Study on short-term network forecasting based on SVM-MFA algorithm
| Autor: | Wen Ma, Xin Yong, Xinyang Zhang, Shenzhang Li |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Mathematical optimization
Fitness function Artificial neural network Computer science Stability (learning theory) 020207 software engineering 02 engineering and technology Power (physics) Support vector machine Mean absolute percentage error Rate of convergence Signal Processing 0202 electrical engineering electronic engineering information engineering Media Technology 020201 artificial intelligence & image processing Firefly algorithm Computer Vision and Pattern Recognition Electrical and Electronic Engineering |
| Zdroj: | Journal of Visual Communication and Image Representation. 65:102646 |
| ISSN: | 1047-3203 |
| DOI: | 10.1016/j.jvcir.2019.102646 |
| Popis: | Accurate prediction of power supply load is vital in power industry, which provides economic operation decision for the power operation department. For the unpredictability and periodicity of power load, nonlinear intelligent forecasting method is adopted. A modified firefly algorithm (MFA) combined with support vector machine (SVM) is proposed to predict the load of power supply data in this paper. The nonlinear mapping function is used to deal with the nonlinear regression problem in SVM, in which the parameters affect the accuracy of load prediction, so the MFA method is adopted to optimized the parameters of SVM. In order to verify the accuracy of SVM-MFA, mean absolute percentage error (eMAPE) was used as the fitness function for simulation and comparison experiment. The results show that the SVM-MFA proposed in this paper has stronger global search ability and faster convergence rate than the traditional artificial neural network, and it is verified that the method proposed in this paper has higher accuracy and higher stability of network load prediction. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect