Algoritme Genetika untuk Peningkatan Prediksi Kebutuhan Permintaan Energi Listrik

Oman Somantri (Politeknik Harapan Bersama Tegal)
Catur Supriyanto (Universitas Dian Nuswantoro)

Jurnal Nasional Teknik Elektro dan Teknologi Informasi
Vol 5 No 2: Mei 2016

Predicting the demand of electrical energy with a high degree of accuracy is expected. Application of an appropriate model using exact method will greatly affect the level of accuracy result. Neural Network (NN) and Support Vector Machine (SVM) models are used to predict the needs of electricity demand. Those models have weaknesses. Both are still difficult in determining the value of parameters used, thus, affecting the level of accuracy. Genetic Algorithm (GA) is proposed as a method to optimize the value of NN and SVM parameters in predicting the demand of electrical energy. The result shows that the NN and GA models have a better accuracy than the SVM and GA.