Conference Paper

Using an agent-oriented framework for supervision, diagnosis and prognosis applications in advanced automation environments

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

This paper demonstrates how a generic agent-oriented framework can be used in advanced automation environments, for systems analysis in general and supervision, diagnosis and prognosis purposes in particular. The framework’s background and main application areas are briefly described. Next, the function-oriented method Multilevel Flow Modeling (MFM) and its reasoning mechanisms that have proven strength in qualitative planning, modeling and diagnosis activities are introduced. The main enhancements of the framework, as well as an MFM editor based on the framework and towards function-oriented supervision, diagnosis and prognosis purposes are equally explained. Finally, the paper sums up by also addressing plans for further enhancement and in that respect integration with other tailor-made tools for joint treatment of various modeling and analysis activities upon advanced automation environments.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The research described in this and earlier papers [1] [2] began by realizing challenges involved in the application of highly developed supervision and control technologies within the nuclear industry. Building advanced automation environments for current and future nuclear reactors indeed requires an understanding of the risks and benefits of the increased complexity for all automation-related activities. ...
... This led to the development of an MFM-editor based on the framework. The updated version of ShapeShifter along with the initial plans for a prototype of the editor was described in a paper published in the ESREL 2011 proceedings [1] . ...
... The background and development history of the original ShapeShifter framework was described in [6] . The initial development of an updated version of the framework was described in [1] [2] . This chapter summarizes the development and describes recent updates that have been made to the framework. ...
Conference Paper
The paper summarizes the development of the Java-based ShapeShifter framework, previously described in papers published in the proceedings of the ESREL 2010, 2011 and 2012 conferences. The framework is based on the Capability-Oriented Agent Theory, advocating a multi-purpose view of socio-technical systems as made of human, organizational and technical agents and their assets. The paper provides a brief description of the base classes, visual components and essential functionality, but primarily focuses on how the framework is being used in the development of a graphical editor supporting the Multilevel Flow Modeling (MFM) method, to be utilized for Supervision, Diagnosis, and Prognosis (SDP) applications in advanced automation environments. The paper describes the currently implemented functionality of the MFM Editor, including a ShapeShifter-based Process Modeler module, which in combination with the reasoning capabilities of the MFM Workbench developed by the Technical University of Denmark, will facilitate the analysis of process equipment discrepancies, their root causes, and their possible propagations.
... The research described in this and an earlier paper [1] began by realizing challenges involved in the application of highly developed supervision and control technologies within the nuclear industry. Building advanced automation environments for current and future nuclear reactors indeed requires an understanding of the risks and benefits of the increased complexity for all automation-related activities. ...
... This led to the development of an MFM-editor based on the framework. The updated version of ShapeShifter along with the initial plans for a prototype of the editor was described in a paper published in the ESREL 2011 proceedings [1]. ...
... The background and development history of the original ShapeShifter framework was described in [5]. The initial development of an updated version of the framework was described in [1]. This chapter describes the updates that have been made to the framework. ...
Conference Paper
The paper describes the continued development of the Java-based ShapeShifter framework, previously described in papers published in the ESREL 2010 and 2011 proceedings, including updated and new base classes, visual components and essential functionality. The framework is based on the Capability-Oriented Agent Theory, advocating a multi-purpose view of socio-technical systems as made of human, organizational and technical agents and their assets. The paper further reports on how the framework is being used in the development of a graphical editor supporting the Multilevel Flow Modeling (MFM) method, to be utilized for Supervision, Diagnosis, and Prognosis (SDP) applications in advanced automation environments. The recently started development of the ShapeShifter-based Process Modeler, which in combination with the MFM Editor and the reasoning capabilities of the MFM Workbench developed by the Technical University of Denmark will constitute the MFM Suite, is also described.
... Limitations inherent in this version necessitated a complete rewrite of the framework in order to fulfil the requirements of a graphical MFM Editor. The initial development of this updated version was described in [1], while some subsequent updates were described in [5], [6] and [7]. This chapter summarizes the development and describes recent updates that have been made to the framework. ...
... The attributes, basic functionality and usage of the Shape class are described in [1] and [5]. To summarize, the basic Shape class is associated with (among others) the attributes given in Table 1. ...
