Medium term load forecasting using evolutionary programming-least square support vector machine

• Published in: ARPN Journal of Engineering and Applied Sciences
• Main Author: Yasin Z.M.; Zakaria Z.; Razak M.A.A.; Aziz N.F.A.
• Format: Article
• Language: English
• Published: Asian Research Publishing Network 2015
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-84949953826&partnerID=40&md5=20496159e2f14bee2ab3bcb3d96e3fbb

This paper presents new intelligent-based technique namely Evolutionary Programming- Least-Square Support Vector Machine (EP-LSSVM) to forecast a medium term load demand. Medium-term electricity load forecasting is a difficult work since the accuracy of forecasting is influenced by many unpredicted factors whose relationships are commonly complex, implicit and nonlinear. Available historical load data are analyzed and appropriate features are selected for the model. Load demand in the year 2008 until 2010 are used for features in combination with day in months and hour in days. There are 3 inputs vectors for this proposed model consists of day, month and year. As for the output, there are 24 outputs vectors for this model which represents the number of hour in a day. In EP-LSSVM, the Radial Basis Function (RBF) Kernel parameters are optimally selected using Evolutionary Programming (EP) optimization technique for accurate prediction. The performance of EP-LSSVM is compared with those obtained from LS-SVM using crossvalidation technique in terms of accuracy. The experimental results show that the proposed approach gives better performance in terms of Mean Absolute Percentage Error (MAPE) and coefficients of determination (R2) for the entire period of prediction. © 2006-2015 Asian Research Publishing Network (ARPN).