Design of nonlinear PID controller and nonlinear model predictive controller for a continuous stirred tank reactor

Area of Science:

• Chemical Engineering
• Control Systems Engineering
• Nonlinear Dynamics

Background:

• Nonlinear systems present significant challenges in process control due to their complex dynamics.
• Traditional linear control methods often fail to provide adequate performance for inherently nonlinear processes.
• Accurate modeling is crucial for designing effective control strategies for nonlinear systems.

Purpose of the Study:

• To develop and evaluate advanced control schemes for nonlinear dynamic processes.
• To implement a Nonlinear PID controller based on a family of local linear models.
• To introduce a Nonlinear Model Predictive Controller (NMPC) utilizing a family of local linear state-space models (F-NMPC).

Main Methods:

• Representing the nonlinear system as a family of local linear state-space models.
• Designing local PID controllers based on these linear models.
• Combining local controller outputs via a weighted sum to form the Nonlinear PID controller.
• Developing the F-NMPC using the identified family of local linear models.

Main Results:

• The proposed Nonlinear PID controller effectively managed the nonlinear process.
• The developed F-NMPC demonstrated robust control performance.
• Both control strategies were validated on a Continuous Stirred-Tank Reactor (CSTR) process exhibiting dynamic nonlinearity.

Conclusions:

• The presented control approaches offer viable solutions for managing nonlinear dynamic systems.
• The F-NMPC and Nonlinear PID controller show significant potential for industrial applications involving complex processes.
• The methodology of using a family of local linear models is effective for nonlinear system control.