Measurement selection for linear multivariable control systems (Q1119224)

This paper examines the selection of measurement variables for control system design in deterministic and stochastic systems. The observation selection problem is formulated as a nonlinear 0-1 integer programming problem and is shown that constraints and performance measures satisfy certain monotone conditions which permit application of the Lawler-Bell algorithm. A new algorithm is proposed which reduced the number of iterations needed to solve the problem via the Lawler-Bell algorithm. A boiler example is used to illustrate the techniques.