论文部分内容阅读
为克服最小二乘法或归一化最小二乘法在二阶Volterra建模时参数选择不当引起的问题,在最小二乘法基础上,应用一种基于后验误差假设的可变收敛因子技术,构建了一种基于Davidon-Fletcher-Powell算法的二阶Volterra模型(DFPSOVF).给出参数估计中自相关逆矩阵估计的递归更新公式,并对其正定性、有界性和τ(n)的作用进行了研究.将DFPSOVF模型应用于Rssler混沌序列的单步预测,仿真结果表明其能够保证算法的稳定性和收敛性,不存在最小二乘法和归一化最小二乘法的发散问题.
In order to overcome the problems caused by improper parameter selection in second-order Volterra modeling by least square method or normalized least squares method, based on the least square method, a variable convergence factor technique based on posterior error assumption A Second Order Volterra Model Based on Davidon-Fletcher-Powell Algorithm (DFPSOVF). The recursive updating formula of the inverse auto-correlation matrix estimation in the parameter estimation is given. And its positive definiteness, boundedness and τ (n) The DFPSOVF model is applied to the single step prediction of Rssler chaotic sequence, and the simulation results show that it can guarantee the stability and convergence of the algorithm, and there is no problem of the least squares method and the normalized least squares method.