In this context, we are taking a close interest in the
optimization of wind energy production. It consists in designing
simple to implement control strategies of a wind energy
conversion system, connected to the network based on the Double
Fed Induction Generator (DFIG) driven by the Converter
Machine Side (CSM) in order to improve the performance of
Direct Torque Control (DTC) and Direct Power Control (DPC).
For this purpose, the artificial neural networks (ANNs) is used.
Hysteresis comparators and voltage vector switching tables have
been replaced by a comparator based on artificial neural
networks. The same structure is adopted to build the two neural
controllers, for the DTC - ANN and for the DPC - ANN. The
simulation results show that the combination of classical and
artificial neural network methods permit a double advantage:
remarkable performances compared to the DTC-C and DPC-C
and a significant reduction of the fluctuations of the output
quantities of the DFIG and especially the improvement of the
harmonics rate currents generated by the machine.