Optimization of Solar Energy Using Recurrent Neural Network Controller

Abstract

The use of solar panels has some advantages over other conventional electrical generating methods, as there is no sound pollution in collecting solar energy using solar panels, and also it has a minimum need for maintenance. In addition, it helps in the greenhouse effect which does not contribute to any CO2 pollution, as the conversion of light to electricity does not contain any chemical reactions. Using photovoltaic (PV) systems that are connected to a load will require a Maximum Power Point Tracker (MPPT) to maintain the highest possible efficiency of power generated. The resistance of the PV panels is different from the load resistance, the MPPT will control the duty cycle of the Insulated Gate Bipolar Transistor (IGBT) in the DC-DC converter to match the PV and load resistance for best efficacy. However, the use of MPPT with the connection to a controller collecting the maximum power generated from the PV system. In this paper, we design and implement a Recurrent Neural Network (RNN) based MPPT method to improve the efficiency of the power observation for the PV system for any value of irradiation (G) and temperature (T). Mainly, we compare two controller methods, using 104 sets of data for an ANN controller that was designed and tested in the past, with the same 104 sets of data to train the proposed RNN controller, as ANN used prediction in its calculations to find the best output efficiency, RNN will use a recurrent connection in the hidden layers that allow information to flow from one input to another.