Prediction of solar energy radiation using adaptive neuro fuzzy inference system in the tropical region

Abstract

The research aims to develop a prediction system of solar energy radiation in the tropical region for home energy needs. The tropics get sunlight all year round. The Solar energy radiates abundantly in the tropic. The Solar radiation Energy is influenced by weather factors. The input data in this study use weather factors such mximum temperature, minimum temperature, Precipitation, wind speed and relative humidity. The input data retrieves in 3 regions are Klaten, Sragen and Sukoharjo. The prediction system was built by using soft computing. The method used in this prediction system is an adaptive neuro fuzzy inference system (ANFIS) using hybrid Hybrid optimization method was chosen because it produces a calculation of RMSE (root mean square error) which is lower than backpropagation. methods. Hybrid optimization method was chosen because it produces a calculation of RMSE (root mean square error) which is lower than backpropagation. The prediction results were the values of solar energy in each region.The research aims to develop a prediction system of solar energy radiation in the tropical region for home energy needs. The tropics get sunlight all year round. The Solar energy radiates abundantly in the tropic. The Solar radiation Energy is influenced by weather factors. The input data in this study use weather factors such mximum temperature, minimum temperature, Precipitation, wind speed and relative humidity. The input data retrieves in 3 regions are Klaten, Sragen and Sukoharjo. The prediction system was built by using soft computing. The method used in this prediction system is an adaptive neuro fuzzy inference system (ANFIS) using hybrid Hybrid optimization method was chosen because it produces a calculation of RMSE (root mean square error) which is lower than backpropagation. methods. Hybrid optimization method was chosen because it produces a calculation of RMSE (root mean square error) which is lower than backpropagation. The prediction results were the values of solar energy in each region.