pH control in biological process using MMPC based on neuro-fuzzy model by LOLIMOT algorithm

pH control is considered as one of the most important issues in chemical and biological processes. Although process has simple components but control of pH in output effluent is difficult in application. The main reasons of this difficulty are highly nonlinearity and time varying nature of the process. Multi-model predictive controller using neuro-fuzzy model based on LOLIMOT algorithm is employed to control pH value in this study. The distinctive features of these controllers that can be expressed are the ability to generalize to multi-variable systems, design in time domain, the ability to handle system with delay, nonlinear and non-minimum phase processes. For this purpose nonlinear process is divided into local linear model using LLNF (Local Linear Neuro-fuzzy) model, each linear model is in CARIMA format and generalized model predictive controller is designed for each linear model and final control input is weighted of controller output of each linear model. Finally by the implementation of designed controller on experimental setup, improvement of responses can be observed.