Neural Network Technique for Direct Torque Control of Induction Motor Used in Electrically Driven Marine Propeller

Purwahyudi, Bambang and Ashari, Mochamad (2011) Neural Network Technique for Direct Torque Control of Induction Motor Used in Electrically Driven Marine Propeller. INTERNATIONAL JOURNAL Of ACADEMIC RESEARCH, 3 (5). pp. 358-364. ISSN 2075-4124 eISSN 2075-7107

	Text Neural_2011_IJAR_V3_N5.pdf Download (781kB)
	Text 3_NEURAL_NETWORK_TECHNIQUE_FOR_DIRECT_TORQUE_CONTROL-1.pdf Download (1MB)
	Text Coresponding IJAR_Neural.pdf Download (482kB)

Abstract

In the electric propulsion system, induction motor is used to drive the marine propeller. The control of induction motor is very complex and totally difference compared with in industries. This control is affected by ocean surface waves, ocean current, wind, weather and also ship motions such as yaw, pitch and rolls. In this paper, a neural network (NN) controller is designed to control the speed of direct torque controlled induction motor. NN is a machine like human brain with basic properties of learning capability and generalization. This network requires a lot of training to understand the characteristic of nonlinear dynamic system mapping. NN controller is trained using the input and output data of PI controller under normal and disturbance conditions. To examine the robustness of NN controller, the induction motor is used for electric propulsion system. In this system, load torque applied to the induction motor depends on rotor speed of propeller and also speed of induction motor depends on torque output of propeller. Simulation results of the designed NN controller show that the induction motor rapidly achieves the speed reference, without overshoot and small steady state error. And also, load disturbance is rapidly rejected.

Item Type:	Article
Uncontrolled Keywords:	Induction Motor, Direct Torque Control, Neural Network, electric propulsion system.
Subjects:	624 Civil Engineering
Divisions:	Faculty of Engineering > Bachelor of Electrical Engineering
Depositing User:	Perpus Ubhara Surabaya
Date Deposited:	13 Sep 2022 02:38
Last Modified:	14 Feb 2023 06:28
URI:	http://eprints.ubhara.ac.id/id/eprint/1372

Actions (login required)

View Item