Transcript presentation in ppt

Supporting Conceptual Design
Innovation through Interactive
Evolutionary Systems
I.C. Parmee
Advanced Computation in Design and Decision-making
CEMS, University of the West of England Bristol, BS16 4QY
Email:
[email protected]
A far more detailed description of aspects of this work can be found in the following references:
Parmee I. C., 2002, Improving Problem Definition through Interactive Evolutionary Computation. Journal of Artificial
Intelligence in Engineering Design, Analysis and Manufacture,16 (3), Cambridge Press.
Parmee I. C., Cvetkovic C., Watson A. H., Bonham C. R., 2000, Multi-objective Satisfaction within an Interactive
Evolutionary Design Environment. Journal of Evolutionary Computation. 8 (2), MIT Press, pp 197 – 222.
Cvetkovic D., Parmee I. C., 2001, Preferences and their Application in Evolutionary Multiobjective Optimisation. IEEE
Transactions on Evolutionary Computation, 6(1), pp 42 - 57.
Cvetkovic D., Parmee I. C., Agent-based Support Within an Interactive Evolutionary Design System. Artificial Intelligence
for Engineering Design, Analysis and Manufacturing Journal; Cambridge University Press, Vol.16 No.5, (Nov 2002 - in
press).
Parmee I. C. Evolutionary and Adaptive Computing in Engineering Design. Springer Verlag, London, (2001).
Interactive evolutionary design strategies support:
• the extraction of optimal design information;
• its presentation to the designer;
• subsequent human-based modification of the problem domain based
upon knowledge gained from the information received.
Iterative designer / evolutionary search processes result in:
• a better understanding of the problem;
• improved machine-based representation of the design domain;
And through continuous user interaction strongly support new
concept formulation, innovation, discovery and creativity.
Prototype Interactive Evolutionary Design System (IEDS)
On-line Database
Rule-Based
Preferences
Scenario
(A)
Evolution
Information
gathering
processes
Machine-Based
Agents
Scenario
(B)
Evolution
Scenario
(C)
Evolution
External Agents
(Design Team)
Information Gathering via Cluster-oriented
Genetic Algorithms
How?
• Highly explorative GA / GAs
• Solutions extracted and passed through Adaptive
Filter
• Better solutions pass into Final Clustering Set defines HP regions
Application of COGA to Preliminary Airframe Design
1
2
3
4
Figures 1 to 4 show the effect of increasing the filter threshold setting. Low settings of
figure 1 result in a large cluster of medium fitness solutions increasing the filter setting
results in the identification of the two disjoint clusters of figure 4.
High-performance regions relating to various objectives
a
b
c
Lines define boundaries of the high performance regions for each
objective - shaded area defines common region containing HP solutions
that satisfy more than one objective. (a) Common region containing high
performance solutions for Ferry Range and Turn Rate identified but
Specific Excess Power(SEP)cannot be satisfied.; (b)Relaxing filter
threshold for SEP allows lower fitness SEP solutions through, boundary
moves towards feasible region; (c) Further relaxation results in the
identification of a feasible region for all objectives.
Objective Preferences
• Simple linguistic rules facilitate direct preference
manipulation by the designer e.g:
relation

<
<<
>
>>
intended meaning
is equally important
is less important
is much less important
is more important
is much more important
• Ranked preferences relating to multi-objectives can be
introduced and altered during an evolutionary run.
• Designer only required to answer a minimal set of
straightforward questions
• Preferences transformed into numerical objective weightings
Co-evolutionary Multi-objective Satisfaction
•Concurrent GA processes each optimise one objective
• Fitness measure for individuals within each GA is adjusted
by comparing distance between solutions of one objective with
those of others
• Penalty relating to the degree of diversity of variables of
each objective process is imposed
• Initial convergence upon individual objectives leads to
overall convergence of all processes upon a single
compromise design region.
(a) Ferry Range is much more important
(b) All objectives are of equal importance
(c) Ferry Range is much less important
Agents for Scenario / Dynamical Constraint
Satisfaction
• Designer likely to have several ideal scenarios such as: ‘I would like
objective A to be greater than 0.6 and objective C to be less than 83.56;
objectives B, D, E should be maximised; variable 2 should have a value
of between 128.0 and 164.5; a value graeter than o.32 is prefered for
variable 7 ’
• Incremental Agent operates as follows:
1 Use designer’s original preferences for both objectives and scenarios
and run optimisation process
2 If some scenarios are not fulfilled, agent suggests increase in their
importance of these scenarios
3 If some scenarios still not fulfilled even when classed as ‘most
important’ agent suggests change to variable ranges in scenario.
4 If some scenarios still not fulfilled agent reports to designer and asks
for assistance.
Closed preference / scenario loop