A neural network based pwm type AC voltage restorer simulation is proposed in this paper. The objective is to apply the neural network switching control technique to the AC voltage restorer to decrease time delays during the switching conditions, to reduce switching losses. Thus, the proposed AC voltage restorer has some advantageous such as quick switching response, simpler and intelligent structure, better output waveform. With unknown and/or variable nonlinear loads, the semi-conductor's triggering and extinction angles depend on the load parameters in a highly nonlinear fashion. The Artificial Neural Network techniques have proven suitable for parameter identification and control of such nonlinear systems. The transient condition of the AC voltage restorer is improved via neural network based control technique. The simulation of the switching conditions is shown, and MATLAB is used to simulate and test its performance. Simulation results were performed to verify the effectiveness of the proposed control strategies.