Modelling load–settlement behaviour of piles using high-order neural network (HON-PILE model)

A. Ismail, D-S. Jeng

    Research output: Contribution to journalArticle

    22 Citations (Scopus)

    Abstract

    An accurate estimation of pile response to loading is a challenging task due to the complexity of the soil–pile interactions and uncertainties in the soil properties. Conventional methods of predicting pile load–settlement relationship either oversimplify the problem or require the parameters that are difficult to determine in the laboratory. In this study, a high-order neural network (HON) is developed to simulate the pile load–settlement curve using properties of the pile and SPT data along the depth of pile embedment as inputs. The results indicated a significant improvement in the quality of HON predictions over that of BPN, RBF and GRNN models. Based on the comparisons with the predictions of elastic and hyperbolic models, the proposed HON model provides better predictions than existing theoretical models.
    Original languageEnglish
    Pages (from-to)813-821
    Number of pages9
    JournalEngineering Applications of Artificial Intellegence
    Volume24
    Issue number5
    DOIs
    Publication statusPublished - Aug 2011

    Fingerprint

    Piles
    Neural networks
    Soils

    Keywords

    • Artificial neural network
    • Higher-order neural network
    • Soil-pile interaction
    • SPT
    • Load-settlement behaviour

    Cite this

    @article{09a4c36060974d478cd4f0a0e7156b6e,
    title = "Modelling load–settlement behaviour of piles using high-order neural network (HON-PILE model)",
    abstract = "An accurate estimation of pile response to loading is a challenging task due to the complexity of the soil–pile interactions and uncertainties in the soil properties. Conventional methods of predicting pile load–settlement relationship either oversimplify the problem or require the parameters that are difficult to determine in the laboratory. In this study, a high-order neural network (HON) is developed to simulate the pile load–settlement curve using properties of the pile and SPT data along the depth of pile embedment as inputs. The results indicated a significant improvement in the quality of HON predictions over that of BPN, RBF and GRNN models. Based on the comparisons with the predictions of elastic and hyperbolic models, the proposed HON model provides better predictions than existing theoretical models.",
    keywords = "Artificial neural network, Higher-order neural network, Soil-pile interaction, SPT, Load-settlement behaviour",
    author = "A. Ismail and D-S. Jeng",
    year = "2011",
    month = "8",
    doi = "10.1016/j.engappai.2011.02.008",
    language = "English",
    volume = "24",
    pages = "813--821",
    journal = "Engineering Applications of Artificial Intellegence",
    issn = "0952-1976",
    publisher = "Elsevier",
    number = "5",

    }

    Modelling load–settlement behaviour of piles using high-order neural network (HON-PILE model). / Ismail, A.; Jeng, D-S.

    In: Engineering Applications of Artificial Intellegence, Vol. 24, No. 5, 08.2011, p. 813-821.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Modelling load–settlement behaviour of piles using high-order neural network (HON-PILE model)

    AU - Ismail, A.

    AU - Jeng, D-S.

    PY - 2011/8

    Y1 - 2011/8

    N2 - An accurate estimation of pile response to loading is a challenging task due to the complexity of the soil–pile interactions and uncertainties in the soil properties. Conventional methods of predicting pile load–settlement relationship either oversimplify the problem or require the parameters that are difficult to determine in the laboratory. In this study, a high-order neural network (HON) is developed to simulate the pile load–settlement curve using properties of the pile and SPT data along the depth of pile embedment as inputs. The results indicated a significant improvement in the quality of HON predictions over that of BPN, RBF and GRNN models. Based on the comparisons with the predictions of elastic and hyperbolic models, the proposed HON model provides better predictions than existing theoretical models.

    AB - An accurate estimation of pile response to loading is a challenging task due to the complexity of the soil–pile interactions and uncertainties in the soil properties. Conventional methods of predicting pile load–settlement relationship either oversimplify the problem or require the parameters that are difficult to determine in the laboratory. In this study, a high-order neural network (HON) is developed to simulate the pile load–settlement curve using properties of the pile and SPT data along the depth of pile embedment as inputs. The results indicated a significant improvement in the quality of HON predictions over that of BPN, RBF and GRNN models. Based on the comparisons with the predictions of elastic and hyperbolic models, the proposed HON model provides better predictions than existing theoretical models.

    KW - Artificial neural network

    KW - Higher-order neural network

    KW - Soil-pile interaction

    KW - SPT

    KW - Load-settlement behaviour

    U2 - 10.1016/j.engappai.2011.02.008

    DO - 10.1016/j.engappai.2011.02.008

    M3 - Article

    VL - 24

    SP - 813

    EP - 821

    JO - Engineering Applications of Artificial Intellegence

    JF - Engineering Applications of Artificial Intellegence

    SN - 0952-1976

    IS - 5

    ER -