en
×

分享给微信好友或者朋友圈

使用微信“扫一扫”功能。
通讯作者:

王神龙,E-mail:shenlongwang@usst.edu.cn

中图分类号:O324;O322

文献标识码:A

文章编号:1672-6553-2023-21(5)-035-009

DOI:10.6052/1672-6553-2022-044

参考文献 1
LLOYD A.Seakeeping:ship behavior in rough weather [M].Horwood,1989.
参考文献 2
SHINOZUKA M,DEODATIS G.Simulation of stochastic processes by spectral representation [J].Applied Mechanics Reviews,1991,44(4):191-203.
参考文献 3
KIM D S,LEE W K.Analysis of random ship rolling using partial stochastic linearization [J].Journal of the Society of Naval Architects of Korea,1995.
参考文献 4
GU J Y.Probability analysis of ship nonlinear roll-motion excited by white noise [J].Journal of Ship Mechanics,2006,10(3):17-25.
参考文献 5
CHAI W,NAESS A,LEIRA B J.Stochastic nonlinear ship rolling in random beam seas by the path integration method[J].Probabilistic Engineering Mechanics,2016,44(apr):43-52.
参考文献 6
刘坤峰,靳艳飞.相关白噪声激励下双稳态 Duffing Van der Pol 系统的随机分岔[J].动力学与控制学报,2020,18(4):12-18.LIU K F,JIN Y F.Stochastic bifurcation in bistable Duffing Van der Pol system driven by correlated white noises[J].Journal of Dynamics and Control,2020,18(4):12-18.(in Chinese)
参考文献 7
COTTONE G,MD PAOLA,IBRAHIM R,et al.Stochastic ship roll motion via path integral method[J].International Journal of Naval Architecture and Ocean Engineering,2010,2(3):119-126.
参考文献 8
VAIPAN I U.Probability density function solution to nonlinear ship roll motion excited by external Poisson white noise [J].Science China Technological Sciences,2011,(5):1121-1125.
参考文献 9
MAMIS K I,ATHANASSOULIS G A,PAPADOPOULOS K E.Generalized FPK equations corresponding to systems of nonlinear random differential equations excited by colored noise,revisitation and new directions [J].Procedia Computer Science,2018,136:164-173.
参考文献 10
ATHANASSOULIS G A,KAPELONIS Z G,MAMIS K.Numerical solution of generalized fpk equations corresponding to random differential equations under colored noise excitation.The transient case [C].8th Conference on Computational Stochastic Mechanics(CSM8),2018.
参考文献 11
MAMIS K I,ATHANASSOULIS G A,KAPELONIS Z G.A systematic path to non-markovian dynamics:new response pdf evolution equations under Gaussian coloured noise excitation [J].The Royal Society,2018,(475):1-18.
参考文献 12
FALZARANO J M,SU Z,JAMNONGPIPATKUL A,et al.Solving the problem of nonlinear ship roll motion using stochastic dynamics:risk of capsizing [M].2019.
参考文献 13
ZHU W Q,CAI G Q.On bounded stochastic processes [M].Springer New York,2013:3-24.
参考文献 14
朱位秋.非线性随机动力学与控制-Hamilton理论体系框架 [M].北京:科学出版社,2003.ZHU W Q.Nonlinear stochastic dynamics and control-Hamilton theoretical framework [M].Beijing:Science Press,2003.(in Chinese)
参考文献 15
ROBERTS J B.A stochastic theory for nonlinear ship rolling in irregular seas[J].Journal of Ship Research,1982,26:229-245.
参考文献 16
张强,王剑龙,李扬,等.有界噪声激励下系统随机共振的改善 [J].动力学与控制学报,2021,19(6):52-58.ZHANG Q,WANG J L,LI Y,et al.Improvement of stochastic resonance under bounded noise excitation [J].Journal of Dynamics and Control,2021,19(6):52-58.(in Chinese)
