Control of pH processes using fuzzy modeling of titration curve

Model-based pH control is a difficult problem due to its inherent nonlinearity and time-varying characteristics. This is specially the case when the pH regulation of streams consisting hundreds of constituents with varying concentrations is desired. The pH process can be modeled by using conservation laws and neutrality condition. An alternative to this mechanistic modeling is using nonlinear empirical modeling such as fuzzy modeling. This paper proposes a new approach of fuzzy modeling of titration curves, which can be used for control purposes. The Takagi–Sugeno model was used for the modeling of titration curve. Two control algorithms, which are designed based on the titration curve, namely, Strong Acid Equivalent (SAE) and Multiple Model, Switching and Tuning (MMST) have been used. The efficacy of the proposed approach is demonstrated via simulation and experimental study of a pH control system.