Technical Report
This report is a continuation of HWR-993, and describes how an agent-oriented framework, ShapeShifter, as a common supporting base for various methodologies and their tools can be used for Supervision, Diagnosis and Prognosis (SDP) applications in advanced automation environments. The framework itself is developed on the basis of a theory assuming that all socio-technical systems are multi-purpose and made of human, organizational and technical agents that together with their assets can fulfill various purposes, depending on their different manners of interrelations and interactions. The report describes the recent additions to and extensions of the ShapeShifter framework, but primarily focuses on the framework’s use as a basis for the continued development of a graphical editor for Multilevel Flow Modelling (MFM), i.e. an MFM Editor and the inclusion of MFM reasoning functionality.
... The MFM Editor contains two essential graphical modelers, the MFM modeler and the process modeler. As both are based on the ShapeShifter framework [3][4], they share a lot of functionality. ...
Conference Paper
This paper presents the status of a software system, the MFM Suite, dedicated to the design and analysis of MFM models related to diagnostic and prognostic analysis of physical processes. New and updated features of the system are described, as well as some examples of its practical use. The paper also briefly describes how the system facilitates the collaboration between control room and field operators via the Android-based MFM Viewer app.
... The MFM Editor contains two essential graphical modelers, the MFM modeler and the process modeler. As both are based on the ShapeShifter framework (Thunem et al, 2011, Thunem, 2012, they share a lot of functionality. ...
Conference Paper
This paper presents the status of a software system, the Multilevel Flow Modeling (MFM) Suite, dedicated to the design and analysis of MFM models related to diagnostic and prognostic analysis of physical processes. New and updated features of the system are described, as well as some examples of its practical use. The paper also briefly describes how the system facilitates the collaboration between control room and field operators via the Android-based MFM Viewer app.
... itis built on the 2nd generation of the Shape Shifter framework (Thunem et al, 2011 ). The background for and initial development of the tool has been described in Thunem (2012Thunem ( , 2013). in addition, the paper describes how the MFM Suite is used to develop an MFM model of the primary side of a PWR. Preliminary results of performing MFM reasoning on a LOCA situation are also included. ...
Conference Paper
This paper reports on the results from the practical application of the ShapeShifter framework on the continued development of a graphical editing suite, the MFM Suite, for MFM and process model design and analysis. The primary use of the MFM Suite is diagnosis and prognosis of anomalies in physical processes. One of the Halden Reactor Project’s advanced NPP simulators based on a PWR is used to demonstrate the applicability of the suite in realistic situations. The paper presents a summary and suggests some plans for future research and development.
... In 2006,a functional HAZOP tool based on MFM [51] was proposed which provided an efficient paradigm for reasoning on potential hazards in safety critical operation for assisting in HAZOP studies. In addition, the MFM Editor [52] was being developed by Institute for Energy Technology to provide a platform for designing an MFM model and was integrated with MFM reasoning engine developed by research groups in Technical University of Denmark to analyze the MFM model. The result can be displayed visually. ...
Thesis
Full-text available
Process safety is of considerable concern for society, in order to reduce the risk for major accidents with severe consequences for human lives and economy. The accidents also demonstrated process complexity as a major challenge, for process safety. Presently process safety is evaluated using qualitative methods which rely upon careful bookkeeping for reevaluation when process modifications and improvements are considered. Consequently it is desirable to develop a more systematic modeling methodology which may be applied for safety assessment and which conveniently may be reused when necessary. A representative qualitative modeling framework is Multilevel Flow Modeling (MFM) which is based on functional modeling. It has been suggested that MFM can deal with the complexity of design and operation of process engineering systems with a promising application future. The purpose of the PhD project is to develop innovative modeling methods for automated analysis and evaluation of safety in industrial processes, especially in oil and gas industry. Validation of functional models is a key issue dealt with in the thesis. The thesis conducts in-depth research on modeling, reasoning, validation and safety analysis applications in the MFM modeling framework. On the basis of an abstraction hierarchy theory, the foundation of MFM theory is introduced and an MFM modeling procedure is proposed for preventing the modeler from making errors. Dynamic simulator of a three phase separation process is established. By following the proposed modeling procedure, an MFM model is built for the first time. The modeling of the assumed thermodynamic gas-liquid phase equilibrium in the separator is discussed as well. The case study demonstrates the applied modelling procedure and also the strength of MFM for modeling of a real technical oil and gas process. Based on the existing research of reasoning rules of MFM, a new reasoning strategy of extended MFM based on roles is proposed. The reasoning strategy is applied for an “untraditional” HAZOP study. The case study shows that the study extends the MFM model expression and reasoning ability. By including roles the discrimination between different types of causes for failure is improved. To deal with the MFM model validation problem, a scientific-based validation method is proposed. With the application of the method for validation of the proposed MFM model for the three phase separation process, the qualitative confidence of the model is assured. To systematically identify cause and evaluate the potential effect of a failure, an integrated qualitative and quantitative modeling framework for HAZOP studies that uses MFM with a knowledge-based reasoning system, together with a risk matrix, and quantitative dynamic simulation for verification and validation risks has been proposed. The integrated framework is successfully applied to a realistic three-phase separation process system. The results demonstrate the importance of the formulation of MFM models to represent the physical system for acquisition of HAZOP knowledge in the qualitative part of the overall methodology. From this point of view, the quantitative analysis based on the dynamic simulation complements and enhances the MFM model based process safety analysis of the system in particular with regard to the transient dynamics of the system. The integrated methodology could be best suitable for FEED (Front End Engineering Design) stage of process development.
