Publication

Abstract : A new dynamic genetic algorithm (DGA) was developed using a modified form of the search space reduction technique (SSRT). The new algorithm was applied to optimise rule curves for the Ubonratana multi-purpose reservoir in Thailand. Comparison of the new DGA and a standard GA showed that the new algorithm produced the optimal solution at half of the computation time required by the standard algorithm. Besides, the value of the optimum fitness function, i.e. the sum of the squared water deficits, was a mere 6,021 (Mm3 ) 2 , compared with 10,271 (Mm3 ) 2 obtained with the standard algorithm. Finally, performance of the reservoir in terms of reliability, vulnerability, resilience and sustainability with the DGA derived rule curves was far superior to that of the standard algorithm.