【摘要】 压电陶瓷执行器物理结构复杂,通过参数化方法辨识经典Preisach模型(CPM)描述其迟滞特性时,难以找到合适的Preisach函数,模型预测误差较大。为了提高建模精度,定义了均值迟滞模型作为CPM的补充,通过将均值迟滞模型与CPM加权叠加,得到混合Preisach模型(MPM),并将权值定义为迟滞度参数,用以描述模型的迟滞非线性强烈程度。证明了MPM具有类似于CPM的擦除特性和一致特性,给出了MPM的表示定理。最后结合神经网络方法完成了MPM的辨识过程。实验结果表明,MPM及其预测误差最大值较CPM降低了1.20μm,同时MPM的两种特性也得到了验证。更多还原

【Abstract】 It is difficult to find a suitable Preisach function in the Classical Preisach Model’s(CPM) identification by the parametric strategy because the piezoceramics actuator has complex construction.To improve the modeling accuracy,this paper presented an average hysteresis model as compensation of the CPM.The Mixed Preisach Model(MPM) was obtained by the weight sum of the CPM and the average hysteresis model,and the weight parameter defined as hysteresis degree was used to describe the degree of the hysteresis nonlinearity.As the CPM was,the MPM had the wiping-out and congruency properties,which gave the representation theorem.The neural network was applied to accomplish the identification of the MPM.Experimental results illustrated that the max predicted error of MPM reduces 1.20 μm compared with the CPM and the two properties of the new model have been proven at the same time.更多还原