Design of Combinatorial Structure for Nonlinear and/or Large-Scale Systems: Use of Convex Optimization and Sensitivity Analysis

2014 Fiscal Year Final Research Report

• Project/Area Number: 24560558
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Control engineering
• Research Institution: Nanzan University
• Principal Investigator: OISHI Yasuaki 南山大学, 理工学部, 教授 (80272392)
• Co-Investigator(Renkei-kenkyūsha): CHEN Gan 南山大学, 理工学部, 准教授 (00295718) ; OKUMURA Yasuyuki 南山大学, 理工学部, 教授 (90434604)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 制御理論 / 非線形制御 / ２乗和多項式 / モデル予測制御 / 量子化制御 / サンプル値制御 / 区分構造 / 分岐

Outline of Final Research Achievements

For control of nonlinear and/or large-scale systems, new approaches are explored through structure design of sum-of-squares polynomials and control systems. In particular, a new method based on a piecewise polynomial is investigated for design of a model predictive controller in order to avoid computation of a piecewise structure involved in the problem. Parameter control for a nonlinear dynamical system is considered with a stability index and a matrix inequality in order to suppress undesirable bifurcation and chaos. Moreover, in order to control a nonlinear system with a quantized input or a discrete-time input, new methodology is proposed using sum-of-squares polynomials, linear-like representation, the Picard iteration, and the contraction mapping theorem.

Free Research Field

工学