提出一种线性电路参数化宏模型自动化建模方法。传统的参数化宏建模采用重心插值实现,将非参数化极点-留数宏模型拓展成参数化宏模型;但是需要人工凭经验确定插值节点,效率低下,建模代价大。针对这一问题,通过引入插值步长迭代平分缩小的策略,在满足一定建模精度的情况下,实现自动确定参数化宏模型插值节点,提高了建模效率。最后,通过微带低通滤波器实例验证了所提方法的有效性。
A method for automatically generating parametric macro-model of linear circuit is proposed. The traditional parametric macro-model is created by barycentric interpolation of several univariate macro-models for different values of the design variables. But the values of the design variables of univariate macro-models are need to be determined artificially. It will result in low efficiency of modeling. To solve the problem,the values of the design variables of univariate macro-models are iteratively selected by gradually narrow the interpolation step with binary dividing algorithm until the model precision is satisfied with the predefined tolerance. An illustrative example involving a low-pass filter is presented for validation of the method proposed.