This paper deals with the compensation of nonlinearities in dynamical systems using nonlinear polynomial autoregressive models with exogenous inputs (NARX). The compensation approach is formulated for static and dynamical contexts, as well as its adaptation to hysteretic systems. In all of these scenarios, identified NARX models are used. The core idea is to rewrite the model as an algebraic polynomial whose roots are potential compensation inputs. A procedure is put forward to choose the most adequate root, in cases where more than one is possible. Both numerical and experimental results are presented to illustrate the method. In the experimental case the method is compared to other approaches. The results show that the proposed methodology can provide compensation input signals that practically linearize the studied systems using simple and representative models with no more than five terms.

中文翻译：

---

基于确定的NARX多项式模型的非线性补偿

本文利用带有外源输入的非线性多项式自回归模型（NARX）来处理动力系统中的非线性。补偿方法是针对静态和动态环境以及其对滞后系统的适应性而制定的。在所有这些情况下，都使用已识别的NARX模型。核心思想是将模型重写为一个以潜在补偿输入为根的代数多项式。如果可能的根数超过一个，则提出一种选择最合适的根的程序。数值和实验结果都可以说明该方法。在实验情况下，将该方法与其他方法进行了比较。