参考文献 17
ROBERTS J B,VASTA M.Markov modeling and stochastic identification for nonlinear ship rolling in random waves [J].Philosophical Transactions of the Royal Society of London A,2000,358:1917-1941.
参考文献 18
WANG Y G,HUANG Z L,TAN J H.First passage probability of nonlinear ship rolling in random seas [J].Journal of Ship Mechanics,2008,12:870-879.
参考文献 19
WANG S L,JIN X L,HUANG Z L.Analytical evaluation of first passage failure for nonlinear ship rolling excited by random wave [J].Journal of Vibration Engineering & Technologies,2016,4(4):291-299.
参考文献 20
WANG L Y,TANG Y G,YAN L I,et al.Studies on stochastic parametric roll of ship with stochastic averaging method [J].China Ocean Engineering,2020,34(2):289-298.
参考文献 21
CVETICANIN L.Strongly nonlinear oscillators [M].Analytical Solutions,Springer,2014.
参考文献 22
TIEN W M,NAMACHCHIVAYA N S,COPPOLA V T.Stochastic averaging using elliptic functions to study nonlinear stochastic systems [J].Nonlinear Dynamic,1993,4:373-387.
参考文献 23
COPPOLA V T,RAND R H.Averaging using elliptic functions:approximation of limit cycles [J].Acta Mechanica,1990,81(3-4):125-142.
参考文献 24
OKABE T,KONDOU T,OHNISHI J.Elliptic averaging methods using the sum of Jacobian elliptic delta and zeta functions as the generating solution[J].International Journal of Non-Linear Mechanics,2011,46(1):159-169.
参考文献 25
OKABE T,KONDOU T.Improvement to the averaging method using the Jacobian elliptic function[J].Journal of Sound & Vibration,2009,320(1-2):339-364.
参考文献 26
RAKARIC Z,KOVACIC I.An elliptic averaging method for harmonically excited oscillators with a purely non-linear non-negative real-power restoring force[J].Communications in Nonlinear Science and Numerical Simulation,2013,18(7):1888-1901.
参考文献 27
ELÍAS ZÙÑIGA A.Exact solution of the cubic-quintic Duffing oscillator[J].Applied Mathematical Modelling,2013,37(4):2574-2579.
参考文献 28
徐文俊,郑丽文,马品奎.一种改进的基于Jacobi椭圆函数的随机平均法[J].振动工程学报,2019,032(003):444-451.XU W J,ZHENG L W,MA P K.An improved stochastic averaging method based on Jacobian elliptic function [J].Journal of Vibration Engineering,2019,032(003):444-451.(in Chinese)
参考文献 29
WANG S L,XIE Z,ZHONG R,et al.Stochastic analysis of a predator-prey model with modified Leslie-Gower and Holling type Ⅱ schemes [J].Nonlinear Dynamic,2020,101:1245-1262.
参考文献 30
TRIPATHI J P,MEGHWANI S S,THAKUR M,et al.A modified Leslie-Gower predator-prey interaction model and parameter identifiability [J].Communications in Nonlinear Science & Numerical Simulation,2017,54:331-346.
参考文献 31
SOURAV,KUMAR,SASMAL,et al.Dynamics of a predator-prey system with fear and group defense-ScienceDirect [J].Journal of Mathematical Analysis & Applications,2020,481(1):123471.
参考文献 32
WANG S L,WANG Z,XU C,et al.Sensitivity analysis and stationary probability distributions of a stochastic two-prey one-predator model[J].Applied Mathematics Letters,2020,116:106996.
参考文献 33
朱位秋.随机振动[M].北京;科学出版社,1998.ZHU W Q.Random vibration[M].Beijing:Science Press,1998.(in Chinese)
目录contents

