摘要
针对伊藤型非线性随机系统,对基于模糊双曲正切模型的H∞滤波问题进行了研究.首先通过定义非线性随机系统的模糊双曲规则基,推导出系统的随机模糊双曲正切模型(SFHM);与Takagi-Sugeno(T-S)模糊模型相比,SFHM不需要前提结构的辨识和完备的前提参数空间,尤其是当所需模糊规则数很多时,采用SFHM明显比T-S模型计算负担小;然后基于该模型进行了滤波器的设计,把非线性随机H∞滤波设计中难以求解的二阶汉密尔顿-雅可比不等式问题转化为线性矩阵不等式问题.仿真结果验证了所提出方法的有效性.
Studies the problem of fuzzy H∞ filter for Ito-type nonlinear stochastic systems. With the fuzzy hyperbolic rule base defined, the stochastic fuzzy hyperbolic model （SFHM） is developed. The superiority of SFHM over Takagi-Sugeno（T-S） fuzzy model in practice is mainly that no identification of preconditional structure and complete parameter space are needed, especially SFHM costs obviously less than T-S fuzzy model in computation when a lot of fuzzy rules are needed. Furthermore, based on SFHM, the H∞ filter is designed to transform the second-order nonlinear Hamilton-Jacobi inequality problem which is difficult to solve in the design of nonlinear stochastic H∞ filter into the problem of linear matrix inequality, and the fuzzy hyperbolic H∞ filter is therefore given by solving the linear matrix inequalities instead. Simulation example is provided to illustrate the effectiveness of the proposed approach.
出处
《东北大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2009年第2期158-161,共4页
Journal of Northeastern University(Natural Science)
基金
国家自然科学基金资助项目(60521003
60774048
60728307)
高等学校学科创新引智计划项目(B08015)
教育部长江学者和创新团队发展计划项目(IRT0421)
