System Approximation Using Pole Spectrum Analysis (PSA) and Modified Cauer Continued Fraction Method

The manuscript presents an advanced approach for reducing large-scale Linear Time-Invariant (LTI) dynamic systems, utilizing a combined framework that integrates Dominant Pole Retention (DPR), Pole Spectrum Analysis (PSA), and Modified Cauer Continued Fraction (MCCF) method. The denominator dynamics of the reduced-order model are extracted via the PSA integrated with a pole dominance technique, while the numerator dynamics are obtained through MCCF method. Notably, for higher-order stable systems, the approach significantly guarantees the stability of the simplified system. To substantiate the efficacy and fidelity of the proposed technique, a set of numerical exemplars sourced from literature has been rigorously analyzed. Comparative evaluations have been conducted, leveraging performance indices and response congruence metrics to benchmark the alignment between the original and simplified system responses, thereby affirming the robustness and precision of the proposed reduction method.