基于凸空间收缩滤波的噪声不确定时滞系统状态估计

Convex space contraction filtering based state estimation for noise uncertainty time-delay systems

针对受未知但有界噪声干扰的噪声不确定时滞系统,提出了一种基于凸空间收缩滤波的系统状态估计方法.首先,利用凸空间定义包裹系统真实状态的可行集,求解下一时刻的凸空间体形状矩阵;随后从凸空间收缩角度,利用当前时刻噪声和扰动构造带空间,得到满足状态预测和量测更新条件的凸空间结构;进而,依据时滞系统约束条件构造线性规划不等式方程组,利用线性规划求解该凸空间,得到包裹状态可行集的最紧致凸空间体;最后,通过数值仿真与电池化成工艺变换器案例仿真,验证了本文所提方法解决不确定时滞系统状态估计问题的有效性和准确性.

For noise uncertainty time-delay systems,a convex space contraction filtering based state estimation algorithm is studied.First,a convex space is denoted to contain the feasible set of real state of the noise uncertainty time-delay systems,before obtaining the convex space shape matrix in next moment.Then,from the perspective of convex space contraction,the noise term and the disturbance at the present time are used to construct the strip space and to get a convex space structure that satisfies the conditions of state prediction and measurement update.Furthermore,linear programming inequalities are constructed based on the proposed constraints,the convex space is then solved by linear programming and the most compact convex space of the feasible set for wrapping the states is obtained.Finally,a numerical simulation and a case simulation of the battery forming process converter show the effectiveness and accuracy of the proposed algorithm on solving the state estimation problem for uncertain time-delay systems.