    摘要

    本文提出了一种改进的基于Jacobi椭圆函数的随机平均法,用于求解有界噪声激励下船舶横摇系统的随机响应问题.对谐波及有界噪声激励下的船舶横摇模型进行了局部灵敏度分析,研究了系统参数对船舶横摇角的影响,为之后随机分析奠定了基础.通过引入Jacobi椭圆函数,导出随机微分方程,应用随机平均法得到了关于横摇运动幅值和相位的伊藤方程,进一步求解相应的Fokker-Planck-Kolmogorov (FPK)方程,得到幅值概率密度.将该方法应用于有界噪声激励下船舶横摇系统,研究了系统的随机响应特性,通过与蒙特卡洛数值模拟结果比较验证了该方法的可行性与准确性.该方法所得的随机响应结果与灵敏度分析结果一致:激励幅值E,外激频率Ω,线性刚度系数γ对船舶横摇系统产生较大的影响,其余参数不影响船舶横摇运动的振幅,只对概率峰值的高度有影响.

    Abstract

    In this paper, a stochastic averaging method based on elliptic function is proposed to evaluate the probabilistic solution of ship roll motion excited by bounded noises. We carry out local sensitivity analysis on the rolling model to analyze the influence of main parameters, thus laying a foundation for the random analysis. By introducing the Jacobi elliptic functions (JEFs), we acquire the stochastic differential equation and the It equation regarding the amplitude and phase of the roll motion. Then, we obtain the probabilistic solutions of the rolling system by solving the corresponding Fokker-Planck-Kolmogorov (FPK) equation. Then apply this method to investigate the probabilistic solution of ship’s nonlinear rolling motion under bounded noise excitation, which shows perfect agreement with the sensitivity quantification. Finally, the feasibility and accuracy of the proposed technique are verified by the Monte Carlo simulations.

  • 引言

  • 船舶作为海洋运输最重要的载体,其运载功能的保障及航行的安全历来受到高度重视.船舶在风浪中航行,其运动状态非常复杂,其中,风浪引起的横摇运动严重威胁着船舶和乘员的安全,是船舶抗倾覆设计和耐波性研究中的热点和难点课题[1].由于船舶所受载荷的随机性,近些年来,对其横摇运动的研究大多集中在随机激励下船舶动态行为预测和响应分析.

  • 对于随机风浪激励,比较普遍的处理方法是将其简化为高斯白噪声[2-6].然而,由于实际噪声功率谱不具有无限带宽,高斯白噪声仅为一种理想化模型.泊松噪声是一种比较重要的非高斯白噪声,通常用来描述瞬间的地震以及旋涡阵风等实际载荷.Cottone等[7]用泊松白噪声来描述随机脉冲载荷,研究了船舶横摇振动的响应问题; Vaipan等[8]分析了泊松白噪声激励下船舶非线性横摇运动的稳态概率密度解.实际上,真实的随机波浪是具有有限谱宽的窄带随机过程,而有界噪声是具有有理谱密度的有色噪声,其谱宽为有限值,因此考虑用有界噪声来模拟随机波浪载荷更为贴近.在模拟有界噪声的过程中,概率分布和功率谱密度是最关键的特性.近年来,对于任意给定的(测量的)频谱,发展了一种基于广义Fokker-Planck-Kolmogorov(FPK)方程来构造有界噪声,该方法可以构造平滑相关的噪声[9-11].Falzarano等[12]提出了一种滤波方法,可用于平滑相关的激励噪声建模.此外,有学者提出用谐波函数来模拟有界噪声的过程,Zhu等[13-15]提出了一种基于随机调和模型描述有界的随机过程,研究了模型的概率分布和功率谱密度,并提出了随机过程的系统参数选择方案.在本文中,有界噪声激励模型即该模型.

  • 在船舶非线性横摇的研究方法中,随机平均法是较为有效的一种方法. Roberts[16]基于能量包线的随机平均法研究了非线性船舶横摇运动,将系统的能量表示为一维马尔科夫过程,通过求解相应的FPK方程得到系统的稳态概率密度,之后基于标准随机平均法研究了非线性船舶横摇系统的稳态响应问题[17].Wang等[18]将功率谱密度视为指数型,基于广义谐和函数随机平均法研究了随机波浪作用下船舶非线性横摇的稳定性问题.Wang等[19]基于广义谐和函数随机平均法研究了非线性船舶横摇运动在随机激励下的响应和稳定性问题; Wang和Tang等[20]基于广义谐和函数的随机平均法研究了参激横摇船舶模型.

  • 然而,与广义谐和函数相比,椭圆函数是非线性系统的精确解[21].Tien等[22]最早提出了基于椭圆函数的随机平均法,Coppola 和Rand[23]将该方法进一步扩展到确定性系统,此后,确定性椭圆函数得到进一步研究与发展[24-26].Elías等[27]应用Jacobi椭圆函数求解一类多阶杜芬振子问题,该模型可应用于船舶横摇系统.作者此前引入基于椭圆函数的随机平均法,用于解杜芬系统的有界噪声激励下的单自由度非线性系统,结果表明它比基于广义调和函数的随机平均具有更高的精度[28].由于椭圆余弦函数是相应保守杜芬系统的精确解,因此引入基于椭圆函数的随机平均法来解船舶横摇问题.

