• Peer Review
  • Non-profit
  • Global Open Access
  • Green Channel for Rising Stars
Volume 3 Issue 3
Jul.  2020
Turn off MathJax
Article Contents

Xiaomeng LI. Study on the Seismic Reliability of Steel Containment[J]. SOUTHERN ENERGY CONSTRUCTION, 2016, 3(3): 79-84. doi: 10.16516/j.gedi.issn2095-8676.2016.03.017
Citation: Xiaomeng LI. Study on the Seismic Reliability of Steel Containment[J]. SOUTHERN ENERGY CONSTRUCTION, 2016, 3(3): 79-84. doi: 10.16516/j.gedi.issn2095-8676.2016.03.017

Study on the Seismic Reliability of Steel Containment

doi: 10.16516/j.gedi.issn2095-8676.2016.03.017
  • Received Date: 2016-07-22
  • Publish Date: 2020-07-17
  • In recent years, more and more high-power nuclear power plants were constructed in China to meet the increasing needs of electricity. The volume of the containments in nuclear power plants increases with the increase of the power level. Therefore, accessing the ability of steel containment to resist strong earthquakes has attracted more attention recently. So far, the most seismic analysis methods are based on deterministic methods. In fact, the parameters in the engineering projects, such as geometry, material and load value, are usually uncertain. The seismic loads aslo have a complex statistical change, which is one of the natural properties of seismic wave. Therefore, the traditional deterministic methods are difficult to verify the safety of the structure. In this article, the seismic reliability of the large steel containment was studied. The random seismic response spectrum under different probabilities was established according to the stationary stochastic model. The parameters of the stationary stochastic model were obtained from Chinese seismic code. Two methods of seismic reliability analysis were considered. One is under deterministic seismic intensity, the other is under the certainty probabilities. Finally, the sensitivity parameters were calculated based on the Monte Carlo method. This article will provide some references for the seismic reliability design of the steel containment.
  • [1] 孙广俊,李鸿晶. 平稳随机地震地面运动过程模型及其统计特征[J]. 地震工程与工程振动,2004, 6(24): 21-26.

    SUN G J, LI H J. Stationary models of random earthquake ground motion and their statistical properties[J]. Earthquake Engineering and Engineering Vibration, 2004, 6(24): 21-26.
    [2] Housner G W. Characteristics of strong motion earthquakes[J]. Bull. of Seism. Soc. Of Am., 1947, 37(1): 19-31.
    [3] Tajimi H. A statistical method of determining the maximum response of a building structure during an earthquake [A]. 2rd ed. WCEE , Tokyo, Japan, 1960.
    [4] 欧进萍,牛荻涛,杜修力. 设计用随机地震动的模型及其参数确定[J]. 地震工程与工程振动,1991, 11(3): 45-54.

    OU J P, NIU D T, DU X L. Random earthquake ground motion model and its parameter determination used in a seismic design[J]. Earthquake Engineering and Engineering Vibration, 1991, 11(3): 45-54.
    [5] 杜修力,胡晓,陈原群. 强震地运动随机过程模拟[J]. 地震学报,1995, 17(1): 103-109.

    DU X L, HU X, CHEN Y Q. The simulation of random earthquake ground motion[J]. ACTA Seismologica Sinica, 1995, 17(1): 103-109.
    [6] GB 18306-2015,中国地震动参数区划图 [S].
    [7] 欧进萍,刘会仪. 基于随机地震动模型的结构随机地震反应谱及其应用[J]. 地震工程与工程振动,1994, 14(1): 14-22.

    OU J P, LIU H Y. Random seismic response spectrum and its application based on the random seismic ground motion[J]. Earthquake engineering and engineering vibration, 1994, 14(1): 14-22.
    [8] 洪峰,江近仁,李玉亭. 地震地面运动的功率谱模型及其参数的确定[J]. 地震工程与工程振动,1994, 14(2): 46-52.

    HONG F, JIANG J R, LI Y T. Determination of the model of earthquake ground motion power spectrum and its parameters[J]. Earthquake Engineering and Engineering Vibration, 1994, 14(2): 46-52.
    [9] 王亚勇,李虹. 考虑场地特征的强震地面运动参数的统计分析[J].地震工程与工程振动,1986,6(3): 11-19.

    WANG Y Y, LI H. Statistical analysis of the seismic ground motion parameters considering with the site characterization[J]. Earthquake engineering and engineering vibration, 1986, 6(3): 11-19.
    [10] 刘会仪. 结构随机地震反应谱理论与应用 [D]. 哈尔滨:哈尔滨建筑工程学院,1993.
    [11] 叶勇,郝艳华,张昌汉. 基于ANSYS的结构可靠性分析[J]. 机械工程与自动化,2004,12(6): 63-66.

