Iranian Association of Chemical Engineering (IAChE) Iranian Journal of Chemical Engineering (IJChE) 1735-5397 Articles in Press 2026 07 31 Fuzzy Logic-based Gain Scheduling for PI control of a Nonlinear pH Process 92 111 248331 10.22034/ijche.2026.564760.1581 EN Sina Maleki Department of Chemical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Tehran 159163-4311, Iran 0009-0000-1651-7037 Ali Mohammad Sahlodin Department of Chemical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Tehran 159163-4311, Iran 0000-0002-2871-0923 Narges Fallah Department of Chemical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Tehran 159163-4311, Iran Journal Article 2025 12 10 Abstract – pH control is crucial in biological and water treatment processes, yet it poses significant challenges due to its nonlinear characteristics. Conventional Proportional-Integral-Derivative (PID) controllers, while operator-friendly, often struggle to maintain satisfactory performance in highly nonlinear systems, especially when subjected to disturbances, measurement delay, and measurement noise. On the other hand, gain scheduling is a technique to cope with the process nonlinearity while maintaining the PID simplicity for the operators. In this paper, a gain-scheduled digital proportional-integral (PI) controller for pH control is presented, where the controller parameters are adapted using fuzzy logic. The control error and its numerical derivative are fed to the fuzzy inference unit with 7 membership functions, fuzzifying the severity of the control deviation. Then, the PI proportional and integral gains are updated in the defuzzification step. A simulation of a benchmark pH process with measurement delay was carried out under various scenarios. The results show that the proposed fuzzy-PI controller significantly outperforms the conventional PI controller. This is especially true in cases of feed disturbance and considerable setpoint changes, where the nonlinear process deviates significantly from the nominal point. In a combined disturbance/setpoint change scenario, the fuzzy-PI controller reduces the integral of absolute error by 54%.