定性信号在混合状态与参数估计中的应用  

Estimates of hybrid states and parameters using qualitative signal data

作　　者：莫以为[1] 萧德云[1] 

机构地区：[1]清华大学自动化系,北京100084

出　　处：《清华大学学报（自然科学版）》2004年第10期1356-1359,共4页Journal of Tsinghua University(Science and Technology)

基　　金：国家"八六三"高技术项目(2002AA412510;2002AA412420)

摘　　要：为了利用实际工程中常可获得的各种测量信号,该文分析定性信号的特点,在某些时刻提取出定量信息,给出将所提取出的定量信息与连续测量信号共同用于混合系统状态估计的方法。同时,针对用于状态估计的模型难以避免误差的情形,在定性信号提取出的定量信息帮助状态估计的前提下,提出一种基于进化粒子滤波的方法,估计模型中存在误差的参数。仿真结果表明,该文提出的混合状态与参数估计方法是有效的,可以充分利用所获得的定性信号。An algorithm was developed to analyze qualitative measurement signals obtained in engineering test for hybrid state estimation to extract exact quantitative information at specified instants in time. A hybrid states estimation approach was then developed using both this extracted quantitative information and the continuous measurement signals. Previous model for state estimates have rarely been accurate. The state estimates were improved with the quantitative information acquired from qualitative signals usingan evolutionary particle filter to estimate imprecise model parameters. Simulation results demonstrate the feasibility of the proposed approaches for hybrid state and parameter estimates with both approaches taking full advantage of the qualitative information.

关 键 词：状态估计 参数估计 混合系统 粒子滤波 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]