Direct Torque Control of Induction Motors Direct Torque Control of Induction Motors
by Use of the GMR Neural Network by Use of the GMR Neural Network
Rated power Prated [kW] 2.2 Rated voltage Urated [V] 220 Rated frequency frated [Hz] 50
Pole-pairs 2
Stator resistance Rs [] 3.88 Stator inductance Ls [mH] 252 Rotor resistance Rr [] 1.87 Rotor inductance Lr [mH] 252 3-phase magnetizing inductance Lm [mH] 236 Moment of inertia J [Kgm2] 0.0266
In this paper an application of the GMR to the control of an electrical drive with an induction motor has been presented. More specifically a particular strategy of the direct torque control has been taught to the GMR and the goodness of its learning capability has been verified by using experimental data obtained on a purposely developed test bench. The results show that this novel neural network gives at least the same results as those obtained in literature with other neural networks; but whereas those works aimed at solving an error minimisation problem by using well known feedforward neural networks, this work instead, considering that the task to be accomplished is the learning of a simple look-up table, exploits the pattern recognition capability of the GMR.
200 400 600 800 1000 1200
-4 -2 0 2 4
Samples
Output error
TEST SET
0 200 400 600 800 1000 1200 1400 1600 1800 2000
-4 -2 0 2 4 6
Samples
Output error
TRAINING SET
Giansalvo Cirrincione Chuan Lu
Department of Electrical Engineering Katholieke Universiteit Leuven University of Picardie-Jules
Verne Afd. ESAT SCD ( SISTA)
Maurizio Cirrincione, Marcello Pucci
I.S.S.I.A.-C.N.R. Section of Palermo (former CE.RI.S.E.P.) (Institute on Intelligent Systems for the Automation)
From the classical DTC to the GMR based DTC
input
neuron
first layer weights second layer weights
chain
3 inputs:
3 inputs: torque error, stator flux linkage error and sector in which it lies 1 output:
1 output: voltage space vector to be generated by the inverter.(6 discrete values corresponding to 6 switching patterns of power devices)
The test set-up for the experimental application of the GMR based DTC
Experimental results
5 10 15 20 25
-150 -100 -50 0 50 100 150
time (s)
speed (rad/s)
5 10 15 20 25
-20 -10 0 10 20 30
time (s)
torque (Nm)
The Generalized Mapping Regressor (GMR) Neural Network:
incremental, self organizing, transforms the mapping problem f : x y into a pattern recognition problem in the augmented space Z, z=[xT yT]T
The weights: 1st layer, continuous, computed by a multiresolution quantization; 2nd layer, discrete (chains between neurons), representing the mapping branches. computed via PCA techniques.
