A remark on ‘Model reduction for singular systems via covariance approximation’

This note comments on the results of the paper, ‘Model reduction for singular systems via covariance approximation’, which studied model reduction for singular system via covariance approximation. Although the proposed new error criterion reflects the capacity of the impulsive behaviour for singular systems, there exist shortcomings due to the fixed matrix Br in the process of optimization, which remarkably matters. In order to avoid this drawback, the model reduction problem is reformulated and a genetic algorithm is used to deal with the optimization problem. A numerical example is provided to show the effectiveness and improvement of the proposed algorithm.