Comparative Analysis and Implementation of an Ant Colony and Cuckoo Search Algorithm for the Time Delay System

Abstract

PID controllers are the most usually utilized controllers in the industries. The reason behind this is because of its easy configuration and reliability. Nonetheless, when the plant to be managed is highly nonlinear or is subjected to disturbances, or we have less knowledge about it, under these conditions, poor performance is achieved when we utilize fixed-parameter PID as a controller. Therefore, a professional supervisor has necessitated online tuning its controller parameters. Other than because of the recognized shortcomings of a human supervisor, such as chances of poor tuning or harm to the plant caused due to inattention of the supervisor, automatic tuning is preferred. Using an automation algorithm for the supervisor’s role is a good option because of its known qualities of mimicking human operators and robustness. However, an expert is still necessary to decide the parameters of membership functions; good information about the system is also required for tuning parameters. So, we can use an optimization technique that can adjust the parameters of the plant, which, as a result, gives us a superior outcome. The paper utilizes an ant algorithm (AA) as a new nature-inspired technique to tune a PID. For tuning purposes, the functions that are used are as follows: Integral Squared Error (ISE) as well as Integral Absolute Error (IAE); the result of the ant algorithm (AA) PID tuning process is also compared with the outcomes of the cuckoo, search Internal Model Control, Iterative Feedback Tuning methods, and Ziegler-Nichols. The PID controllers optimized with the ant algorithm and the Cuckoo search algorithms provide performance in the form of a simulation Graph. An earlier settling time, less or almost no overshoot, and higher robustness were accomplished with the proposed method. The new tuning process is triumphant even in the presence of high disturbance.