1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Title:	Intelligent controller for multi three-phase induction motor drive using heuristic optimisation techniques to enhance performance efficiency
Authors:	Jamal Abd Ali Abdulridha (P73194)
Supervisor:	Mahammad A. Hannan, Prof. Dr.
Keywords:	Dynamic configuration Heuristic techniques Performance efficiency Intelligent control systems
Issue Date:	9-May-2016
Description:	Induction motor drive is widely used in many load applications in approximately 60% of the total industrial load. The dynamic configuration of the induction motor, especially a three-phase induction motor (TIM) is a nonlinear system which is not easy to explain theoretically due to sudden changes in load or speed. Thus, advanced and robust controller is needed to enhance the performance efficiency of the TIM. Fuzzy logic controller (FLC) has gained increasing popularity in designing complex TIM control system due to its simplicity and adaptability. However, the performance of FLC depends on the membership functions (MFs) variables which are determined by a heuristic procedure which is time consuming. To overcome this problem, a novel quantum-inspired lightning search algorithm (QLSA) is proposed to avoid the exhaustive conventional heuristic procedure for obtaining the MFs. To evaluate the reliability and efficiency of the developed QLSA, it is tested using fourteen benchmark functions with various characteristics. The QLSA is applied to improve the design of two control systems, namely, scalar and vector control in the multi induction motor drive. In scalar control system, QLSA is used for voltage/frequency control design with space vector pulse width modulation (SVPWM) to generate adaptive input and output MFs implemented in the fuzzy speed controller. In vector control system, QLSA is used to improve the indirect field-oriented fuzzy-proportional integral (PI) controller design with SVPWM to generate the best input and output fuzzy MFs and PI control parameters implemented in the speed and current controllers. The objective function considers the mean absolute error (MAE) of the rotor speed response for the TIM. Thus, an optimal QLSA-based FLC is employed to minimise the MAE, thereby improving the performance of the TIM with changes in speed and mechanical load. The designed QLSA-based FLC simulation model is implementation in the MATLAB/Simulink environment. Accordingly, QLSA-based optimal fuzzy speed controller is tested by experiment by using a fully integrated single DSP-TMS320F28335 for controlling five independent TIMs drive system. In addition, the random forest (RF) regression-based SVPWM is implemented for a twolevel inverter to minimise the complex online computation of conventional SVPWM. To validate the RF-based SVPWM, it is compared with the adaptive neuro fuzzy inference system and artificial neural network based SVPWM. Simulation results of the QLSA-based FLC are compared with the lightning search algorithm-based FLC, backtracking search algorithm-based FLC, gravitational search algorithm-based FLC, and particle swarm optimisation-based FLC. It is found that the accuracy of the QLSA-based FLC for speed control is superior compared to the other controllers in terms of transient response, damping capability and reduction of the MAE, root mean square error and standard deviation under different speeds and loads. The simulation results for maximum overshoot and settling time of speed response of the proposed controller are 0.4 % and 0.048 sec whereas the experimental results are 0 % and 0.5 sec, respectively. The performance of the proposed QLSA-based FLC speed controller has been validated in which the experimental results under different conditions show closed matching of the speed responses and stator currents with the simulation results.,Certification of Master's/Doctoral Thesis" is not available
Pages:	316
Call Number:	TJ217.5.A236.2016 3 tesis
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.