    YE Y, HAO Y H, ZHANG C H. Struture reliability analysis in ANSYS[J]. Mechanical Engineering & Automation, 2004, 12(6): 63-66.
    [12] GB 50267-1997.核电厂抗震设计规范 [S].
  • 通讯作者: 陈斌, bchen63@163.com
    • 1. 

      沈阳化工大学材料科学与工程学院 沈阳 110142

    1. 本站搜索
    2. 百度学术搜索
    3. 万方数据库搜索
    4. CNKI搜索

Figures(10)  / Tables(6)

Article Metrics

Article views(157) PDF downloads(12) Cited by()

Related

Study on the Seismic Reliability of Steel Containment

doi: 10.16516/j.gedi.issn2095-8676.2016.03.017

Abstract: In recent years, more and more high-power nuclear power plants were constructed in China to meet the increasing needs of electricity. The volume of the containments in nuclear power plants increases with the increase of the power level. Therefore, accessing the ability of steel containment to resist strong earthquakes has attracted more attention recently. So far, the most seismic analysis methods are based on deterministic methods. In fact, the parameters in the engineering projects, such as geometry, material and load value, are usually uncertain. The seismic loads aslo have a complex statistical change, which is one of the natural properties of seismic wave. Therefore, the traditional deterministic methods are difficult to verify the safety of the structure. In this article, the seismic reliability of the large steel containment was studied. The random seismic response spectrum under different probabilities was established according to the stationary stochastic model. The parameters of the stationary stochastic model were obtained from Chinese seismic code. Two methods of seismic reliability analysis were considered. One is under deterministic seismic intensity, the other is under the certainty probabilities. Finally, the sensitivity parameters were calculated based on the Monte Carlo method. This article will provide some references for the seismic reliability design of the steel containment.

Xiaomeng LI. Study on the Seismic Reliability of Steel Containment[J]. SOUTHERN ENERGY CONSTRUCTION, 2016, 3(3): 79-84. doi: 10.16516/j.gedi.issn2095-8676.2016.03.017
Citation: Xiaomeng LI. Study on the Seismic Reliability of Steel Containment[J]. SOUTHERN ENERGY CONSTRUCTION, 2016, 3(3): 79-84. doi: 10.16516/j.gedi.issn2095-8676.2016.03.017
  • 压水堆核电厂以往大多采用混凝土安全壳,而且只承担安全屏障功能。在第三代核电厂AP1000设计中增加了一层钢制安全壳,它既承担安全屏障功能,又是非能动冷却系统的重要载出热量环节。然而,目前国内对于这种安全壳的设计以及扩容后的设计经验还很缺乏,还需要投入大量的精力进行研究。

    另外,对于核电站设备的抗震分析大多基于确定论方法,而工程实践中,不可避免的会出现各种随机因素,如结构的物理参数、几何尺寸的随机性,载荷的随机性等。显然,常规的确定性结构在确定性载荷作用下的分析模型和方法已经不适用[1]。为了向核电站设计、施工单位提供更为可信、真实的数据,必须借助基于概率论的可靠性分析方法对核电站关键设备进行研究。

    本文基于随机地震动模型,利用现行抗震规范获得部分模型参数,建立了随机地震反应谱,并应用于大容积钢制安全壳的抗震可靠性分析中。同时还利用蒙特卡洛模拟法对钢制安全壳进行了可靠性及敏感性计算。

  • 目前,一般将地震动加速度过程视为平稳随机过程,则可用功率谱函数来描述结构随机地震反应计算的输入模型。可通过假定功率谱模型,再根据地震动参数确定模型中待定参数的方法将地震动基本参数转换成相应的功率谱[2]

    自1947年Housner[3]首次提出地震动加速度过程的白噪声模型至今,国内外学者在白噪声模型的基础上已提出了多种地震动模型。如过滤白噪声模型即金井清谱[3]、过滤有色白噪声模型[4]等。本文选取另外一种杜修力模型[5],这种模型考虑震源机制,将地震学中低频模拟方法和工程学中高频模拟方法相结合,其功率谱密度为:

    ((1))

    式中:S0为基岩地震波的功率谱密度,大小要根据地震强度来定;ωg为场地土卓越频率;ζg为场地土阻尼比;D为反映基岩特性的谱参数,一般取0.04 s;ω0为低频拐角频率,ω0=2π/TrTr为断层的破裂持时,Tr与震级M的统计关系:

