Local Nonlinear Control of a Process Actuator

Abstract

Through the use of nonlinear control techniques, a locally intelligent actuator strategy is developed such that a control valve is optimized independent of the central digital control system. A state-space model of a pneumatic process valve is developed using force balances on valve components and effective friction effects. Nonlinear compensation is implemented through the use of input-output linearization to cancel friction nonlinearities, closed-loop observers to estimate non-measurable parameters, and methods for compensation of adaptive friction parameters.