Design and Implementation of a Novel Self-adaptive Fuzzy Logic Controller for a pH Neutralization Process

Abstract

This paper presents a self-adaptive fuzzy logic controller for a pH neutralization process which has been designed and implemented in real time process. The controller deals with the minimization of both the steady state error and time taken to reach steady state under varying operating conditions. The interesting fact about this control technique is that it operates a full-scale pH neutralization plant using the same equipment like that in real time full-scale industrial pH neutralization plant. The proposed self-adaptive technique updates the control parameters and the normalized universe of discourse of output fuzzy membership functions using varying scaling factors based on error and change of error values. A combination of both the coarse control and fine control was used for the overall control action. The speed of the response of the self-adaptive controller is handled by the coarse control technique while the amount of deviation of the measured values with respect to the setpoint under steady state is handled by the fine control technique. The performance of the envisaged selfadaptive scheme is tested for pH control in real time, and results are encouraging. MATLAB TM SIMULINK TM software is used for control, display as well as online communication with the DCS system through an OPC server configured on a PC. The controller is easily implementable and user-friendly and can, therefore, be adopted by process engineers. Lastly, the control technique envisaged herein can be implemented on preexisting control loops, hence alleviating financial constraints and great savings benefits of the controller.

Proof-of-Concept Design of an In-Line pH Neutralization System with Coarse and Fine Adjustments for the Continuous Manufacturing of Pharmaceuticals

“…To overcome such control problems, process models and sophisticated algorithms have been developed. Examples include genetic adaptive PI control using internal model control (IMC), adaptive nonlinear feedback control, model predictive control (MPC) based on a dynamic matrix (DMC), model-free learning control (MFLC), Wiener and Hammerstein models, strong acid equivalent control, fuzzy control, ,, neural networks, and hybrid models that integrate multiple control strategies. − Although these proposed models have demonstrated an excellent ability to precisely control pH, most of them have only been examined via simulation. They do not account for potential disturbances and other technical difficulties associated with experimental and physical industrial implementation in real-time manufacturing processes ( e.g ., nonuniformity of the feed stream, flow rate disturbance, and variations in equipment performance).…”

“…17 To overcome such control problems, process models and sophisticated algorithms have been developed. Examples include genetic adaptive PI control using internal model control (IMC), 18 adaptive nonlinear feedback control, 19 model predictive control (MPC) based on a dynamic matrix (DMC), 20 model-free learning control (MFLC), 21 Wiener and Hammerstein models, 22 strong acid equivalent control, 23 fuzzy control, 17,24,25 neural networks, 26 and hybrid models that integrate multiple control strategies. multiple tank reactors in series to eliminate nonlinear characteristics.…”