... Another advantage of using Jess as programming language except its fast algorithm is that it is fully integrated with Java program and can reason about Java objects (as Jess facts) directly. The rule-based system developed by the authors' research group is now integrated with a MFM model editor, a Java based model building tool developed by Thunem (et. al. 2011) so that the reasoning result can be displayed graphically with the models. ...
Conference Paper
Consequence reasoning is a major element for operation support system to assess the plant situations. The purpose of this paper is to elaborate how Multilevel Flow Models can be used to reason about consequences of disturbances in complex engineering systems. MFM is a modelling methodology for representing process knowledge for complex systems. It represents the system by using means-end and part-whole decompositions, and describes not only the purposes and functions of the system but also the causal relations between them. Thus MFM is a tool for causal reasoning. The paper introduces MFM modelling syntax and gives detailed reasoning formulas for consequence reasoning. The reasoning formulas offers basis for developing rule-based system to perform consequence reasoning based on MFM, which can be used for alarm design, risk monitoring, and supervision and operation support system design.
... 21,22 In this study, a graphical editor based on the Java-based ShapeShifter framework supporting MFM modeling (called the MFM Editor) developed by the Institute for Energy Technology in Norway is used. 23,24 This MFM Editor integrated with the MFM reasoning system developed by DTU can generate cause and consequence trees for a given deviation in a system function. ...
Article
The paper proposes a novel practical framework for computer assisted Hazard and Operability (HAZOP) that integrates qualitative reasoning about system function with quantitative dynamic simulation in order to facilitate detailed specific hazard and operability analysis. The practical framework is demonstrated and validated on a case study concerning a three phase separation process. The multilevel flow modeling (MFM) methodology is used to represent the plant goals and functions. First, means-end analysis is used to identify and formulate the intention of the process design in terms of components, functions, objectives and goals on different abstraction levels. Based on this abstraction, qualitative functional models are constructed for the process. Next MFM specified causal rules are extended with systems specific features to enable proper reasoning. Finally systematic hazard and operability analysis is performed to identify safety critical operations, its causes and consequences. The outcome is a qualitative hazard analysis of selected process deviations from normal operations and their consequences as input to a traditional HAZOP table. The list of unacceptable high risk deviations identified by the qualitative HAZOP analysis is used as input for rigorous analysis and evaluation by the quantitative analysis part of the framework. To this end, dynamic first-principles modeling is used to simulate the system behavior and thereby complement the results of the qualitative analysis part. The practical framework for computer assisted HAZOP studies introduced in this paper allows the HAZOP team to devote more attention to high consequence hazards.
Conference Paper
For complex engineering systems, there is an increasing demand for safety and reliability. Decision support system (DSS) is designed to offer supervision and analysis about operational situations. A proper model representation is required for DSS to understand the process knowledge. Multilevel Flow Modeling (MFM) represents complex system in multiple levels of means-end and part-whole decompositions, which is considered suitable for plant supervision tasks. The aim of this paper is to explore the different possible functionalities by applying MFM to DSS, where both currently available techniques of MFM reasoning and less mature yet relevant MFM concepts are considered. It also offers an architecture design of task organization for MFM software tools by using the concept of agent and technology of multiagent software system.
