The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON INFORMATION SCIENCE AND APPLICATIONS
Transactions ID Number: 29-757
Full Name: Milena Petković
Position: Assistant
Age: ON
Sex: Female
Address: Trg Dositeja Obradovića
Country: YUGOSLAVIA
Tel: +381641267703
Tel prefix: +381
Fax: +38121458873
E-mail address: milena5@uns.ac.rs
Other E-mails: milenapetrujkic@gmail.com
Title of the Paper: Electrical Energy Consumption Forecasting in Oil Refining Industry Using Support Vector Machines and Particle Swarm Optimization
Authors as they appear in the Paper: Milena Petković, Milan Rapaić, Boris Jakovljević
Email addresses of all the authors: milena5@uns.ac.rs, rapaja@uns.ac.rs, bjakov@uns.ac.rs
Number of paper pages: 10
Abstract: In this paper, Support Vector Machines (SVMs) are applied in predicting electrical energy consumption in the atmospheric distillation of oil refining at a particular oil refinery. During cross-validation process of the SVM training Particle Swarm Optimization (PSO) algorithm was utilized in selection of free SVM kernel parameters. Incorporation of PSO into SVM training process has greatly enhanced the quality of prediction. Furthermore, various (different) kernel functions were used and optimized in the process of forming the SVM models.
Keywords: Support vector machines (SVM), Kernel functions, Particle swarm optimization (PSO), Electrical energy prediction, Oil refining
EXTENSION of the file: .doc
Special (Invited) Session: Energy consumption forecasting in process industry using support vector machines and particle swarm optimization
Organizer of the Session: 617-344
How Did you learn about congress: Prof. Filip Kulic (kulic@uns.ac.rs), Prof. Zoran Jelicic (jelicic@uns.ac.rs)
IP ADDRESS: 147.91.173.31
