说明：  针对 Prony 算法辨识传递函数的模型阶数选取问题，提出了一种基于 SNR 值及留数值的 新方法。SNR 值体现的是拟合性能，留数值体现的是模式参与程度。首先选取一个阶数初始值， 然后在模型阶数取初始值条件下对输出信号进行 Prony 分析，最终依据 SNR 值及留数模值，得到 适合的模型阶数。对典型传递函数的仿真分析验证了所提方法的有效性。
(A new method based on SNR and residual value is proposed to solve the problem of model order selection in Prony algorithm. The SNR value reflects the fitting performance, while the remaining value reflects the degree of mode participation. First, an initial value of order is selected, and then Prony analysis is performed on the output signal under the condition of taking the initial value of model order. Finally, according to SNR value and residual modulus value, the appropriate model order is obtained. The simulation analysis of typical transfer functions verifies the effectiveness of the proposed method.)