Modified fuzzy gain scheduling speed controller for BLDC with seamless speed reversal using direct commutation switching scheme

Pothorajoo, Satishrao (2018) Modified fuzzy gain scheduling speed controller for BLDC with seamless speed reversal using direct commutation switching scheme. Masters thesis, Universiti Malaysia Pahang (Contributors, Thesis advisor: Hamdan, Daniyal).

[img]
Preview
Pdf
Modified fuzzy gain scheduling speed controller for BLDC with seamless speed reversal using direct commutation switching scheme.wm.pdf

Download (4MB) | Preview

Abstract

Over the past decade the Brushless Direct Current (BLDC) motors have gained popularity in multiple sectors such as transportation, robotics, and automation. This is due to its high efficiency, low maintenance and high-power density. BLDC motor applications in automation, aerospace and industrial automation requires the motor to be operated bidirectionally. Although many BLDC drive controllers have been developed, most solutions are focusing only on forward motoring instead of bidirectional. A 4-quadrant BLDC controller was developed by S.Joice was the first to address this issue, although it is unclear whether the controller able to achieve bidirectional operations due to lack of evidence in the literature. To assess this bidirectional capability of S.Joice controller, a test platform was developed in MATLAB Simulink simulation. It is found that the controller is incapable of achieving reference speed in third quadrant and has 67.5 % of speed error during quadrant transient operations. To overcome this limitation, this study proposes a new BLDC commutation scheme, called direct commutation switching (DCS) scheme. A PID speed controller coupled with DCS scheme is tested for two test cases. From the test cases, it can be concluded that DCS scheme is able to drive the BLDC motor bidirectionally. Further analysis points out that the PID exhibits the typical unsatisfying performance under nonlinear load conditions. This is a classic problem that have lead many different types techniques to be developed, including Fuzzy Logic and Artificial Neural Network (ANN). Among others, fuzzy logic optimization technique is preferable due to simplicity compared other intelligent speed controller. This study attempts to develop a new BLDC controller by modifying fuzzy gain, hence proposes M.F.G.S speed controller. This proposed controller’s step responses are compared to PID and S.T.Fuzzy speed controllers under six test cases. The proposed controller has the shortest recovery time during load changes from no load to 5 Nm load. It is also able to adapt with sudden speed changes by achieving lowest steady state error. As for quick reversal operation, BLDC motor requires transient capabilities between quadrants. It is necessary to determine the instance when the rotor is ideally positioned for reversal to prevent standstill position. In order to examine the quadrant transient capabilities, M.F.G.S speed controller together with PID and S.T.Fuzzy speed controllers were evaluated under four cases of quadrant transient. From the study, the M.F.G.S controller had the lowest overshoot and steady state error of 0.015 % while transiting from first quadrant to second quadrant under loaded conditions. Overall, M.F.G.S Speed Controller for BLDC outperforms the other two controllers in this study. Hence, the M.F.G.S controller has potential to be used as bidirectional drive in highly dynamic load conditions.

Item Type: Thesis (Masters)
Additional Information: Thesis (Master of Science) -- Universiti Malaysia Pahang – 2018, SV: DR. HAMDAN BIN DANIYAL, NO. CD: 11573
Uncontrolled Keywords: Speed controller; Brushless Direct Current (BLDC); direct commutation switching
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculty/Division: Faculty of Electrical & Electronic Engineering
Depositing User: Mrs. Sufarini Mohd Sudin
Date Deposited: 29 May 2019 07:14
Last Modified: 24 May 2023 02:36
URI: http://umpir.ump.edu.my/id/eprint/24591
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item