  • 本文提出了一种基于Jacobi椭圆函数的随机平均法,用于求解有界噪声激励下船舶横摇运动幅值的概率密度.利用局部灵敏度分析,研究了系统中各个参数对船舶横摇幅值响应的影响.通过引入基于Jacobi椭圆函数的变换,导出相应的随机微分方程,应用随机平均原理将响应的幅值近似为一个马尔科夫扩散过程并建立相应的伊藤方程,通过求解FPK方程,得到概率密度分布.将该方法应用于有界噪声激励下的船舶横摇系统,研究其随机响应,并通过蒙特卡洛模拟验证了该方法的可行性和准确性.同时对系统在有界噪声激励下各参数的灵敏度进行分析,进一步确认方法的有效性.

  • 1 灵敏度分析

  • 船舶横摇运动微分方程可近似描述为单自由度系统[19],其在谐和激励下可表示为

  • φ¨+αφ˙+βφ˙|φ˙|+γφ-μφ3=EcosΩt
    (1)
  • 其中φt)表示t时刻的横摇角.该模型中包含绝对值阻尼项和负非线性刚度项,其6个系统参数的定义和取值可以参考表1.

  • 本节研究模型(1)中船舶横摇角对表1所示参数的局部灵敏度,实现了模型的不确定性量化,第i个响应变量(pi=φφ˙φ¨)随第j个参数值(qj={αβγμEΩ})变化的灵敏度[29-31]可表示为

  • Si,j=qjuiqj
    (2)
  • 表1 模型(1)的参数及物理意义

  • Table1 Physical meanings and values of the system parameters in model (1)

  • 其中Si,j是半相对灵敏度函数.通过中心差分法和龙格-库塔法[32],可得到横摇角φ对系统参数的灵敏度,如图1所示.由从图1可知,参数ΩγE的变化对船舶横摇角度有比较明显的影响,例如Ωγ的扰动可以使横摇角φ变化超过300%,而参数μαβ的变化对φ几乎没有影响.

  • 图1 船舶横摇角φ对各参数的灵敏度分析

  • Fig.1 Sensitivity analysis of roll angle φ to system parameters

  • 通过局部灵敏度分析,初步评估模型中各参数对横摇角的影响程度.然而,有界噪声激励下的系统参数将如何影响横摇角φ,以及如何获得响应统计量的概率解,将在接下来的两节中讨论.

  • 2 基于Jacobi椭圆函数的随机平均法

  • 对于有界噪声激励下的非线性系统,其运动方程可描述为 [28]

  • φ¨+c1φ+c3φ3+εg(φ,φ˙)=Eξ(t)
    (3)
  • 其中c1c3分别为系统的线性和非线性刚度系数,表示小量,gXX˙)表示系统的阻尼,E表示激励的幅值,ξt)是有界噪声激励[1314],可表示为

  • ξ(t)=cos(Ωt+σB(t)+χ)
    (4)
  • 其中Ωσ2分别为激励的平均频率和有界噪声的强度,Bt)为单位维纳过程,χ为[0,2π]区间上均匀分布的随机变量,ξt)是均值为零的广义平稳随机过程,其功率谱密度和自相关函数分别为:

  • S(ω)=σ24πω2+Ω2+σ4/4ω2-Ω2-σ4/42+σ4ω2
    (5)
  • R(τ)=12exp-σ22|τ|cosΩτ
    (6)
  • 本文中,σ取较小的值时,随机激励为窄带随机过程,当ε=E=0时,系统(3)退化为保守的杜芬振子对应的问题,即

  • φ¨+c1φ+c3φ3=0
    (7)
  • 其解为Jacobi椭圆函数,表示如下[25]:

  • φ=aep(ν-,κ-)
    (8)
  • φ˙=aλ-ω-ep'(ν-,κ-)
    (9)
  • 其中ep'ν-κ-为Jacobi椭圆函数,根据线性与非线性刚度的正负号不同,可以定义snν-κ-cnν-κ-dnν-κ-,如表2所示; aω-分别为椭圆函数的幅值和频率; ep'ν-κ-表示/v-,表3中给出了具体的求导结果; ν-λ-可以表示为

