Peningkatan Sistem Prediksi Kebutuhan Energi Jangka Pendek Menggunakan Algoritma RVGA-ENM

Wahab Musa, Sardi Salim, Ifan Wiranto


Short term energy demand forecasting system is a tool in short term energy planning such as monthly energy needs in an area. This study aims to improve the ability of short-term prediction systems using the RVGA-ENM algorithm. The integration of the RVGA (real value genetic algorithm) and ENM (extended nelder mead) algorithm is a hybrid of two algorithms that complement each other. The ability of RVGA to explore the search for global optimal solutions and ENM in exploiting optimal local solutions, when combined will improve the accuracy of predictive systems. To test the performance of the proposed short-term energy demand system, the monthly electricity demand data in the Gorontalo Province area is used. Electrical energy needs data were obtained from PT. PLN Gorontalo Branch for a period of 72 months from January 2012 to December 2017. The test results showed a significant increase in the system of short-term energy demand prediction marked by MAPE error 3%, MSE 0.4% and RMSE 0.6%.


Sistem prediksi kebutuhan energi jangka pendek (short term energy demand forecasting) merupakan suatu alat bantu dalam perencanaan energi jangka pendek seperti kebutuhan energi setiap bulan disuatu wilayah. Penelitian ini bertujuan untuk meningkatkan kemampuan sistim prediksi jangka pendek menggunakan algoritma RVGA-ENM. Integrasi algoritma RVGA (real value genetic algorithm) dan ENM (extended nelder mead) merupakan hibrida dua algoritma yang saling melengkapi satu sama lainnya. Kemampuan RVGA dalam mengeksplorasi pencarian solusi optimal global dan ENM dalam mengeksploitasi solusi optimal local, ketika digabungkan akan meningkatkan akurasi sistim prediksi. Untuk menguji unjuk kerja sistim prediksi kebutuhan energi jangka pendek yang diusulkan, digunakan data kebutuhan energi listrik bulanan di wilayah Provinsi Gorontalo. Data kebuthan energi listrik diperoleh dari PT. PLN Cabang Gorontalo untuk rentang waktu 72 bulan dari Januari 2012 hingga Desember 2017. Hasil pengujian menunjukkan adanya peningkatan yang signifikan sistim prediksi kebutuhan energi jangka pendek ditandai dengan error MAPE 3%, MSE 0.4% dan RMSE 0.6%.


Full Text:



Khazem, H.A (2008). Using artificial neural networks to forecast the futures prices of crude oil, ProQuest Dissertations and Theses; 2008;

Musa, W., & Tansa, S. (2016). HAM: A Hybrid Algorithm Prediction Based Model For Electricity Demand. Journal of Theoretical and Applied Information Technology,92(2), 303-310.

Musa, W. (2016). A Hybrid Modern and Classical Algorithm for Indonesian Electricity Demand Forecasting, ARPN Journal of Engineering and Applied Sciences, 11(3), 1782-1786

Ozturk, H. K., & Ceylan, H. (2005). Forecasting total and industrial sector electricity demand based on genetic algorithm approach: Turkey case study. International Journal of Energy Research, 29, 829-840.

Piltan, M., Shiri, H., & Ghaderi., S.F. (2012). Energy Demand Forecasting in Iranian Metal Industry using Linear and Nonlinear Models Based on Evolutionary Algorithms. Energy Conversion and Management, 58, 1-9.

Musa, W., & Ridwan, W (2017). Aplikasi Algoritma Hibrida rvGA-eNM Untuk Prediksi Harga Energi Takterbarukan. Prosiding FORTEI 2017, 273-276 .


  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.