Identification and control of nonlinear processes in the presence of unmeasured load disturbances

Identification and control of a nonlinear process in the presence of unmeas ured load disturbances is important, because most chemical processes are pe rturbed by load disturbances that are often not measured. In this paper, th e absorption principle is first extended to develop an effective identifica tion strategy for a feedforward neural network representation of the proces s input-output relation in the presence of an unmeasured load disturbance. This developed model can provide an accurate output prediction, irrespectiv e-of the load disturbances, as long as the disturbances can be reasonably a pproximated by piecewise polynomials. Second, a predictive control scheme i s developed on the basis of genetic algorithm optimization, using the above -identified model, for the nonlinear process under the influence of unmeasu red loads. Finally, simulations are provided to illustrate the effectivenes s of the proposed identification and control scheme.