A nural network indentification method for the inverter

The paper investigates the use of neural networks for identification of power converter. A nonparametric model of the AC-to-DC inverter implemented by means of a neural network identifies the inverter dynamics in cases of uncertainty in the input voltage supply and/or load parameter. The circuit consists of a main circuit, control system, together with the source inductance and the output filter. The ultimate object is used in the analysis of power system. Simulation results in a PWM inverter under large-signal operation illustrate the applications and detect the characteristics of the circuit.