Multi-Model Convolutional PFC for the Control of Activated Sludge Reactors

Abstract

This paper investigates the use of Predictive Functional Control algorithm by replacing the classical internal model with an impulse response of the process under control as Convolutional Predictive Functional Control Conv-PFC. Convolution functional control has been successfully applied in industry. The use of the Conv-PFC to a reduced bio-reactor activated sludge ASM1 model, which describes the biological degradation of an activate sludge reactor, is designed based on several simplifications, as a TS model, witch structure is based on a set of linear sub models, covering the process input-output space, interpolated by a nonlinear weighting function. The Conv-PFC controller is compared to a benchmark PID in terms of error and response dynamics.