混沌高效遗传算法在马斯京根模型参数优选中的应用

Chaos high efficient genetic algorithm for parameter optimization of Muskingum routing model

本文利用混沌映射的遍历性和实编码遗传算法的全局优化性,通过在遗传进化过程中加入混沌变异操作,在变量的定义域内投放混沌初始群体,用优秀个体群来逐步缩小搜索空间,提出了求解马斯京根模型参数优选问题的一种新方法——混沌高效遗传算法(CHEGA)。应用该方法对5个经典非线性测试函数进行了仿真,在收敛速度和全局优化方面好于现有的简单遗传算法和改进的遗传算法。并将CHEGA用于求解实际马斯京根水文模型参数优选问题。与实编码加速遗传算法、传统非线性规划方法等相比,CHEGA可以遍历到整个区域的内部和边界,较好的保持了种群的多样性,精度高、收敛速度快,对求解实际水文模型参数优选问题非常有效。

In this paper, the chaos high efficient genetic algorithm（CHEGA） is proposed for parameter optimization of Muskingum routing model, in which the initial population are generated by chaos algorithm, and the new chaos mutation operation is used for the shrinking of searching range. CHEGA gradually directs to an optimal result with the excellent individuals obtained by real-value genetic algorithm. It is Very efficient in maintaining the population diversity during the evolution process of genetic algorithm. Its efficiency is verified by application of five test functions compared with standard binary-encoded genetic algorithm and improved genetic algorithm. Compared with real-valued accelerating genetic algorithm and traditional optimization methods, CHEGA can get to the whole searching range, it has rapider convergent speed and higher calculation precision. It is efficient for the global optimization in the practical hydrological models.