Prediction output current and output power of photovoltaic system using artificial neural network / Muhamad Azim Mohd Wadzir

Global climate change has already had observable effects on the environment. Glaciers have shrunk, ice on rivers and lakes is breaking up earlier, plant and animal ranges have shifted and trees are flowering sooner. Human activities such as burning fossil fuel and changes in land use, release large amounts of carbon to the atmosphere, causing C02 concentrations in the atmosphere to rise. The burning of fossil fuels has contributed to a 40% increase in the concentration of carbon dioxide in the atmosphere. In our country, the government already do an action to reduce the amounts of carbon to the atmosphere by using the renewable energy as a electric source to reduce the dependency on fossil fuel. Under the 8th Malaysia Plan (2001 - 2005) the government of Malaysia had changed the Four Fuel Policy to the Five Fuel Policy energy mix with the addition of renewable energy as the fifth sources of fuel in the year 2000. Solar is one of type of renewable energy, by using Photovoltaic (PV) it can convert solar irradiation into electricity. This paper presents an application of Artificial Neural Network (ANN) for prediction of output current and output power from Photovoltaic (PV). The output current and output power is predicted from PV system from Green Energy Research Centre (GERC) in Uitm Shah Alam. DC input current and voltage from two strings A and B is sets as input data in ANN program in order to predict the output current and output power from PV system. The performance of the system is measured by its regression value. The best performance of the system is when regression value is equal to 1. Usually, ANN will run heuristically for its parameter and in this paper ANN program is set to generate automatically random parameter. The purpose of prediction output current and output power from PV system using ANN is to evaluate the performance of PV system by using historical data from GERC. The results show that ANN program gives an accurate prediction.