ABSTRACT Wiener model structures have attracted considerable attentions due to their powerful modelling abilities to solve industrial problems. This paper considers a new class of Wiener model structure, which consists of a linear transfer function and a radial basis function (RBF) network. Based on this structure, the unstable zero-dynamics, the added disturbances and the approximation errors problems can be dealt with. Since the model parameters are assumed unknown, we develop a new indirect adaptive control scheme. It is a consensus that the certainty equivalence principle is the key that connects the adaptation algorithm with the control law. This paper explains that, for indirect adaptive control of Wiener model, this principle may not be applied directly. The main contribution of this paper is to address this issue by presenting a rigorous theoretical analysis. Representative examples including a simulated pH process control problem are studied to test the control performance. Comparison results indicate that the proposed controller has much wider applicability than some alternative methods, especially for its improved performance and smoother adaptation.

中文翻译：

---

Wiener非线性系统的鲁棒间接自适应滑动模型控制

摘要 Wiener 模型结构因其强大的建模能力来解决工业问题而引起了广泛的关注。本文考虑了一类新的 Wiener 模型结构，它由线性传递函数和径向基函数 (RBF) 网络组成。基于这种结构，可以处理不稳定的零动态、附加扰动和逼近误差问题。由于假设模型参数未知，我们开发了一种新的间接自适应控制方案。确定性等价原理是连接自适应算法与控制律的关键，这是一个共识。本文说明，对于维纳模型的间接自适应控制，该原理可能不能直接应用。本文的主要贡献是通过提出严格的理论分析来解决这个问题。研究了包括模拟 pH 过程控制问题在内的代表性示例，以测试控制性能。比较结果表明，所提出的控制器比一些替代方法具有更广泛的适用性，尤其是其改进的性能和更平滑的适应。