非线性时变系统时滞和参数在线联合估计的SMSA方法

针对输入带纯迟延的一类非线性系统，提出时变时滞和参数在线联合估计的一种SMSA混合算法。该方法有机结合具有概率突跳特性的模拟退火和基于高维欧氏空间中凸多面体结构的单纯形搜索法，在最小化平均平方误差意义下动态寻找时滞和参数的最优估计值。基于实际非线性水箱系统的仿真验证了混合算法对非线性系统时滞和参数在线联合估计的有效性及其一定的抗噪声能力。

SMSA Application for Online Joint Estimation of Time Delay and parameters in Nonlinear Time-varying Systems

This paper presents a kind of SMSA hybrid algorithm for the online joint estimation of time delay and parameters in a class of nonlinear time varying systems with pure time delay in input signal.Combining the simulated annealing with probabilistic jumping property and the Nelder Mead simplex method based on the convex polyhedron in higher dimension Euclid space,the approach dynamically searches the optimal time delay and parameters in the sense of minimizing the mean square error.Simulation results based on a nonlinear water tank demonstrate the effectiveness and the robustness against nosie of the hybrid algorithm for online joint estimaltion of time delay and parameters in nonlinear systems.