PENGARUH JUMLAH DATA LATIH SVM PADA PERAMALAN BEBAN LISTRIK BULANAN DI SEKTOR INDUSTRI
DOI:
https://doi.org/10.30595/techno.v17i2.1219Abstract
Abstrak Peramalanterhadap beban listrik bulanan di industriadalah sangat penting dalam perencanaan dan pengawasan serta monitoring penggunaan listrik di sektor indsutri, sehingga tingkat keakuratan sistem peramalan sangat diperlukan.Suport Vector Machine (SVM) merupakan salah satu metode peramalan yang berbasis kecerdasan buatan dapat memberikan nilai akurasi yang baik, dengan hasil nilai error yang kecil.Sebagai sistem cerdas, SVM memerlukan pelatihan pada sistem nya.Penelitian ini mengkaji pengaruh jumlah data latih yang diberikan pada SVM sebagai metode peramalan beban listrik. Pengujian dilakukan dengan memberikan variabel jumlah data dari 12 data sampai 72 data latih, dan SVM menggunakan fungsi Kernel Gaussian RBF. Hubungan antara jumlah data latih dengan nilai kesalahan peramalan (yang diukur dengan MAPE) dianalisis dengan korelasi Person’s. Penelitian ini menghasilkan nilai Korelasi Person’s r= -0,96 memberikan arti bahwa antara jumlah data latih SVM dengan nilai MAPE yang didapatkan dalam prakiraan SVM terdapat hubungan negatif dengan kekuatan hubungan yang kuat. Kata kunci : Peramalan Beban Listrik, SVM, Data Latih, Korelasi Person’s Abstract Forecasting of the monthly electricity load in the industry is very important in the planning, supervision and monitoring of electric energy use of industrial sector, so the level of forecasting system accuracy is required. Support Vector Machine (SVM) is one forecasting method based on artificial intelligence can give good accuracy, with the result that a small error value. As an intelligent system, SVM requires training on his system. This research is studiedthe influence of training data amount given on SVM as a method of the electrical loadforecasting. Testing of system by given variable amount of data,from 12 datas to 72 datas, and SVM using the Gaussian kernel function of RBF. The relationship between the amount of training data with forecasting error value (as measured by MAPE) was analyzed by Person's correlation. This research resulted in the value of Person's correlation r = -0.96 gives conclusion that the amount of training data SVM with MAPE values obtained in the forecast SVM negative associations with the strength of relationship Key-word: forecasting, Electric Load, SVM, Train Data, Person’s CorrelationReferences
Biegel, John E., 1999, Pengendalian Produksi Suatu Pendekatan Kuantitatif, Akademika Presindo, Jakarta.
Cristianini N., Taylor J.S., 2000, An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods, Cambridge Press University, 2000
Hahn, H., Nieberg, S.M., Pickl, S., 2009 , Electric Load Forecasting Methods: Tools for Decision Making, European Journal of Operational Research 199, 902–907
Hong, W.C., 2009, Chaotic particle swarm optimization algorithm in a support vector regression electric load forecasting model, Energy Conversion and Management, 50, 105–117
Maali, Y., Al-Jumaily, A., 2013, Self-Advising Support Vector Machine, Knowledge-Based Systems 52, 214–222.
Makridakis, Spyross, Steven C., Whellwright, Victore Mcgee, Untung Sus Adriyanto, Abdul Basith, 1988, Metoda dan Aplikasi Peramalan, Jilid Satu Edisi Kedua, Erlangga.
Misbahuddin, Hasan, I., 2013, Analisis Data Penelitian dengan Statistik, Bumi Aksara, Jakarta
Vapnik, V.N., 1999, The Nature of Statistical Learning Theory 2nd edition, Springer-Verlag, New York Berlin Heidelberg
Downloads
Published
Issue
Section
License
Authors who publish with this journal agree to the following terms:
Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
Techno (Jurnal Fakultas Teknik, Universitas Muhammadiyah Purwokerto) is licensed under a Creative Commons Attribution 4.0 International License.
