IJRR

International Journal of Research and Review

| Home | Current Issue | Archive | Instructions to Authors | Journals |

Research Paper

Year: 2018 | Month: October | Volume: 5 | Issue: 10 | Pages: 101-107

Optimization of PID Controller Using Crow Search Algorithm and Bat Algorithm

A. A. A. Wahab, S. S. N. Alhady, N. H. B. Sofik, W. A. F. W. Othman

School of Electric and Electronic Engineering, Universiti Sains Malaysia (USM) Transkrian, Nibong Tebal, Pulau Pinang, Malaysia

Corresponding Author: S. S. N. Alhady

ABSTRACT

Engineering field usually requires the best design for an optimum performance, thus optimization plays an important part in this field. Among optimization algorithms, crow search algorithm (CSA) and bat algorithm (BA) were chosen for this study. In this study, the effectiveness of CSA and BA were compared by analyzing the minimization of objective function and the step responses of the closed loop systems with proportional-integral-differential (PID) controller. The four criteria of objective functions are comprised as overshoot, settling time, rise time and steady state error. These are compared with acceptable solutions and the best one is selected with respect to optimized design. The parameters of PID controller i.e. Kp, Ki and Kd must be properly selected as the selection affects the transient response of a system. The best combination of parameters will reduce problems such as nonlinearities encountered by industrial plants. Then, comparisons and investigations were made based on the average and standard deviation of cost returned by the cost function. It can be seen that the ability to subdivide automatically in BA search was a short one in finding the optimal solutions. The results show improvement 1.06% in overshoot percentage after modification. The objective function value is increased for BA 1.443e-04 and the CSA is 7.5113e-07 considering the performance parameter needs to be optimized to minimize the error signal. Even though BA had a lower convergence rate compared to CSA but it managed to find the optimal solutions.

Key words: PID controller; Crow Search algorithm; Bat Algorithm; Error criteria

[PDF Full Text]