提出一种对扩展的非线性状态空间模型进行识别的自组织序贯蒙特卡洛滤波方法.通过在原始状态空间模型中增加未知待识别参数来定义扩展的状态空间模型,该模型为自组织非线性状态空间模型,适用于解决噪声分布的自校正问题;同时该系统识别方法通过引入局部似然函数,可以从有限的有效数据中进行最优参数选取.给出了Bouc-Wen滞同系统识别的数值算例,证明了该方法的有效性.%A self-organizing sequential Monte Carlo filtering method for a nonlinear state-space model was pro- posed. An expanded state-space model was defined by augmenting its state vector with the unknown parameters of the orig- inal state-space model. The self-organizing state-space model could also be applied to self-tuning of noise dispersions. A local likelihood was introduced and used to select the optimal parameter from a finite number of possible values. Examples of hysteretic system identification were shown to verify the effectiveness of the proposed method.
展开▼
