Premium
Robust synthesis via bilinear matrix inequalities
Author(s) -
Goh K.C.,
Safonov M. G.,
Ly J. H.
Publication year - 1996
Publication title -
international journal of robust and nonlinear control
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.361
H-Index - 106
eISSN - 1099-1239
pISSN - 1049-8923
DOI - 10.1002/(sici)1099-1239(199611)6:9/10<1079::aid-rnc270>3.0.co;2-#
Subject(s) - bilinear interpolation , matrix (chemical analysis) , inequality , mathematics , computer science , mathematical optimization , algebra over a field , pure mathematics , materials science , mathematical analysis , statistics , composite material
The μ/K m ‐synthesis formulation for robust controller synthesis with frequency dependent scalings/multipliers is shown to be equivalent to a bilinear matrix inequality (BMI) feasibility problem. Formulating the non‐convex μ/K m ‐synthesis problem as a BMI problem allows local joint controller and multiplier synthesis using finite dimensional LMI‐like optimization methods and also the application of global optimization techniques. The BMI approach improves on the standard D , G (SINGLE BOND) K ‐iteration, M (SINGLE BOND) K ‐iteration approaches to μ/K m ‐synthesis, and additionally allows controller order and structure constraints.