【摘要】 针对多变量受控自回归自回归滑动平均(M-CARARMA)系统，利用滤波辨识理念和递阶辨识原理，研究和提出了滤波递阶广义增广随机梯度辨识方法、滤波递阶多新息广义增广随机梯度辨识方法、滤波递阶广义增广递推梯度辨识方法、滤波递阶多新息广义增广递推梯度辨识方法、滤波递阶递推广义增广最小二乘辨识方法、滤波递阶多新息广义增广最小二乘辨识方法。这些滤波递阶广义增广辨识方法可以推广到其他有色噪声干扰下的线性和非线性多变量随机系统中。更多还原

【Abstract】 For multivariable controlled autoregressive autoregressive moving average(M-CARARMA) models, which are also called multivariable equation-error autoregressive moving average(M-EEARMA) models, this paper investigates and proposes filtered hierarchical generalized extended stochastic gradient identification methods, filtered hierarchical multi-innovation generalized extended stochastic gradient identification methods, filtered hierarchical generalized extended recursive gradient identification methods, filtered hierarchical multi-innovation generalized extended recursive gradient identification methods, filtered hierarchical generalized extended least squares identification methods, and filtered hierarchical multi-innovation generalized extended least squares identification methods from available input-output data by using the filtering identification idea and the hierarchical identification principle. These filtered hierarchical generalized extended identification methods can be extended to other linear and nonlinear multivariable stochastic systems with colored noises.更多还原