Subspace geometry and performance tradeoffs of modal blending for active flutter suppression

Abstract Modal blending maps high-dimensional sensor and actuator arrays onto low-dimensional virtual inputs and outputs for targeted modal control. While $$\mathcal {H}_2$$ H 2 -optimal single-input single-output (SISO) blending has been proposed for active flutter suppression, the geometric and performance properties of the resulting virtual plant remain incompletely characterized. This paper develops a unified framework connecting subspace geometry, transmission-zero placement, and modal observability and controllability. Closed-loop performance depends on the blending subspace rather than its basis; an equivalence on the Grassmann manifold. An energy-normalized minimum singular value metric quantifies independent modal observability and controllability. An analytical zero map shows that the transmission zero known to be introduced by SISO blending of an oscillatory mode is governed by quadrature mismatch, which $$\mathcal {H}_2$$ H 2 -optimal direction selection does not prescribe. On a generic rectangular-wing model, multi-input multi-output (MIMO) blending spanning the full two-dimensional flutter subspace outperformed SISO blending in all configurations studied under a common structured $$\mathcal {H}_\infty $$ H ∞ synthesis. The advantage is explained jointly by subspace completeness and projection-induced zero placement.