Conference Paper
Full-text available
Process safety is of considerable concern for society, in order to reduce the risk for major accidents with severe consequences for human lives and economy. The accidents also indicated process complexity as a major challenge for process safety. Presently process safety is evaluated using qualitative methods which rely upon careful bookkeeping for reevaluation when process modifications and improvements are considered. Consequently it is desirable to develop a more systematic modeling methodology which may be applied for safety assessment and which conveniently may be reused when necessary. An established qualitative modeling framework is Multilevel Flow Modeling (MFM) which is based on functional modeling. It has been suggested that MFM can deal with the complexity of design and operation of process engineering systems with a promising application future. Qualitative modeling and reasoning as implemented with MFM can with advantage be combined with quantitative methods in order to automate analysis and evaluation of safety in industrial processes, especially in oil and gas industry with increased coverage of the analysis or for validation purpose. The paper will point out the difference and connections between the qualitative modeling (e.g. functional modeling) and quantitative modeling (e.g. differential and algebraic equations, DAEs) in the process safety context. Then the MFM method will be introduced. A recent HAZOP study of an oil and gas separation plant is summarized. It is shown that validation is a key issue here. It has been investigated how the reasoning results from an MFM model could be validated by comparing it with simulation using a quantitative model. However, due to the complexity of advanced industrial process system, MFM still faces many challenges in industrial process safety application. Finally, the suggested future work within the aspects of supporting for MFM modeling construction, reasoning, validation and counteraction planning are discussed.
Conference Paper
In control design, fault-identification and fault tolerant control, the controlled process is usually perceived as a dynamical process, captured in a mathematical model. The design of a control system for a complex process, however, begins typically long before these mathematical models become relevant and available. To consider the role of control functions in process design, a good qualitative understanding of the process as well as of control functions is required. As the purpose of a control function is closely tied to the process functions, its failure has a direct effects on the process behaviour and its function. This paper presents a formal methodology for the qualitative representation of control functions in relation to their process context. Different types of relevant process and control abstractions are introduced and their application to formal analysis of control failure modes from a process perspective is presented. Finally anticipated applications in context of offline analysis and online supervisory control are discussed.
Conference Paper
This paper summarizes the development to date of the Java-based MFM Editor, a graphical editor supporting the Multilevel Flow Modeling (MFM) method, to be used for Supervision, Diagnosis and Prognosis (SDP) applications in advanced automation environments. The editor builds on the ShapeShifter graphical framework, previously described in the proceedings of the ESREL 2010, 2011 and 2012 conferences. The paper focuses on the editing capabilities and the recently added reasoning functionality (provided by the Technical University of Denmark) required to perform cause and consequence analyses (diagnosis and prognosis) of MFM models. Also, a recently developed process design module is described in some detail.
Conference Paper
Full-text available
Introduction of distributed generation, deregulation and distribution of control has brought new challenges for electric power system operation, control and automation. Traditional power system models used in reasoning tasks such as intelligent control are highly dependent on the task purpose. Thus, a model for intelligent control must represent system features, so that information from measurements can be related to possible system states and to control actions. These general modeling requirements are well understood, but it is, in general, difficult to translate them into a model because of the lack of explicit principles for model construction. Available modeling concepts for intelligent control do not assist the model builder in the selection of model content i.e. in deciding what is relevant to represent for a particular reasoning task and thereby faced with a difficult interpretation problem. In this paper, we present our work on using explicit means-ends model based reasoning about complex control situations which results in maintaining consistent perspectives and selecting appropriate control action for goal driven agents.
Article
Full-text available
A HAZOP methodology is presented where a functional plant model assists in a goal oriented decomposition of the plant purpose into the means of achieving the purpose. This approach leads to nodes with simple functions from which the selection of process and deviation variables follow directly. The functional HAZOP methodology lends itself directly for implementation into a computer aided reasoning tool to perform root cause and consequence analysis. Such a tool can facilitate finding causes and/or consequences far away from the site of the deviation. A functional HAZOP assistant is proposed and investigated in a HAZOP study of an industrial scale Indirect Vapor Recompression Distillation pilot Plant (IVaRDiP) at DTU-Chemical and Biochemical Engineering. The study shows that the functional HAZOP methodology provides a very efficient paradigm for facilitating HAZOP studies and for enabling reasoning to reveal potential hazards in safety critical operations.
Chapter
Automatic, computerized diagnosis can be based on several different search strategies, e.g. a search for a match between a pattern of measured data and some stored symptom patterns, or a search to locate a change in the system’s functional state with reference to a stored model of normal or specified plant state. The latter strategy has a number of basic advantages: it is independent of the prediction and analysis of specific faults and events; the reference for search, the normal state, can be derived from actual plant operation by the computer; the strategy can be based on invariate relations such as conservation laws; etc.
