A PSO based NN-SVM for short-term load forecasting

Zhenyuan Xu, Junzo Watada, Jiliang Xue

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    Load forecasting has become one of the core research topics in the power system. As power load has time-variant characteristics and nonlinear characteristics, different computational intelligent techniques, neural networks (NN) in particular, are used in short-term load forecasting (STLF) to make it more effective. This study proposes a Particle Swarm Optimization (PSO) - based neural network with support vector machine (NN-SVM) model to predict the power load in short-term forecasting by using a radial-basis-function neural network (RBFNN), SVM and PSO. There are two stages in the proposed model. The first stage applies the RBFNN to predict monthly variations, and the second stage trains the SVM through hourly data to obtain the final forecast for short-term load forecasting (STLF). In the process of SVM training and NN learning, PSO is used to find the optimal parameters. The results of several experiments show that this new model performs more accurately and stably than some conventional models including RBFNN, RGA-SVM, Karman filter in STLF.

    Original languageEnglish
    Title of host publicationFrontiers in Artificial Intelligence and Applications
    PublisherIOS Press
    Pages219-227
    Number of pages9
    Volume262
    ISBN (Print)9781614994046
    DOIs
    Publication statusPublished - 2014

    Publication series

    NameFrontiers in Artificial Intelligence and Applications
    Volume262
    ISSN (Print)09226389

    Fingerprint

    Particle swarm optimization (PSO)
    Neural networks
    Support vector machines
    Experiments

    Keywords

    • particle swarm optimization (PSO)
    • radial-basis-function neural network (RBFNN)
    • short-term forecasting (STLF)
    • support vector machine (SVM)

    ASJC Scopus subject areas

    • Artificial Intelligence

    Cite this

    Xu, Z., Watada, J., & Xue, J. (2014). A PSO based NN-SVM for short-term load forecasting. In Frontiers in Artificial Intelligence and Applications (Vol. 262, pp. 219-227). (Frontiers in Artificial Intelligence and Applications; Vol. 262). IOS Press. https://doi.org/10.3233/978-1-61499-405-3-219

    A PSO based NN-SVM for short-term load forecasting. / Xu, Zhenyuan; Watada, Junzo; Xue, Jiliang.

    Frontiers in Artificial Intelligence and Applications. Vol. 262 IOS Press, 2014. p. 219-227 (Frontiers in Artificial Intelligence and Applications; Vol. 262).

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Xu, Z, Watada, J & Xue, J 2014, A PSO based NN-SVM for short-term load forecasting. in Frontiers in Artificial Intelligence and Applications. vol. 262, Frontiers in Artificial Intelligence and Applications, vol. 262, IOS Press, pp. 219-227. https://doi.org/10.3233/978-1-61499-405-3-219
    Xu Z, Watada J, Xue J. A PSO based NN-SVM for short-term load forecasting. In Frontiers in Artificial Intelligence and Applications. Vol. 262. IOS Press. 2014. p. 219-227. (Frontiers in Artificial Intelligence and Applications). https://doi.org/10.3233/978-1-61499-405-3-219
    Xu, Zhenyuan ; Watada, Junzo ; Xue, Jiliang. / A PSO based NN-SVM for short-term load forecasting. Frontiers in Artificial Intelligence and Applications. Vol. 262 IOS Press, 2014. pp. 219-227 (Frontiers in Artificial Intelligence and Applications).
    @inproceedings{4af065f5a75749acaca99daa36a8667c,
    title = "A PSO based NN-SVM for short-term load forecasting",
    abstract = "Load forecasting has become one of the core research topics in the power system. As power load has time-variant characteristics and nonlinear characteristics, different computational intelligent techniques, neural networks (NN) in particular, are used in short-term load forecasting (STLF) to make it more effective. This study proposes a Particle Swarm Optimization (PSO) - based neural network with support vector machine (NN-SVM) model to predict the power load in short-term forecasting by using a radial-basis-function neural network (RBFNN), SVM and PSO. There are two stages in the proposed model. The first stage applies the RBFNN to predict monthly variations, and the second stage trains the SVM through hourly data to obtain the final forecast for short-term load forecasting (STLF). In the process of SVM training and NN learning, PSO is used to find the optimal parameters. The results of several experiments show that this new model performs more accurately and stably than some conventional models including RBFNN, RGA-SVM, Karman filter in STLF.",
    keywords = "particle swarm optimization (PSO), radial-basis-function neural network (RBFNN), short-term forecasting (STLF), support vector machine (SVM)",
    author = "Zhenyuan Xu and Junzo Watada and Jiliang Xue",
    year = "2014",
    doi = "10.3233/978-1-61499-405-3-219",
    language = "English",
    isbn = "9781614994046",
    volume = "262",
    series = "Frontiers in Artificial Intelligence and Applications",
    publisher = "IOS Press",
    pages = "219--227",
    booktitle = "Frontiers in Artificial Intelligence and Applications",

    }

    TY - GEN

    T1 - A PSO based NN-SVM for short-term load forecasting

    AU - Xu, Zhenyuan

    AU - Watada, Junzo

    AU - Xue, Jiliang

    PY - 2014

    Y1 - 2014

    N2 - Load forecasting has become one of the core research topics in the power system. As power load has time-variant characteristics and nonlinear characteristics, different computational intelligent techniques, neural networks (NN) in particular, are used in short-term load forecasting (STLF) to make it more effective. This study proposes a Particle Swarm Optimization (PSO) - based neural network with support vector machine (NN-SVM) model to predict the power load in short-term forecasting by using a radial-basis-function neural network (RBFNN), SVM and PSO. There are two stages in the proposed model. The first stage applies the RBFNN to predict monthly variations, and the second stage trains the SVM through hourly data to obtain the final forecast for short-term load forecasting (STLF). In the process of SVM training and NN learning, PSO is used to find the optimal parameters. The results of several experiments show that this new model performs more accurately and stably than some conventional models including RBFNN, RGA-SVM, Karman filter in STLF.

    AB - Load forecasting has become one of the core research topics in the power system. As power load has time-variant characteristics and nonlinear characteristics, different computational intelligent techniques, neural networks (NN) in particular, are used in short-term load forecasting (STLF) to make it more effective. This study proposes a Particle Swarm Optimization (PSO) - based neural network with support vector machine (NN-SVM) model to predict the power load in short-term forecasting by using a radial-basis-function neural network (RBFNN), SVM and PSO. There are two stages in the proposed model. The first stage applies the RBFNN to predict monthly variations, and the second stage trains the SVM through hourly data to obtain the final forecast for short-term load forecasting (STLF). In the process of SVM training and NN learning, PSO is used to find the optimal parameters. The results of several experiments show that this new model performs more accurately and stably than some conventional models including RBFNN, RGA-SVM, Karman filter in STLF.

    KW - particle swarm optimization (PSO)

    KW - radial-basis-function neural network (RBFNN)

    KW - short-term forecasting (STLF)

    KW - support vector machine (SVM)

    UR - http://www.scopus.com/inward/record.url?scp=84902328682&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84902328682&partnerID=8YFLogxK

    U2 - 10.3233/978-1-61499-405-3-219

    DO - 10.3233/978-1-61499-405-3-219

    M3 - Conference contribution

    AN - SCOPUS:84902328682

    SN - 9781614994046

    VL - 262

    T3 - Frontiers in Artificial Intelligence and Applications

    SP - 219

    EP - 227

    BT - Frontiers in Artificial Intelligence and Applications

    PB - IOS Press

    ER -