Determining optimal configuration for turbine generator cooler

Configuration problems with hierarchical decision-tree structures are difficult to encode for solution using simple genetic algorithms. The chromosomes typically require fitness-evaluating schemes with steep gradients to optima. Solutions get stuck at local optima. We used a GA that can control the expression of over-specified chromosomes for exploring the multilevel search-space. Experiments to configure a turbine generator cooler are performed and results are reported.

[1]  Christopher M. Foley,et al.  Evolutionary algorithm for structural optimization , 1999 .

[2]  D. Dasgupta,et al.  A MORE BIOLOGICALLY MOTIVATED GENETIC ALGORITHM: THE MODEL AND SOME RESULTS , 1994 .

[3]  Dipankar Dasgupta,et al.  Engineering Optimizations Using the Structured Genetic Aigorithm , 1992, ECAI.

[4]  D. R. McGregor,et al.  Designing application-specific neural networks using the structured genetic algorithm , 1992, [Proceedings] COGANN-92: International Workshop on Combinations of Genetic Algorithms and Neural Networks.

[5]  Dipankar Dasgupta,et al.  Handling deceptive problems using a different genetic search , 1994, Proceedings of the First IEEE Conference on Evolutionary Computation. IEEE World Congress on Computational Intelligence.