  • ν-=λ-(ω-t+θ-),λ-=λ-(κ-)=nK(κ-)π
    (10)
  • θ-κ-分别表示相角和模; Kκ-为椭圆函数的第一类完全积分.对式(9)求导,并将其与式(8)代入式(7),可以得到aω-κ-的关系式[23]

  • λ-2ω-2ρ2(κ-)+c1=0,2λ-2ω-2ρ1(κ-)+c3a2=0
    (11)
  • 其中ρ1κ-ρ2κ-随表3中不同椭圆函数而变化,通过式(11),频率ω-和模κ-可以分别表示为关于幅值的函数ω-=fωaκ-=fκa.

  • 表2 系统(7)对应的精确解

  • Table2 The exact solution of the system (7)

  • 表3 不同函数ep对应的和

  • Table3 Definition of and depending on the JEFs

  • 当ε≠0为小量时,系统(3)可以近似为

  • φ=A(t)ep(ν,κ)
    (12)
  • φ˙=A(t)ep'(ν,κ)
    (13)
  • 其中

  • ν=λu,λ=λ(κ)=nK(κ)π,κ=fκ(A)u=Ψ+Θ,dΨdt=ω,ω=fω(A)
    (14)
  • 为了简化,将Jacobi椭圆函数ep(νκ)和Zνκ)记作ep和Z,可得[25]

  • A˙ep+Θ˙Aλep'+A˙dfκdAAψ1fκ1-fκ2=0
    (15)
  • 其中

  • ψ1=-Zep'+ρ1ep-ep3
    (16)
  • 对(13)式求导可得

  • φ¨=A˙λωep'+Aλ2ω(ω+Θ˙)ψ2+κ˙Aλωκl2ψ3+Aλep'dωdAA˙
    (17)
  • 其中

  • ψ2=ep2ρ1ep2+ρ2
    (18)
  • ψ3=EK-l2-ρ3ep'-2ρ1ep2ep'-Zψ2
    (19)
  • 将式(12)-(14)及(17)代入到式(3)中,可得

  • A˙λfωep'+Θ˙Aλ2fωψ2+A˙A˙kdAAλfωψ3fk1-fk2+Aλep'dfωdAA˙=Eξ(t)-η-εgAep,Aλfωep'
    (20)
  • 其中

  • η=λ2fω2ρ2+γAep+2λ2fωω2ρ1-μA2Aep3
    (21)
  • 联立随机微分方程(15)和(20),解得

  • A˙=F1(A,Θ,Ωt+σB(t)+χ)Θ˙=F2(A,Θ,Ωt+σB(t)+χ)
    (22)
  • 其中A是小量,由式(21)知η也是小量,AΘ是慢变过程.考虑外共振系统在窄带激励下的响应问题,引入一个新的变量Γ,其中qp为小的整数.于是,可将式(22)变换成如下的随机微分方程

  • dA=F1dtdΓ=Ω-qpfω-qpF2dt+σdB(t)
    (23)
  • u进行平均,可得如下伊藤随机微分方程

  • dA=m1(A,Γ)dtdΓ=m2(A,Γ)dt+σdB(t)
    (24)
  • AΓ)是二维的扩散过程,式(24)对应的FPK方程如下:

  • -am1(a,γ)p-γm2(a,γ)p+σ222pγ2=0
    (25)
  • 其中

  • m1(A,Γ)=12π02π Eξ(t)+εg+ηλfωG1ep'dum2(A,Γ)=12π02π Ω-qpfω-qpF2duF2=[Eξ(t)-εg-η]ep+dfkdAAψ1fk1-fk2Aλ2fωG1G1(A,Θ)=ψ2ep-(ep)'+dfκdAAfκ1-fκ2ψ1ψ2-ψ3ep'-dfωdAAep'2fω
    (26)
  • γ对应的具有周期性的边界条件为:

  • pγ+2nπ=pγ(p/γ)γ+2nπ=(p/γ)γ
    (27)
  • a的边界条件为:

  • (28)
  • 根据FPK方程(26)以及边界条件(27)和(28),利用数值方法求得随机激励下系统的联合概率密度分布.通过联合概率密度在0-2π上对相位进行积分,可得关于幅值的随机响应.

