Research Article
Artificial Neural Network Based Rotor Flux Estimation and Fuzzy-Logic Sensorless Speed Control of an Induction Motor
@INPROCEEDINGS{10.1007/978-3-030-93709-6_19, author={Tefera T. Yetayew and Rahel S. Sinta}, title={Artificial Neural Network Based Rotor Flux Estimation and Fuzzy-Logic Sensorless Speed Control of an Induction Motor}, proceedings={Advances of Science and Technology. 9th EAI International Conference, ICAST 2021, Hybrid Event, Bahir Dar, Ethiopia, August 27--29, 2021, Proceedings, Part I}, proceedings_a={ICAST}, year={2022}, month={1}, keywords={ANN Fuzzy logic Induction motor drive Sensor less indirect field-oriented control}, doi={10.1007/978-3-030-93709-6_19} }- Tefera T. Yetayew
Rahel S. Sinta
Year: 2022
Artificial Neural Network Based Rotor Flux Estimation and Fuzzy-Logic Sensorless Speed Control of an Induction Motor
ICAST
Springer
DOI: 10.1007/978-3-030-93709-6_19
Abstract
This paper aims to design rotor flux estimation based on artificial neural network (ANN) and fuzzy-logic based sensor less speed control of control for an induction motor drive. Induction motors are widely used for industrial applications with better reliability compared with DC-motor drives. However, control techniques for wide range speed control are complex. To achieve wide speed control range, field oriented control techniques are recommended. From the areas of application point of view, field sensors may not operate properly may be due to frequent failure that needs sensor less control technique. Thus, the research in this paper focuses on application of artificial neural network for flux estimation and fuzzy logic sensor less speed control of induction motor drive system. Performance evaluation of the control and flux estimation is done using MATLAB tool. The training of ANN for the flux estimation converged with epochs of 1000 and mean squared error of 0.00061617. The simulation results for the reference step input of 100 rad/s, the system with PI controller showed 5.851% percentage overshoot and 0.2 s settling time. Whereas the fuzzy based system resulted in 0.505% percentage overshoot and 0.085 s settling time. In summary, the controller performance reveals better dynamic response can be achieved using fuzzy logic based system than the system based on the conventional proportional integral (PI) controller.
