Work Area: Basic Aspects of Multiple Computing Agents
Keywords large-scale systems, distributed/decentralised measurement and control structures, hierarchical coordination, modelling methodologies, analysis, evaluation criteria
Start Date: to be announced / Status: starting
[ participants / contact ]
Abstract SESDIP aims at developing modelling and analysis tools for the evaluation of alternative distributed measurement and control structures for industrial processes. A dominant characteristic is that the approach is based on structural system methodologies. The results will prove the advantages of the structural framework and tools in the analysis of control and measurement schemes of industrial processes and will specify the future design needs.
The design of large-scale industrial processes involves distributed measurement, actuation and control schemes and is frequently characterised by hierarchical information processing. SESDIP addresses the problems of evaluation of alternative distributed measurement and control schemes using system structure methodologies. As such the project aims at : (i) Evaluating the alternative methodologies, (ii) Unifying and expanding the relevant modelling tools, (iii) Developing certain system theoretic structural approaches for the analyis of control and measurement schemes, (iv) Producing CAD tools to assist the (ii)-(iii) activities and (v) Specifying the future needs for an integrated analysis and design framework of large scale distributed processes.
The underlying philosophy of the project is based on the following: (i) The potential of achieving good system design depends on its inherent structure and (ii) The selection of systems of measurement and actuation variables affects the overall system structure. The approach is characterised by the exploration of structural characteristics of composite models. System structure can be seen as the interaction of (i) Subsystem model parameters and characteristics, (ii) Interconnection graph and (iii) Selection of local measurement and actuation schemes.
The consortium brings together complementary skills from the different areas of structural system approaches, Hierarchical Control, Measurement, Industrial applications and CAD. Existing modelling and analysis tools will be evaluated in terms of their relevance to the analysis and design of control and measurement schemes. Algebra-geometric, implicit, mixed system and graph approaches will be developed with emphasis on deriving systematic analysis tools. The potential of discrete event and neural network tools for overall system operation modelling will also be examined. It is envisaged that CAD tools will be developed to assist the modelling and analysis aspects and a framework for simultaneous measurement and control design will be specified.
The conceptual framework, tools and techniques which will emerge from these fundamental studies will provide engineers a new set of analytic, manipulative and interpretative tools based on structure indicators and overall performance characteristics, for the assessment of alternative control and measurement schemes for distributed large scale processes. The spin-offs of the work include the specification of systematic design procedures for control and, measurement structure selection.
City University - UK
Control Engineering Centre
UK- London EC1V OHB
Technical University of Denmark - DK
Ecole Centrale de Nantes - F
University of Patras - GR
Intrasoft S.A - GR
Prof. N. Karcanias
tel +44/71 477 8125
fax +44/71 477 8568
SESDIP - 8924, August 1994
please address enquiries to the ESPRIT Information Desk
html version of synopsis by Nick Cook