论文信息 - Global optimization of measurement strategies for linear stochastic systems

Global optimization of measurement strategies for linear stochastic systems

Abstract In this paper, nonlinear optimization problems arising from the determination of optimal measurement strategies for linear stochastic systems are analyzed. Since the minimum value of the mean square error is included in a further optimization process, a Riccati-type equation appears naturally as a constraint. In spite of its nonlinear nature, it presents a convexity property which makes possible the development of a global convergent algorithm even in the case of logical alternative decisions being under consideration. Two numerical examples are included.

José Claudio Geromel

[1] Michael Athans,et al. 1972 IFAC congress paper: On the determination of optimal costly measurement strategies for linear stochastic systems , 1972 .

[2] Y. Sawaragi,et al. Measurement optimization with sensitivity criteria for distributed parameter systems , 1980 .

[3] E. Balas. An Additive Algorithm for Solving Linear Programs with Zero-One Variables , 1965 .

[4] R. K. Mehra. Optimization of measurement schedules and sensor designs for linear dynamic systems , 1976 .

[5] J. Geromel. Convex analysis and global optimization of joint actuator location and control problems , 1989 .

[6] A. M. Geoffrion. Generalized Benders decomposition , 1972 .

[7] S. Omatu,et al. Optimal sensor location problem for a linear distributed parameter system , 1978 .

[8] José Claudio Geromel,et al. Nonlinear Programs with Complicating Variables: Theoretical Analysis and Numerical Experience , 1986, IEEE Transactions on Systems, Man, and Cybernetics.