Building an integrated hybrid model for short-term and mid-term load forecasting with genetic optimization

Jiliang Xue, Zhenyuan Xu, Junzo Watada

    Research output: Contribution to journalArticle

    8 Citations (Scopus)

    Abstract

    The enhancement of load forecasting has become one of the core research topics in the energy field. Because power load has both time-variant and nonlinear characteristics, different types of methods, neural networks (NN) in particular, have been applied to power load forecasting. This study proposes a real-valued genetic algorithm (RGA)- based neural network with support vector machine (NN-SVM) model to predict the power load in both short-term and mid-term forecasting by using a radial-basis-function neural network (RBFNN), SVM and RGA. The model consists of two stages. In short-term load forecasting (STLF), 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. Similar operations are used in mid-term load forecasting (MTLF). In the process of SVM training and NN learning, RGA 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. Also it is able to function well in MTLF.

    Original languageEnglish
    Pages (from-to)7381-7391
    Number of pages11
    JournalInternational Journal of Innovative Computing, Information and Control
    Volume8
    Issue number10 B
    Publication statusPublished - 2012 Oct

    Fingerprint

    Load Forecasting
    Integrated Model
    Hybrid Model
    Radial Basis Function Neural Network
    Genetic Algorithm
    Short-term Load Forecasting
    Optimization
    Neural networks
    Neural Networks
    Genetic algorithms
    Predict
    Optimal Parameter
    Model
    Forecast
    Forecasting
    Support Vector Machine
    Enhancement
    Filter
    Support vector machines
    Energy

    Keywords

    • Mid- term load forecasting (MTLF)
    • Radial-basis-function neural network (RBFNN)
    • Real-valued genetic algorithm (RGA)
    • Short-term load forecasting (STLF)
    • Support vector machine (SVM)

    ASJC Scopus subject areas

    • Computational Theory and Mathematics
    • Information Systems
    • Software
    • Theoretical Computer Science

    Cite this

    Building an integrated hybrid model for short-term and mid-term load forecasting with genetic optimization. / Xue, Jiliang; Xu, Zhenyuan; Watada, Junzo.

    In: International Journal of Innovative Computing, Information and Control, Vol. 8, No. 10 B, 10.2012, p. 7381-7391.

    Research output: Contribution to journalArticle

    @article{a1357a20e17a49c59e08f1361738c879,
    title = "Building an integrated hybrid model for short-term and mid-term load forecasting with genetic optimization",
    abstract = "The enhancement of load forecasting has become one of the core research topics in the energy field. Because power load has both time-variant and nonlinear characteristics, different types of methods, neural networks (NN) in particular, have been applied to power load forecasting. This study proposes a real-valued genetic algorithm (RGA)- based neural network with support vector machine (NN-SVM) model to predict the power load in both short-term and mid-term forecasting by using a radial-basis-function neural network (RBFNN), SVM and RGA. The model consists of two stages. In short-term load forecasting (STLF), 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. Similar operations are used in mid-term load forecasting (MTLF). In the process of SVM training and NN learning, RGA 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. Also it is able to function well in MTLF.",
    keywords = "Mid- term load forecasting (MTLF), Radial-basis-function neural network (RBFNN), Real-valued genetic algorithm (RGA), Short-term load forecasting (STLF), Support vector machine (SVM)",
    author = "Jiliang Xue and Zhenyuan Xu and Junzo Watada",
    year = "2012",
    month = "10",
    language = "English",
    volume = "8",
    pages = "7381--7391",
    journal = "International Journal of Innovative Computing, Information and Control",
    issn = "1349-4198",
    publisher = "IJICIC Editorial Office",
    number = "10 B",

    }

    TY - JOUR

    T1 - Building an integrated hybrid model for short-term and mid-term load forecasting with genetic optimization

    AU - Xue, Jiliang

    AU - Xu, Zhenyuan

    AU - Watada, Junzo

    PY - 2012/10

    Y1 - 2012/10

    N2 - The enhancement of load forecasting has become one of the core research topics in the energy field. Because power load has both time-variant and nonlinear characteristics, different types of methods, neural networks (NN) in particular, have been applied to power load forecasting. This study proposes a real-valued genetic algorithm (RGA)- based neural network with support vector machine (NN-SVM) model to predict the power load in both short-term and mid-term forecasting by using a radial-basis-function neural network (RBFNN), SVM and RGA. The model consists of two stages. In short-term load forecasting (STLF), 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. Similar operations are used in mid-term load forecasting (MTLF). In the process of SVM training and NN learning, RGA 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. Also it is able to function well in MTLF.

    AB - The enhancement of load forecasting has become one of the core research topics in the energy field. Because power load has both time-variant and nonlinear characteristics, different types of methods, neural networks (NN) in particular, have been applied to power load forecasting. This study proposes a real-valued genetic algorithm (RGA)- based neural network with support vector machine (NN-SVM) model to predict the power load in both short-term and mid-term forecasting by using a radial-basis-function neural network (RBFNN), SVM and RGA. The model consists of two stages. In short-term load forecasting (STLF), 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. Similar operations are used in mid-term load forecasting (MTLF). In the process of SVM training and NN learning, RGA 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. Also it is able to function well in MTLF.

    KW - Mid- term load forecasting (MTLF)

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

    KW - Real-valued genetic algorithm (RGA)

    KW - Short-term load forecasting (STLF)

    KW - Support vector machine (SVM)

    UR - http://www.scopus.com/inward/record.url?scp=84866051916&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84866051916&partnerID=8YFLogxK

    M3 - Article

    AN - SCOPUS:84866051916

    VL - 8

    SP - 7381

    EP - 7391

    JO - International Journal of Innovative Computing, Information and Control

    JF - International Journal of Innovative Computing, Information and Control

    SN - 1349-4198

    IS - 10 B

    ER -