12,000

We have over 12,000 students, from over 100 countries, within one of the safest campuses in the UK

93%

93% of Lancaster students go into work or further study within six months of graduating

Home > Research > Publications & Outputs > Optimal design synthesis of component-based sys...
View graph of relations

« Back

Optimal design synthesis of component-based systems using intelligent techniques

Research output: Contribution in Book/Report/ProceedingsChapter

Published

Publication date2004
Host publicationDo Smart Adaptive Systems Exist?
EditorsB Gabrys, K Leviska, J Strackelijan
Place of publicationBerlin/Heidelberg
PublisherSpringer
Pages267-284
Number of pages18
Volume173
Edition173
ISBN (Print)3-540-24077-2
Original languageEnglish

Publication series

NameStudies in Fuzziness and Soft Computing
PublisherSpringer

Abstract

In this chapter we have considered a special case of application of intelligent techniques, namely evolutionary algorithms and fuzzy optimization for au- tomatic synthesis of the design of a component-based system. We have also considered the problem of adaptation of the solutions found and the ways it can be done ’smartly’, maximizing the reuse of the previous knowledge and search history. Some useful hints and tips for practical use of these techniques in their application to different problems has been made. The approach we have presented for automatic synthesis of optimal design of component-based systems is novel and original. It can be applied to similar problems in automatic design of complex engineering systems, like petrochem- ical pipelines, electronic circuits design, etc. The complex problem of design synthesis, which involves human decisions at its early stages, is treated as a soft constraints satisfaction problem. EA and problem-specific operators are used to numerically solve this fuzzy optimization problem. Their use is mo- tivated by the complex nature of the problem, which results in the use of different types of variables (real and integer) that represent both the physical and the topological properties of the system. The ultimate objective of this fuzzy optimization problem is to design a feasible and efficient system. The important problem of the design adaptation to the changed speci- fications (constraints) has also been considered in detail. This adds to the flexibility of the approach and its portability. The approach which is pre- sented in this chapter has been tested with real systems and is realized in Java. It fully automates the design process, although interactive supervision by a human-designer is possible using a specialized GUI. (c) Springer