Genetic algorithm has a slow convergence rate and the accuracy is not high. It is easy to premature convergence. Propose a method which can adjust the probability value dynamically. Introduce climbing method, excellent performance in local, into iterative process of genetic algorithm and propose adaptive genetic algorithm based on climbing, which could avoid algorithm falling into local optimal solution and improve the evolutionary efficiency and precision of genetic algorithm. Compare the performance among Simple Genetic Algorithm, Adaptive Genetic Algorithm and Adaptive Genetic Algorithm based on Climbing through simulation experiments to prove that the improved Genetic Algorithm is more excellent in speed and accuracy.Adaptive Genetic Algorithm based on Climbing was utilized in Projection Pursuit model, which was applied to the instance for comprehensive evaluation. The result indicates that the degree of water nutrition in Hengshui Lake is ?, belonging to moderate nutrition degree and providing basis for Water pollution control of Hengshui Lake for governance. |