Solar generation prediction model by using statistical method

Renewable Energy play vital role power generation and become one reliable source of power generation. Various source of renewable energy which introduce by researcher and solar power generation is one of it. The Solar Energy is become one of major contribute for world power generation and the power generation rapidly increase. Malaysia was located with high intensity of light and which continuously up to eleven hours each day. It creates a good path for harvesting the natural and environment free energy for support nation need. The solar prediction modules will give a huge impact on the managing solar power plants and the designation of solar energy system and this paper attempts to develop a program that is able to predict the solar temperature module through the implementation of Statistical Method which is Multiple Discriminant Analysis and Linear Support Vector Method. The data from the weather variables which are the ambient temperature (TA), solar radiation flux (GT), the relative humidity (RH) and local wind speed (Vw) will be used in this project. Multiple Discriminant Analysis and Linear Support Vector Method which can be used for data classification and system approximation and prediction method will have selected to predict solar temperature module by analyzing the data collected based on three different variables and are classified into three condition which is either perfect, good or poor. This project shows that the Multiple Discriminant Analysis identify more accuracy among other technique in term of prediction of solar temperature module in Lestari building in NDUM.