Switching over voltages analysis during shunt reactor energization using ANN

摘要

Transients caused by shunt reactor switching have been an important parameter in the design of the relevant equipment (reactor, circuit breaker, insulation) of power systems. In a small power system that appears in an early stage of a black ystart of a power system, an overvoltage could be caused by core ysaturation on the energization of a reactor with residual flux. Artificial neural network (ANN) is addressed in this work, in order to estimate the temporary overvoltages (TOVs) due to shunt reactor yenergization. yln proposed methodology, Levenberg-Marquardt ymethod is used to train the multilayer perceptron. ANN training is performed for a sample circuit based on equivalent circuit parameters of the network. Thus, trained ANN is applicable to every studied system. The ydeveloped ANN is trained with the worst case of the switching angle and remanent flux, and ytested for typical cases. The simulated results for a partial of 39-bus New England test system, yshow that the proposed technique can estimate the peak values and yduration of switching overvoltages with good accuracy