Application of Artificial Neural Networks in Power Electronics

Power Electronics is defined as electronic circuits used for electrical conversion. The conversion includes rectification and inversion. For a UPS inverter the output voltage obtained for the linear load is sinusoidal whereas for the non linear loading conditions the output voltage waveform is highly distorted. Hence it is necessary to maintain a sinusoidal output voltage for all loading conditions with minimum total harmonic distortion (THD). In this paper a neural network controller for UPS inverter applications is presented. The proposed neural network controller is trained offline using the patterns obtained from a simulated controller, which had an idealized load current reference. A sinusoidal Pulse width modulation (PWM) based switching UPS inverter has been modeled. The error in the output voltage and current are traced especially under non-linear loads. Simulation results shows that the proposed neural network controller can achieve low total harmonic distortion under nonlinear loading condition and good dynamic response under transient loading condition.