Agenetic-neural network algorithmforoptimumdesign is developed. In the algorithm, the global
property of genetic algorithms (AG) and the parallelism of artificial neural networks (AN2) are combined;
GA provides global initial solutions, fromwhich AN2obtains the final solutions. Thus, the defects of slow
convergence with GA and easily falling into local solutions with AN2can be overcome. An applied example
shows that the global property of the new algorithm is better than that ofAN2, and its convergence is better
than that of GA.