This paper considers the identification of a Volterra system and its applications in structural identification of nonlinear block-oriented models. Any order of the Volterra output is estimated separately via multilevel excitations and optimizations. Then, each order of the Volterra kernels is estimated independently with improved accuracy. Finally, relationships between the first and the second order Volterra kernel functions of block-oriented models are exploited to determine the structures of nonlinear block-oriented systems. The simulation studies verify the effectiveness of the proposed Volterra series identification method and the structure identification method for nonlinear block-oriented systems.

中文翻译：

---

Volterra级数辨识及其在非线性块导向系统结构辨识中的应用

本文考虑了 Volterra 系统的识别及其在非线性面向块模型的结构识别中的应用。Volterra 输出的任何阶数都是通过多级激励和优化单独估计的。然后，Volterra 内核的每个阶都以更高的精度独立估计。最后，利用面向块模型的一阶和二阶 Volterra 核函数之间的关系来确定非线性面向块系统的结构。仿真研究验证了所提出的 Volterra 级数识别方法和非线性块导向系统的结构识别方法的有效性。