Confined Frequency-Interval Gramian Framework-Based Balanced Model Reduction

Some of the existent frequency-limited model reduction techniques result in unstable reduced-order models (ROMs). Further, the ROMs of a few methods differ significantly from the actual model, resulting in a significant approximation error. Therefore, this work presents an innovative structure of confined frequency-intervals Gramians for continuous-time systems. The developed ROM ensures stability and minimal inaccuracy in approximation error over the desired frequency intervals. Also, the developed technique provides an easy-to-calculate error-bound. Compared to conventional approaches, the suggested approach offers consistent results, illustrating its usefulness. Further, the numerical cases support the effectiveness of the proposed scheme.