z-logo
open-access-imgOpen Access
Prediksi Konsumsi Energi Pada Bangunan Menggunakan Metode Support Vector Machice Berbasis Algoritma Genetika
Author(s) -
Wida Prima Mustika
Publication year - 2017
Publication title -
jurasik : jurnal riset sistem informasi dan teknik informatika/jurasik (jurnal riset sistem informasi dan teknik informatika)
Language(s) - English
Resource type - Journals
eISSN - 2549-7839
pISSN - 2527-5771
DOI - 10.30645/jurasik.v2i1.14
Subject(s) - mean squared error , support vector machine , genetic algorithm , feature selection , energy consumption , computer science , feature (linguistics) , energy (signal processing) , mistake , selection (genetic algorithm) , algorithm , mathematical optimization , artificial intelligence , mathematics , machine learning , statistics , engineering , linguistics , philosophy , law , political science , electrical engineering
Energy consumption is a demand for the amount of energy that must supply the building at any given time. Building energy consumption has continued increased over the last few decades all over the world, and Heating, Ventilating, and Air-Conditioning (HVAC), which has a catalytic role in regulating the temperature in the room, mostly accounted for of building energy use. Models created for in this study support vector machine and support vector machine-based models of genetic algorithm to obtain the value of accuracy or error rate or the smallest error value Root Mean Square Error (RMSE) in predicting energy consumption in buildings is more accurate. After testing the two models of support vector machines and support vector machines based on the genetic algorithm is the testing results obtained by using support vector machines where RMSE value obtained was 2,613. Next was the application of genetic algorithms to the optimization parameters C and γ values obtained RMSE error of 1.825 and a genetic algorithm for feature selection error RMSE values obtained for 1,767 of the 7 predictor variables and the selection attribute or feature resulting in the election of three attributes used. After that is done the optimization parameters and the importance of the value of feature selection mistake or error of the smallest RMSE of 1.537. Thus the support vector machine algorithm based on genetic algorithm can give a solution to the problems in the prediction of energy consumption rated the smallest mistake or error.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here