzing residuals for the purpose of fault detection. The method incorporates residuals from multiple models using a hybrid dynamic Bayesian network in order to yield a low-cost, complete, diagnostic system. Continuous residuals are used as evidence directly in the network, and this paper discusses options for representing their probability distributions. The Bayesian network is used to model the temporal behavior of the faults, and the assumptions necessary to do this are analyzed. The diagnostic method is demonstrated on a car's handling system and experimental results are presented.