基于嵌套粒子群算法的多涡环微下击暴流模型参数选择方法

Parameters Selection Method of Multiple Vortex-Ring Microburst Model Based on Nested Particle Swarm Optimization

提出一种使用嵌套粒子群算法，根据微下击暴流水平垂直风速最大峰值比选择多涡环微下击暴流模型参数的方法．设计了包括目标群和中间群的嵌套式粒子群结构，使用中间群寻找水平垂直最大风速，并使用目标群最终选择模型参数．用该方法计算了水平垂直风速峰值比在0．3—0．7之间的多组多涡环微下击暴流模型参数，最终生成的微下击暴流场的水平垂直风速峰值比与预设值误差的数量级在10^-4以内．

A parameter selecting method of Multiple Vortex Ring Microburst Model（MVRMM） based on NPSO （Nested Particle Swarm Optimization） is propesed, which can select the parameters of MVRMM by any proportion between maximal hori- zontal velocity and maximal vertical velocity. The nested particle swarm structure is designed which contained parent swarm and subswarms. Find the maximal horizontal and vertical velocity with subswarms, and find the optimal parameters with parent swarm. Calculate several parameters of MVRMM by this method, which the proportion between maximal horizontal velocity and maximal vertical velocity is between 0.3 to 0.7.And the error between the factual proportion and the setting proportion is within 10^-4.