一类改进的Logistic模型参数估计方法及其应用

An Improved Parameter Estimation Method of Logistic Model and Its Application

根据实测数据估计Logistic模型参数时,对已有的数据不满足直接利用三点法、四点法应用条件的问题,提出一类改进的三点法、四点法,即迭代逼近算法.以底部耗氧型结冰湖的溶解氧浓度分布为例,建立冰盖下溶解氧浓度垂直分布的Logistic模型,采用迭代逼近算法估计该模型的参数值.结果表明:改进的三点法、四点法的判定系数都较高,均可用于Logistic模型的参数估计,但改进的四点法整体优于改进的三点法.算法进一步完善了Logistic模型的参数估计方法.

The improved three-point and four-point methods,named as the iterative approximation algorithm,were presented to estimate the parameters of Logistic model according to the measured data in this paper.And they could be used to the data which was not directly satisfied with conditions of the both methods.Taking the dissolved oxygen distribution of an ice-covered lake as an example,the Logistic model of spatial distribution of dissolved oxygen concentration under ice was established and the parameters of the model were estimated with the algorithm.The results show that the both improved methods have high correlation coefficients and can be used to estimate the parameters of the Logistic model.The improved four-point method is better than the improved three-point one.The iterative approximation algorithm improves the parameter estimation method of Logistic model.