In this paper, a novel Set Point Weighted Proportional Integral Derivative (SPWPID) controller has been proposed for the Magnetic levitation (Maglev) system in Simulink and real time. The recently evolved Teaching Learning Based Optimization (TLBO) has been used to identify the suitable controller parameter values by optimizing the objective function. The performance of the SPWPID controller is compared with that of the PID controller, designed using TLBO. Moreover, the performance of the SPWPID controller has also been compared with the performance of the 1-DOF and 2-DOF PID controller designed for the same Maglev plant [15]. The result of the comparison shows that the SPWPID controller outperforms both the 1-DOF and 2-DOF PID controllers in terms of overshoot and settling time. The robustness analysis has also been incorporated to demonstrate the robust behavior of the plant with the SPWPID controller.Keywords: Maglev, Set-Point Weighted PID, PID, TLBO.
IntroductionMaglev is an example of an inherently nonlinear and unstable system. Because of these properties, it becomes difficult to design a controller which will efficiently control this system. The application of Maglev can be found in different fields of research which include high-speed transportation systems [1], photolithography devices for semiconductor manufacturing [2], seismic attenuators for gravitational wave antennas [3], self-bearing blood pumps [4] for use in artificial hearts etc. Because of such vast applications, it becomes extremely important to develop a proper control strategy for the Maglev system. The literature review shows that the design of the controller utilizes different control techniques such as sliding mode control, ∞ control, TID and I-TD control, fractional order control etc. In addition, the use of the evolutionary algorithm, fuzzy logic and neural network can also be found in the literature [5][6][7][8][9][10][11].Because of its simple structure and easy implementation, the PID controller has always been a favorite choice for the control engineers and has been applied in several fields of research [12][13][14]. To provide an overall good performance, different techniques are chosen but a controller is rarely found where it has got the potential to provide a good response in all respects.This paper provides a novel approach, where an SPWPID controller is designed to efficiently control the Maglev system. The recently evolved Teaching Learning Based Optimization (TLBO) has been used to identify the suitable controller parameter values by optimizing the objective function. The performance of the SPWPID controller is compared with that of the PID controller designed using TLBO. Moreover, the performance of the SPWPID controller has also been compared with the performance of the 1-DOF and 2-DOF PID controller designed for the same Maglev plant [15]. The result of the comparison shows that the SPWPID controller outperforms both the 1-DOF and 2-DOF PID controllers in terms of overshoot and settling time.