    ((2))

    式中:d1=-1.325,d2=0.353。该模型综合了两种模型的优点,而且反映了人们对震源机制的理解。

  • 根据《中国地震动参数区划图》(GB 18306-2015)[6],可以给出考虑设计特征周期分区的各类场地特征卓越频率和阻尼比如表1所示。

    场地土类型 I基岩 II中硬 III中软 IV软弱
    ω g 1区 25.13 17.95 13.96 9.67
    2区 20.94 15.71 11.42 8.38
    3区 17.95 13.96 9.67 6.98
    ζg / 0.64 0.72 0.80 0.90

    Table 1.  Dynamic characteristics of site soil

  • 地震持时不仅对结构的累积破坏有很大的影响,而且与地震反应的最大值分布密切相关。根据国内外大量强震记录进行统计回归分析,与震级、震中距和场地类别有关的按90%能量持时定义的地震持时Ts的回归公式见式(3)[7]

    ((3))

    式中:M是震级、R是震中距,Tg是场地土卓越周期,a1a4是回归系数,值如表2所示.

    回归系数 a1 a2 a3 a4
    水平方向 -1.555 0.165 0.831 0.148
    竖直方向 -1.340 0.104 0.982 0.184

    Table 2.  Regression coefficients

  • 谱强度因子S0与地面最大加速度均值之间的关系为[8]:

    ((4))
    ((5))
    ((6))
    ((7))

    式中:r为峰值因子;Td为固有周期;Am为地面最大加速度均值,取中国地震动参数区划图中的值。

  • 根据反应谱定义,单质点振子在地震作用下的最大反应为[9]

    ((8))

    式中:ζT分别是振子阻尼比和固有周期;Ts是地震持续时间。基于随机振动理论和泊松近似,在持续时间τ内,谱值umax (Tζ)不超过u的概率分布函数表达式为:

    ((9))

    最大反应的均值和标准差分别为:

    ((10))
    ((11))

    式中:v是反应过程y(t)向上和向下的期望越零率;ωT分别为系统的固有圆频率和周期;σy是随机地震反应过程u(tTζ)的均方差。

    ((12))
  • 根据上面umax (Tζ)的概率分布函数公式(8)可得到最大反应发生概率为P的概率地震反应谱 [10]

    (#(13))

    式中:γP (T)称为概率峰值因子,按下式计算:

    ((14))

    随机地震反应谱的均值成为平均地震反应谱,其计算式为:

    ((15))

    图1为基本加速度为0.3 g的各类场地第2特征周期分组[6]的平均反应谱和15%、50%、85%的概率反应谱。从图中可以看出,平均反应谱略大于50%的概率反应谱,但两者十分接近。据国内外大量试验研究[10],平均反应谱大概是57%的概率反应谱。而《核电厂抗震设计规范》中的反应谱[12]的概率保证在局部区段低于保证概率为50%的反应谱,说明低估了地震对结构的影响。

    Figure 1.  Average acceleration response spectrum comparison with response spectrum of different probabilities

  • 文章以大容积钢制安全壳为研究对象,建立三维模型。采用三维壳单元Shell181来模拟安全壳的封头和圆筒段部分;三维实体单元Solid185来模拟钢制安全壳的环吊梁及加强肋部分。壳单元与实体单元通过MEARGE操作实现共结点。模型共划分21 051个单元,13 117个结点。图2为钢制安全壳有限元模型。

    Figure 2.  Finite element model of steel containment vessel

    通过定义模型的表面附加质量来模拟加强肋、环吊梁、空气导流板、通道管线、闸门等结构的重量,具体附件质量及位置如表3所示。模型边界条件为底部固定约束。

    设备 标高/m 单位附加质量/(kg·m-2)
    空气导板 ∇46~83 96.529
    通道线路 ∇50~53 301.620
    混凝土加强肋 ∇40 172.855
    设备闸门 ∇43 503.992
    人员闸门 ∇42 601.924

    Table 3.  Additional mass

    阶数 频率/Hz 周期/s 振型参与系数 比率
    1 4.734 0.211 1 644.9 1.000
    2 4.735 0.212 -822.55 0.500
    3 5.331 0.188 -10.472 0.006
    4 5.334 0.187 -1.074 0.000
    5 5.417 0.185 -9.136 0.006
    6 5.420 0.184 -2.845 0.002
    7 5.433 0.183 5.283 0.003
    8 5.435 0.182 -8.076 0.005
    9 5.675 0.176 -0.155 0.000
    10 5.677 0.175 4.570 0.003

