Parametric Eigen-space Characterization via Static Output Feedback in MIMO Control Systems: A New Look to Output Feedback Gain Matrix Design for Turbo-Generator System

This paper presents a novel parametric output feedback strategy for MIMO control systems, enabling full closed-loop eigenvalue assignment through structured transformations without relying on prior state feedback or restrictive assumptions. The method uses a sector matrix to capture spectral characteristics of the closed-loop system and employs a convex LMI-based optimization framework for feedback gain computation. Its performance is validated on a benchmark turbo-generator system characterized by large scale and tightly coupled dynamics. Compared to conventional techniques such as adaptive pole placement and eigenstructure assignment, which often involve high control effort, sensitivity to disturbances, or limited precision, the proposed approach delivers superior results. It achieves significantly lower steady-state errors (RMSE: 0.013 and 0.009), minimal control input magnitudes (maximum of: 14.12 and 3.12 for the two inputs), reduced spectral sensitivity (2.12), and improved robustness to eigenvalue perturbations (relative variation 0.018). These findings underscore the method’s practical applicability, computational efficiency, and robustness, making it a strong candidate for high-performance MIMO control applications.