SDH模型与神经网络串联的谐波减速器混合迟滞建模研究

Research of Hybrid Hysteresis Modeling of Harmonic Reducer based on SDH Model and Neural Network in Series

针对柔性环节含有的谐波减速器所表现出的特殊非线性迟滞特性,构建了由SDH模型与神经网络串联的谐波减速器的混合迟滞模型。以输入与输出信号之间具有与谐波减速器迟滞曲线相似迟滞特性的SDH模型为前置模型,以补偿前置模型在描述迟滞特性时存在的误差的非线性动态RBF神经网络作为后置模型,构成了混合迟滞模型,描述谐波减速器迟滞非线性特性。根据所搭建的实验平台,对不同频率输入信号、不同负载状态下获得的数据进行建模,与经典RBF神经网络模型和SDH模型相对比,实验表明,所构造的混合迟滞模型精度高、适应性强。

A hybrid hysteresis model consisting of SDH model and neural network in series is constructed for the special nonlinear hysteresis characteristics of the harmonic reducer in the flexible link. The SDH model is used as the pre-model of which hysteresis characteristics between the input and output signals is similar to the hysteresis characteristics of the harmonic reducer. The dynamic RBF neural network which can describe the nonlinearity is used as the post-model to make the hybrid hysteresis model to describe the hysteresis nonlinearity characteristic of the harmonic reducer. The data obtained under different frequency input signals and different load are modeled. Compared with the classical RBF neural network model and SDH model, the experimental results show that the constructed hybrid hysteresis model has high precision and strong adaptability.