    Table 4.  Structural natural frequencies and mode participation coefficient

    随机地震反应谱 P=15% P=85% P=50% 平均反应谱 规范反应谱
    结构位移/mm 14.6 18.8 16.5 16.8 16.9
    Mises应力/MPa 62.1 79.8 69.9 71.2 71.6

    Table 5.  The comparison of the calculation

  • 对钢制安全壳进行不同发生保证概率的反应谱计算,即基于确定性地震强度下的条件抗震可靠度分析。图3图7为在不同保证概率的反应谱下钢制安全壳的位移及Mises应力云图。

    Figure 3.  The seismic response of the steel containment with the average response spectrum

    Figure 4.  The seismic response of the steel containment with the P=50% response spectrum

    Figure 5.  The seismic response of the steel containment with standard response spectrum

    Figure 6.  The seismic response of the steel containment with the P=15% response spectrum

    Figure 7.  The seismic response of the steel containment with the P=85% response spectrum

    结果显示,结构抗震分析后的位移、应力图的分布都基本相同。最大位移出现在上封头处,最大应力出现在上封头顶部。另外采用规范反应谱、平均反应谱及保证概率为50%的反应谱计算出的结构位移和应力较为接近,但规范反应谱的计算结果略大于平均反应谱结果,而保证概率为50%的反应谱计算结果为三者中最小。而用保证概率为15%和85%两种反应谱计算出的结果则与其他三种相差较多,保证概率越大,结构的抗震响应越大。因此,对核电厂安全壳的抗震分析必须针对具体的厂址、统计搜集当地的地震历史,充分考虑当地的地震频率,这样才能建立真实的反应谱,得出可靠性高的结果。

  • 本章基于一定期限对结构的抗震性能进行可靠度分析,即考察一定期限内所有地震发生的可能性进行全概率分析。将结构几何、材料、地震特征参数等作为随机输入变量,定义最大位移及最大Mises应力为状态变量,结构总重量为随机输出变量,借助有限元软件的蒙特卡洛模拟进行抗震可靠性分析。

    分析的主要步骤有[11]:(1)创建PDS分析文件,即仿真循环mac文件;(2)运行分析文件,执行初始化的分析过程;(3)定义输入输出参数及其概率分布;(4)选择抽样方法,确定抽样点数;(5)执行概率设计循环仿真计算;(6)敏感性参数判断。

  • 通过观察随机参数的抽样过程及样本值的均值、标准差等,可以看出程序基本按照所假定的参数概率分布进行抽样,样本均值、标准差逐步趋向平稳,说明抽样次数足够,如图8图9所示。

    Figure 8.  Probability distribution of the random input parameters

    Figure 9.  Standard deviation of state variables and output parameters

    通过表6可以看出不管是最大位移还是最大Mises应力值均没有超过设计范围。因此在假设的随机变量分布及平均地震反应谱作用下,根据可靠性计算,钢制安全壳最大位移小于18 mm的概率是99.4%,最大Mises应力小于60 MPa的概率是99.7%,结构是可靠的。

    输出随机变量 最大值 最小值 均值
    DMAX/mm 38.3 1.18 17.1
    SMAX/MPa 167.6 0.005 71.8
    WT/106kg 7.57 7.17 7.39

    Table 6.  The results of the reliability analysis

    对钢制安全壳的16个输入随机变量进行敏感性分析,得到的结果如图10所示:

    Figure 10.  Sensitivity parameter identification chart

    可以看出对结构最大位移量敏感的参数有5个,其中影响最大的是地震谱强度因子。对结构最大Mises应力敏感的参数有2个,分别是地震谱强度因子以及人员操作平台的附加载荷。对结构总体质量敏感的参数分别为圆柱段壁厚等。

    通过上述敏感性分析可以看出多种随机参数对输出随机变量均有不同程度的影响,其中地震谱强度因子对结构抗震可靠性影响最大。除了谱强度因子(地震强度)的影响外,结构尺寸参数、附加质量的影响也较大,因此可以对结构进行优化设计,提高可靠性。

  • 本文主要以大容积钢制安全壳为研究背景,分别从基于确定性地震载荷及基于确定性保证概率两方面出发,对其抗震可靠性进行了研究。获得了具有不同发生保证概率的随机地震反应谱,并利用有限元软件的蒙特卡洛模拟方法进行了敏感性参数计算。本文从条件概率以及全概率的角度所提出的两种可靠性分析方法,可为今后我国的钢制安全壳抗震可靠性设计提供参考。

Reference (12)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return