基于并行交叉遗传粒子群算法的水文频率参数估计  

Hydrologic Frequency Parameter Estimation Based on Parallel and Crossed GA-PSO Algorithm

作　　者：佘纬[1] 夏永波[1] She Wei;Xia Yongbo(College of Mathematics and Statistics,South-Central University tbr Nationalities,Wuhan 430074,Chin)

机构地区：[1]中南民族大学数学与统计学学院,武汉430074

出　　处：《中南民族大学学报（自然科学版）》2018年第2期147-150,共4页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：国家自然科学基金资助项目(61603419;61771021)

摘　　要：针对水文频率参数估计问题,提出了基于并行交叉遗传粒子群算法的水文频率优化适线方法.该方法从初始种群的产生和编码、算法的执行方式和数据融合,以及其中的PSO算法的惯性权重三个方面对传统算法进行了改进.为了验证该算法的性能,分别采用矩法、权函数法、概率权重矩法、线性矩法、GA、PSO和文中所提出的算法,对某水文站的年径流量进行了研究分析,得到了各个方法对应的水文频率曲线,实验结果表明:文中提出的并行交叉遗传粒子群算法较其它6种方法,可以得到更小的离差平方和,该算法得到的水文频率曲线可以很好地拟合实测数据.For the problem of the Hydrologic frequency estimation,a hydrologic frequency fitting-curve method based on Parallel and Crossed GA-PSO Algorithm was proposed. In this paper,we devoted to make three improvements： generation and coding of initial population,run mode and data fusion of the algorithm,inertia weight of the PSO algorithm. In order to verify the performance of the proposed algorithm,using moment method,weight function method,probability weighted moment method,linear moment method,GA,PSO and the algorithm proposed in this paper,the annual runoff data of a hydrologic station were analyzed and studied,and we obtained the hydrologic frequency curves according to seven methods. The experiment result shows that the proposed algorithm can get smaller sum of the squares of the vertical deviation than the other six methods. The hydrologic frequency curve obtained by this algorithm can fit the measured data well.

关 键 词：水文频率 参数估计 初始种群 遗传粒子群算法 

分 类 号：TP274.5[自动化与计算机技术—检测技术与自动化装置]