Article
This paper presents a function-oriented system analysis method that gains knowledge about properties of technical systems, including system capabilities that can appear as sources of incipient failures. Such failures have a significant role in connection with process control systems, since they can cause the systems to become overloaded and even unstable, if they remain hidden. The method uses a particular terminology to contribute to the identification of system properties, including goals, functions, and the capabilities. All identified knowledge about the system is then represented by using a tailored combination of two function-oriented methods, Multilevel Flow Modelling (MFM) and Goal Tree–Success Tree (GTST). The features of the method, called Hybrid MFM-GTST, are described and demonstrated by using an example of a process control system. © 1998 John Wiley & Sons, Inc.13: 159–179, 1998
Article
Multilevel flow models (MFM) are graphical models of goals and functions of technical systems. MFM was invented by Morten Lind at the Technical University of Denmark and several new algorithms and implementations have been contributed by the group headed by Jan Eric Larsson at Lund Institute of Technology. MFM has several properties which makes for a relatively easy knowledge engineering task, compared to mathematical models as used in classical control theory and compared to the rule bases used in standard expert systems. In addition, MFM allows for diagnostic algorithms with excellent real-time properties. This paper gives an overview of existing MFM algorithms, and different MFM projects which have been performed, or are currently in progress.
Conference Paper
Multilevel flow models (MFM) are graphical models of goals and functions of technical systems. MFM provides a good basis for computer-based supervision and diagnosis, especially in real-time applications, where fast execution and guaranteed worst-case response times are essential. The expressive power of MFM is similar to that of rule-based expert systems, while the explicit representation of means-end knowledge and the graphical nature of the models make the knowledge engineering effort less and the execution efficiency higher than that of standard expert systems. The paper gives an overview of existing MFM algorithms, and different MFM projects which have been performed or are currently in progress
Article
There seems to be a great potential for using Multilevel Flow Modeling as a framework for reasoning in supervisory control of complex systems. It is a precondition, however, that knowledge about the causal relations between flow functions is represented. Previous attempts have used generic causation rules, specifying possible influences between specific types of flow functions. The problem with this approach is that the generic nature of such rules sometimes leads to invalid reasoning results. This paper presents a method for representing more precisely the actual causal structure of the system being modeled, directly in MFM. The method is based on a set of generic relations, which can be used to make explicit the causal relations hidden in the MFM connection relation. Implications of the method are illustrated by means of simple examples. 1. Introduction In the field of Cognitive Systems Engineering a great deal of research has been concerned with the development of operator support...
Mimir-A Platform for Advanced Condition Monitoring, paper presented at the Enlarged Halden Programme Group meeting
  • Mario Hoffmann
  • Davide Roverso
Hoffmann, Mario, Roverso, Davide (2008), Mimir-A Platform for Advanced Condition Monitoring, paper presented at the Enlarged Halden Programme Group meeting (EHPg 2008), May, Loen, Norway.
Model Based Reasoning using MFM
  • Min Fang
  • Morten Lind
Fang, Min, Lind, Morten (1995), Model Based Reasoning using MFM, Proceedings of Pacific Asian Conference on Expert Systems (PACES), May 16-18, Huangshan China.
Jess in Action Application of Derivation Technique of Possible Counter Actions to an oil Refinery Plant
  • Ernest Friedmann-Hill
Friedmann-Hill, Ernest (2003), Jess in Action, Manning Publishing Co (ISBN 1930110898), USA, http://www. manning.com/friedman-hill/. gofuku, Akio, Tanaka, Yutaka (1999), Application of Derivation Technique of Possible Counter Actions to an oil Refinery Plant, Proceedings of the 4th IJCAI Workshop on Engineering Problems for Qualitative Reasoning, Stockholm Sweden: 77-83.
Application of Derivation Technique of Possible Counter Actions to an oil Refinery Plant
  • Akio Tanaka
gofuku, Akio, Tanaka, Yutaka (1999), Application of Derivation Technique of Possible Counter Actions to an oil Refinery Plant, Proceedings of the 4th IJCAI Workshop on Engineering Problems for Qualitative Reasoning, Stockholm Sweden: 77-83.
Multilevel Flow Modeling of Monju Nuclear Power Plant
  • Morten Lind
  • Yoshikawa
  • Hidekazu
  • Ming Yang
  • Tamayama
  • Kyoichi
Lind, Morten, Yoshikawa, Hidekazu, Yang, Ming, Tamayama, kiyoshi, okusa kyoichi (2011c), Multilevel Flow Modeling of Monju Nuclear Power Plant, Submitted to ICI2011, Daejon korea, August 21-25.