Suboptimal stochastic control of a continuous latex reactor

Abstract A suboptimal control algorithm is derived for a continuous latex reactor subject to deterministic and stochastic disturbances to control the pathological behavior of the reactor. An extended Kalman filter based on an instantaneous local linearization technique is used to estimate the reactor states, and a quadratic performance index is used to derive the suboptimal feedback control law. This study successfully demonstrates the application of the multivariable control theory to the control of a simulated continuous latex reactor operating under limit‐cycle conditions. The mathematical model used for the control studies was previously developed and has been shown to describe the performance of an actual reactor system extremely well over its entire range of operation.