【摘要】 针对传统的求解Jensen模型敏感指数的回归分析法(LR)存在的有偏估计和拟合精度不高等问题,利用粒子群算法(PSO)和单纯形法—粒子群算法(SM-PSO)分别对模型的敏感指数进行求解并与传统方法进行对比。结果表明,回归分析法、PSO算法和SM-PSO算法所得模型计算的相对产量与实际相对产量的平均相对误差分别为3.1%、1.8%和1.4%,说明PSO算法和SM-PSO算法均优于传统算法,尤其是SM-PSO算法收敛速度更快、拟合精度更高,是一种有效的求解Jensen模型敏感指数的方法。更多还原

【Abstract】 Aiming at the problems of partial estimation and low fitting precision for traditional regression analysis algorithm,particle swarm optimization algorithm(PSO) and simplex-particle swarm optimization algorithm(SM-PSO) are applied to solve sensitive index of Jensen model,separately.The results show that the mean relative error between actual relative yield and calculating relative yield with the models of traditional regression analysis algorithm,PSO algorithm and SM-PSO algorithm is 3.1%,1.8% and 1.4% respectively,which indicates that PSO algorithm and SM-PSO algorithm are both superior to the traditional algorithm;especially,SM-PSO algorithm has quickly convergence and higher fitting precision,which can be regard as an effective method for calculating sensitive index of Jensen model.更多还原