Comparison between Neural Network (NN) and Adaptive Neuro Fuzzy Inference System (ANFIS) on sunlight intensity prediction based on air temperature and humidity

Rahmalia, Dinita and Herlambang, Teguh and Kamil, Ahmad Syafiq and Rasyid, Reizano Amri and Yudianto, Firman and Muzdalifah, Lilik and kurniawati, Eriska Fitri (2019) Comparison between Neural Network (NN) and Adaptive Neuro Fuzzy Inference System (ANFIS) on sunlight intensity prediction based on air temperature and humidity. Journal of Physics: Conference Series, 1538. ISSN 1742-6596

Preview	PDF Comparison between Neural Network (NN) and Adaptive Neuro Fuzzy Inference System (ANFIS) on sunlight intensity prediction based on air temperature and humidity.pdf Download (1MB) | Preview
Preview	PDF peer review reizano amri rasyid.pdf Download (849kB) | Preview
Preview	PDF turnitin reizano amri rasyid.pdf Download (2MB) | Preview

Abstract

Weather prediction expecially in predicting sunlight intensityhas important role in energy usage. As effort for controlling petrol-based fuel usage, government manage energy usage by converting solar energy from sunlight intensity to electric through solar cell.Sunlight intensity depends on air temperature and humidity. Two methods on prediction process will be applied : Neural Network (NN) and Adaptive Neuro Fuzzy Inference System (ANFIS). Type of NN used in prediction process is Backpropagation. Backpropagation consists of forward propagation, backward propagation, and update weight matrices. ANFIS uses hybrid method to train consequent parameters and premise parameters. In this research, NN will be compared with ANFIS. From five trials of NN simulations, the number of maximum epoch for making the root of mean square error(RMSE) in training data is very large so that the computation time is very long. From the comparison result of two methods, we can see that ANFIS can make faster prediction than NN with the number of maximum epoches is smaller than NN so that computation time is faster.

Item Type:	Article
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Divisions:	Faculty of Economics and Business > Program Study of Management
Depositing User:	Mr. . Adit
Date Deposited:	14 Sep 2022 04:16
Last Modified:	14 Sep 2022 04:17
URI:	http://repository.unusa.ac.id/id/eprint/8834

Actions (login required)

View Item