Performance Comparison of Starting Speed Control of Induction Motor

Deddy Kusbianto

Abstract


In the induction motor speed control without sensors operated by the method Field Oriented Control (FOC) was required an observer to estimate the speed. Obsever methods have been developed, among others, was the method of Self-Constructing Fuzzy Neural Network (SCFNN) with some training algorithms such as backpropagasi (BP). Levenberg Marquard (LM) etc.. In the induction motor control techniques were also developed methods of Direct Torque Control (DTC) with observer Recurrent Neural Network (RNN). This paper compares the performance of the motor response to initial rotation between SCFNN observer method that uses the LM training algorithm with DTC control technique with RNN observer. From the observation performance of the motor response to initial rotation of the two methods shows that the LM method has better performances than the RNN. This can be seen on both the parameters : overshoot, rise time, settling time, peak and peak time. With the right method, can enhance better performance of the system. With the improvement of system performance, is expected to increase work efficiency in the industrial world, so overall, particularly for systems that require high precision, FNN methodcan be said to be better. Keywords: Motor Speed control without sensors, FOC, SCFNNO, DTC, Levenberg Marquardt and RNN.

DOI: http://dx.doi.org/10.11591/ijpeds.v1i1.66


Full Text:

PDF

Refbacks

  • There are currently no refbacks.


Copyright (c) 2011 Deddy Kusbianto

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.