Output Voltage Regulation Techniques for Solar Fed Cascaded Multilevel Inverter

Abstract — An inverter is a circuit for converting direct current to alternating current. The unique structure of multilevel inverters allows reaching high voltages with low harmonics without the use of transformers, filters or series connected synchronized switching devices. This paper aims at improving the output voltage performance of a solar fed fifteen level cascaded multilevel inverter with Proportional Integral and Artificial Neural Networks based Controllers. The gain of PI (Proportional Integral) controller is tuned to generate a reference signal to regulate a constant output voltage of 325Vp. The proposed ANN (Artificial Neural Network) based controller is trained using Gradient Descent Back Propagation algorithm with various error signals and the obtained result is compared with conventional PI controller. The input solar panel with respect to variable irradiance is simulated in the MATLAB / Simulink. Thus by using the proposed system the output voltage of a solar fed cascaded multilevel inverter was maintained constant inspite of variations in the input