Scribd Logo
Upload

Welcome to Scribd!

  • 31 views

    GEOSNet TG3 Example 8

    Uploaded by

    greenday3
    1) The document describes the reliability analysis of a concrete dam using different system reliability methods including first order multinormal, Cornell's bound, Ditlevsen's bound, adaptive importance sampling, radius-based importance sampling, and Monte Carlo simulation. 2) Three potential failure modes are identified for the dam: sliding, overstress at the upstream heel, and overstress at the downstream toe. 3) The system reliability index calculated from the different methods ranges from 2.172 to 2.188, with the Monte Carlo simulation method estimating it at 2.188 and the adaptive importance sampling method estimating it within 0.5% error.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    GEOSNet TG3 Example 8

    Uploaded by

    greenday3
    31 views3 pages
    1) The document describes the reliability analysis of a concrete dam using different system reliability methods including first order multinormal, Cornell's bound, Ditlevsen's bound, adaptive importance sampling, radius-based importance sampling, and Monte Carlo simulation. 2) Three potential failure modes are identified for the dam: sliding, overstress at the upstream heel, and overstress at the downstream toe. 3) The system reliability index calculated from the different methods ranges from 2.172 to 2.188, with the Monte Carlo simulation method estimating it at 2.188 and the adaptive importance sampling method estimating it within 0.5% error.

    Original Title

    GEOSNet_TG3_Example_8

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    1) The document describes the reliability analysis of a concrete dam using different system reliability methods including first order multinormal, Cornell's bound, Ditlevsen's bound, adaptive importance sampling, radius-based importance sampling, and Monte Carlo simulation. 2) Three potential failure modes are identified for the dam: sliding, overstress at the upstream heel, and overstress at the downstream toe. 3) The system reliability index calculated from the different methods ranges from 2.172 to 2.188, with the Monte Carlo simulation method estimating it at 2.188 and the adaptive importance sampling method estimating it within 0.5% error.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    31 views3 pages

    GEOSNet TG3 Example 8

    Uploaded by

    greenday3
    1) The document describes the reliability analysis of a concrete dam using different system reliability methods including first order multinormal, Cornell's bound, Ditlevsen's bound, adaptive importance sampling, radius-based importance sampling, and Monte Carlo simulation. 2) Three potential failure modes are identified for the dam: sliding, overstress at the upstream heel, and overstress at the downstream toe. 3) The system reliability index calculated from the different methods ranges from 2.172 to 2.188, with the Monte Carlo simulation method estimating it at 2.188 and the adaptive importance sampling method estimating it within 0.5% error.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    Example

    No.
    Authors(s) Dianqing Li., Xiaosong Tang
    Date January 23, 2009
    For the concrete dam shown in Fig. 1, three different failure modes may occur
    as follows (Novak et al. 2001): sliding (failure mode 1), overstress at upstream
    Brief heel (failure mode 2), and overstress at downstream toe (failure mode 3). For
    description illustrative purpose, only the normal operation is considered herein. With the
    identified failure modes, performance functions can be formulated for each
    stability mode as follows.
    9.0 m
    Q
    ▽ 423.0 m (maximum water level)

    ▽ 418.0 m (normal water level)

    ▽ 406.0 m

    1 : 0.75

    Figure

    ▽ 354.0 m (tailwater level)


    ▽ 353.0 m (sediment level)

    ▽ 338.0 m
    5.0 m line of pressure relief drains
    60.0 m

    Figure 1. Main cross-section of concrete dam

    (
    g1 = 2499γ h + Q + ( 0.375 H x2 − 30α H s − 27.5 H x − 2.5 H s ) γ w f '+ )
    60c '− 0.5γ w ( H s2 − H x2 ) − 0.5γ n H n2 tan 2 (45D − θ n / 2) (1)

    g 2 = σ t + 85.7225γ h − 0.0003H s3 − 0.9168α H s + 0.0003H x3 − 0.0102 H x2 +


    Performance
    function 0.9168α H x − H x − 0.0003γ n H n3 tan 2 (45D − θ n / 2) + 0.0592Q − 10.381 (2)

    g3 = σ c + 2.4225γ n − 0.0003H s3 − 0.0832α H s + 0.0003H x3 − 0.0227 H x2 −

    0.0832α H x + H x − 0.0003γ n H n3 tan 2 (45D − θ n / 2) + 0.0258Q − 4.9644 (3)

    Table 1. Summary statistics of basic random variables in the dam stability model
    Inputs Standard
    Variable Description Mean
    deviation
    COV Distribution
    Water level for
    Hs (m) 80.00 4.80 0.06 Normal
    upstream face
    Water level for
    Hx (m) 16.00 0.96 0.06 Normal
    downstream face
    Hn (m) Sediment depth 15.00 2.25 0.15 Normal
    3 Unit weight of dam
    γh(kN/m ) 24 0.72 0.03 Normal
    concrete
    Submerged unit
    γn(kN/m3) 12 0.60 0.05 Normal
    weight of sediment
    Angle of shearing
    θn (0) resistance of 15 1.50 0.10 Extreme type I
    sediment
    Reduction coefficient
    α 0.40 0.10 0.25 Lognormal
    of uplift pressure
    2
    c′(kN/m ) Cohesion 62 21.70 0.35 Lognormal
    f′ Friction coefficient 1.00 0.30 0.30 Lognormal
    Maximum allowable
    σc(kN/m2) compressive stress of 9000 1800 0.20 Lognormal
    dam concrete
    Maximum allowable
    σt(kN/m2) tension stress of dam 1000 250 0.25 Lognormal
    concrete
    Q (kN) Vertical live load 350 122.50 0.35 Extreme type I
    The system reliability analyses are performed using the following system
    reliability methods: the first order multinormal (FOMN) (Hohenbichler and
    Rackwitz 1983), Cornell’s bound (Cornell 1967), Ditlevsen’s bound (Ditlevsen
    Solution
    methods 1979), Adaptive importance sampling (AIS) (Melchers 1989), Radius-based
    importance sampling (ISAMF) (Harbitz 1986), and Monte Carlo simulation
    (MCS).

    Table 2. System reliability indexes of concrete dam using different methods


    Solution FOMN Cornell’s Ditlevsen’s AIS ISAMF MCS
    method bound bound (confidence (n=104) (n=106)
    level=0.95)
    Results β 2.186 2.172~2.308 2.185 2.177 2.197 2.188
    Pf 0.0144 0.0105~0.01 0.0144 0.0147 0.0140 0.0143
    49
    % error in Pf 0.7 -26.6~4.2 0.7 2.8 -2.1 -

    Cornell, C. A. Bounds on the reliability of structural systems. Journal of


    Structural Division, 1967, 93(1): 171-200.
    Ditlevsen, O. (1979). Narrow reliability bounds for structural systems. Journal
    References
    of Structural Mechanics, 7(4): 453-472.
    Harbitz, A. (1986). An efficient sampling method for probability of failure
    calculation, Structural Safety, 3: 109-115.
    Hohenbichler, H., Rackwitz, R. (1983). First-order concepts in system
    reliability. Structural Safety, 1: 177-188.
    Melchers, R. E. (1989). Importance sampling in structural systems. Structural
    Safety, 6(1): 3-10.
    Novak, P., Moffat, A. L. B., Nalluri, C., et al. (2001). Hydraulic Structures,
    third edition. Taylor & Francis Group.

    You might also like