Data-Driven Regularized Time-Limited h2 Model Reduction from Noisy Impulse Responses

TL;DR

This paper introduces a data-driven regularized approach for time-limited h2 model reduction of discrete-time systems using noisy impulse response data, improving accuracy over existing methods.

Contribution

It formulates and solves a regularized time-limited h2 model reduction problem solely from noisy data, with explicit gradient and objective representations.

Findings

The method achieves lower relative errors than alternatives.

It remains effective under noisy conditions where unregularized methods fail.

Numerical experiments validate the approach on benchmark systems.

Abstract

This paper develops a data-driven time-limited h2 model reduction method for discrete-time linear time-invariant systems. Specifically, we formulate and solve a regularized time-limited h2 model reduction problem using only noisy impulse response data. Furthermore, we show that the objective function and its gradient can be represented using only noisy impulse response data. Numerical experiments using SLICOT benchmarks demonstrate that the proposed regularized method achieves lower relative time-limited h2 errors than the tested alternatives and is effective in situations where the unregularized method may deteriorate under noise.