利用改进SA算法估计河流水质参数的仿真实验  

Simulation Experiment on Application of Improvement Simulated Annealing Method to Estimating Water Quality Parameters of River Stream

作　　者：郭建青[1] 李彦[2] 王洪胜[1] 马健[2] 

机构地区：[1]长安大学环境科学与工程学院,西安710054 [2]中国科学院新疆生态与地理研究所,新疆乌鲁木齐830011

出　　处：《系统仿真学报》2003年第12期1750-1752,1762,共4页Journal of System Simulation

基　　金：中国科学院知识创新工程重要方向项目(KZCX3-SW-326)

摘　　要：将改进模拟退火法应用于求解分析河流水团示踪试验数据,确定河流水质参数的函数优化问题。针对标准SA算法收敛速度缓慢的弱点,采取了增加附加约束条件、设置内阈值提前降温和增加记忆功能等措施对算法进行了改进。仿真实验结果表明:1) 改进措施能够明显地提高算法收敛速度,并可得到满意的参数计算结果;2) 内循环次数不会对外循环次数产生明显的影响;3) 内阈值的设置对算法的收敛速度影响非常明显,当其值与外阈值接近或相等时,算法的收敛性最佳;4) 在研究的具体问题情况下,降温指数不宜过大,其在0.4—0.65之间取值为宜。The function optimization problem for analyzing the data of river stream water quality test to estimate such water quality parameters as longitudinal dispersion coefficient, average stream velocity and the other parameter is to be solved with simulated annealing method. Such measures as adding supplement constraint on state variables, setting up inner valve value and adding memory function in inner loops are taken to improve convergence of algorithm. The results of simulation show: 1) that these measures are very effective and the improved algorithm may give better estimation results of parameters, 2) that the times of inner loop bring little effect on that of outer loops, 3) that adding inner valve value may improve the convergence significantly, and the convergence is better if the inner valve value equals to outer valve value, 4) that the cooling rate may also affect the convergence, and smaller value of it may speed up computation.

关 键 词：模拟退化法 算法改进 河流水质参数 参数优化估计 算法控制参数 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]