模式含有开关时遗传算法求解条件非线性最优扰动的有效性研究

THE EFFECTIVENESS OF GENETIC ALGORITHM IN CAPTURING CONDITIONAL NONLINEAR OPTIMAL PERTURBATION WHEN PARAMETERIZATION "ON-OFF" SWITCHES ARE INCLUDED BY MODEL

在基于条件非线性最优扰动(CNOP)的台风适应性观测研究中,针对预报模式的湿物理参数化产生的"on-off"开关导致传统伴随方法不能为最优化过程提供正确梯度这一现象,将模式含有"on-off"开关时求解CNOP的问题视为非光滑最优化问题,引入遗传算法,在给出详细的算法流程后,以一个在强迫项中含"on-off"开关的理想模式,分析了"on-off"开关对求解CNOP的影响,三个数值试验检验了模式含有"on-off"开关时遗传算法求解CNOP的有效性,并分析了不同初始种群对最优化结果的影响。结果显示,所采用的含有"on-off"开关的理想模式下,遗传算法能有效求解CNOP,最后对遗传算法求解CNOP的优缺点进行了详细讨论。

Among the typhoon adaptive observation based on conditional nonlinear optimal perturbation （CNOP）, considering that ＂on-off＂switches, caused by moist physical parameterization in prediction model, disable conventional adjoint method to provide correct gradient during the optimization process, the capture of CNOP, when ＂on-off＂ switches are included in model, is treated as nonsmooth optimization in this study, and genetic algorithm （GA） is introduced, after detailed algorithm procedures are formulated, with an idealized model with parameterization ＂on-off＂ switches in forcing term, the impacts of ＂on-off＂ switches on the capture of CNOP are analyzed, and three numerical experiments are conducted to check the effectiveness of GA in capturing CNOP and analyze the impacts of different initial population on optimization result. The result shows that GA is competent for the capture of CNOP in the context of the idealized model with parameterization ＂on-off＂ switches in this study. Finally, the advantages and disadvantages of GA in capturing CNOP are analyzed in detail.