Interval Type-2 Fuzzy PID Controller Using Disassembled Gradational Optimization

Area of Science:

• Control Systems Engineering
• Computational Intelligence
• Fuzzy Logic Systems

Background:

• Traditional PID controllers struggle with nonlinear systems and uncertainties.
• Interval Type-2 Fuzzy Logic Systems (IT2-FLS) offer enhanced robustness but require complex design.
• Optimization of fuzzy logic controllers is computationally intensive.

Purpose of the Study:

• To introduce a novel Interval Type-2 Fuzzy Proportional-Integral-Derivative (IT2F-PID) controller.
• To develop and validate a new Disassembled Gradational Optimization (D-GO) method for designing IT2F-PID controllers.
• To evaluate the performance and efficiency of the proposed D-GO method against existing optimization techniques.

Main Methods:

• A PID controller is optimized using the D-GO method.
• A Type-1 Fuzzy Logic System (T1-FLS) is integrated and its parameters optimized.
• The T1-FLS is blurred into an IT2-FLS to form the IT2F-PID controller.
• Comparative simulations are conducted against general and concurrent optimization methods.

Main Results:

• The D-GO method reduced optimization time by over 90% compared to the general method and by over 25% compared to the concurrent method.
• Integral-of-Time-Absolute-Error (ITAE) was decreased by 30% (vs. general) and by approximately 95% (vs. concurrent).
• The proposed IT2F-PID controller reduced overshoot by 80% and fluctuation by 67% compared to traditional PID and general IT2F-PID controllers.

Conclusions:

• The D-GO method offers a highly efficient approach for designing IT2F-PID controllers.
• The proposed IT2F-PID controller demonstrates superior performance in handling nonlinear systems, model uncertainty, target uncertainty, and external disturbances.
• This research provides a significant advancement in robust and efficient control system design.