Michał Syfert

Development and Application of Methods for Actuator Diagnosis in Industrial Control Systems (DAMADICS): A Benchmark Study

Abstract This paper provides the description and presentation of a new actuator benchmark for fault diagnosis studies developed in the framework of the EC-funded Research Training Network DAMADICS. The paper also introduces 3 special sessions of papers providing an evaluation of methods and results focussed on the benchmark problem. The benchmark is FDI method independent and based on an in-depth study of the phenomena that can lead to likely faults in valve actuator systems. The study uses a detailed consideration of the physics and electro-mechanical properties of the valve together with typical engineering requirements of a valve operating under challenging industrial conditions. The benchmark study is available on the internet at http://diag.mchtr.pw.edu.pl/damadics/ and provides a set of performance indices for evaluating the results.

Fuzzy Logic Application to Diagnostics of Industrial Processes

Summary The paper presents application of fuzzy logic for fault detection and isolation in industrial processes, and especially for generation of residuals on the grounds of fuzzy models, residual values evaluation, description of a faults-symptoms relation, as well as diagnostic inference. Advantages of diagnostic algorithms based on fuzzy logic have been taken into considerations.

Current Diagnostics of Power Boiler System with Use of Fuzzy Logic

Abstract The paper presents diagnostics of power boiler system performed with use of the DIAG system. Fuzzy and neural local models of particular systems have been applied for fault detection. Obtained results of control valve and attemperator modelling have been presented. Fault isolation has been performed with use of the F-DTS method that bases on fuzzy appreciation of residuals and fuzzy reasoning. An example of fault isolation in a steam superheater and attemperator has been presented.

Advanced monitoring and diagnostic system 'AMandD'

Abstract The paper presents structure and functional properties of the Advanced Monitoring and Diagnostic System ‘AMandD’ dedicated for large scale industrial processes. The system arose within the framework of research project ‘CHEM’ of 5th European Framework Program. Applied in the system up-to-date algorithms of modelling methods and fault detection and isolation were described in the paper. Also a brief description of used IT solutions was presented. As a conclusion, the further directions of development and researches were mentioned.

CURRENT DIAGNOSTICS OF THE EVAPORATION STATION

Abstract The paper presents structure and algorithms of the system of current diagnosis of the evaporation station in the Lublin Sugar Factory. The applied fault detection algorithms are based on the set of partial parametric models, the set of heuristic dependencies describing process behaviour and the hardware redundancy. The isolation algorithm implements fuzzy reasoning based on the diagnostic rules given by the expert. All the diagnostic algorithms are carried out by the modules of Advanced Monitoring and Diagnostic System AMandD. The achieved fault detectability and isolability analysis and some test results are also presented.

On-Line Actuator Diagnosis Based on Neural Models and Fuzzy Reasoning: The DAMADICS Benchmark Study

Abstract This paper described on-line diagnostic system for the final control elements. The system makes possible early fault detection and isolation. The neural models of actuator components, information system theory with faults-symptoms relation and fuzzy logic are applied. The paper presents results of laboratory and industrial tests. The laboratory tests were done with application of Simulink actuator model. For industrial tests real devices working in technological installation in sugar factory where investigated.

On-Line Fault Isolation Algorithm for Industrial Processes

Abstract The paper presents the solutions of main problems that appear in diagnostics of large scale processes in chemical, petrochemical, pharmaceutical, power and other industry utilised in applied diagnostic software. In addition to description of reasoning rules based on fuzzy logic and taking into account the dynamics of symptoms forming the following topics where raised: decomposition of diagnostic system, dynamical creation of FI threads and multiple fault isolation assuming single fault scenarios.

Diagnostic Reasoning Based on Symptom Forming Sequence

Abstract The paper deals with the problem of correct diagnostic reasoning based on symptoms sequence while the dynamics of the symptoms forming is considered. First, the methods of achieving the knowledge about pattern symptoms erasing sequence are discussed. Then, the two versions of the isolation algorithm based on symptoms forming sequence are described: basic and restricted. Key issues are illustrated with simple examples.

Fault isolability with different forms of the faults-symptoms relation

Abstract 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.

Sequential residual design method for linear systems

The paper presents the fault isolation method for linear systems taking into account the residual dynamics. The method is based on the analysis of the sequence of observed symptoms assuming the residual models. Additionally, the method of designing the secondary residuals that allows to achieve particular symptom sequence for particular fault is proposed. Described methods are illustrated with a simple example.