  • 作者之前曾将基于雅克比椭圆函数的随机平均法与基于广义谐和函数的随机平均结果进行比较[28].以杜芬-范德波振子为例,分别研究了基于雅克比椭圆函数的随机平均法与基于广义谐和函数的随机平均法来求解该系统振幅的一阶中心矩的变化; 同时在不同组别的参数下,计算两种方法与蒙特卡洛数值模拟结果的相对误差.研究表明,基于椭圆函数的随机平均法比基于广义谐和函数的随机平均法的结果相对更为准确.

  • 3 数值分析结果

  • 应用第2节中基于Jacobi椭圆函数的随机平均法,求解关于系统(3)得到其概率密度分布,对有界噪声激励下的船舶横摇系统[19],将式(3)改写为

  • φ¨+αφ˙+βφ˙|φ˙|+γφ-μφ3=Ecos(Ωt+σB(t)+χ)
    (29)
  • 通常,船舶横摇系统呈现出负的非线性刚度,对应表2中的软弹簧情形,令其中C1=γC3=-μ,从而椭圆函数为

  • ep(ν,κ)=sn(ν,κ)κ=fκ(A)=-c3A2/2c1+c3A2ω=fω(A)=c1+c3A2/2/α
    (30)
  • 代入上一节中,可得

  • m1(A,Γ)=12π02π -Eξ(t)+ep'αAλω+βAλωep'+ηλfωG1ep'dum2(A,Γ)=12π02π Ω-qpfω-qpF2duF2=Eξ(t)-ep'αAλω+βAλωep'-ηep+dfκdAAψ1fκ1-fκ2Aλ2fωG1
    (31)
  • 图2 参数ΩγEμαβ取不同数值时,船舶横摇系统的随机响应分析(实线是文中的方法,点线是蒙特卡洛模拟结果,各参数选取如表1所示,σ2=0.02)

  • Fig.2 The response statistics densities of the amplitude of ship roll system for different values of Ω, γ, E, μ, α, β, (Solid lines: results of the proposed method, and symbols: results of MCS. The other system parameters are the same as those in Table1 except that σ2=0.02)

  • 同时系统参数选取如表1所示,利用差分法可以解得FPK方程(25)的数值解.假设a∈(0,2),γ∈(0,2π),通过设置差分网格数50×60,将矩形域(0,2)×(0,2π)划分为对应的网格,将偏微分方程(25)转化为差分方程,数值求解得到船舶横摇振幅的概率解.图2显示了表1中6个参数对船舶横摇运动概率解的影响,其中实线表示基于Jacobi椭圆函数的计算结果,点线是蒙特卡洛模拟得到的结果(样本量为10000,时间步长为0.02).结果显示,两种方法具有较好的吻合度.

  • 由图2(a)、图2(b)、图2(d)可知,αβμ三个参数的变化影响概率密度峰值的高度,对响应统计密度的影响不大,即对概率密度在横坐标上的移动影响不大.随着α的值由0.1变化到0.3,响应统计密度峰值对应的幅值由0.5向左移了约0.1; 然而相同变化下的β,其对应的值略有左移,几乎没有变化; 对于μ,与αβ不同的是,对应的幅值随其增大而增大.即,前两个参数与船舶横摇运动振幅负相关,第三个参数与船舶横摇运动振幅正相关,这一影响程度与相关性结果与灵敏度分析的结果相吻合.由此可知,模型中的阻尼系数和非线性刚度系数对船舶横摇幅值的影响不大.由图2(c)、2(e)和2(f)可以看出,随着参数E的大小由0.1变化到0.3,响应统计密度曲线由陡峭变得平滑,响应统计密度峰值和其对应的幅值都发生了一定的变化; Ω变化的规律类似于E; 对于γ,其大小由0.9变化到1.1时,其对应的幅值减小.由此可知,随着EΩ增加或γ减小,概率的峰值向右移动,曲线变得更加平缓,同时三个参数的变化也影响了概率峰的高度.因此,γEΩ因子对船舶横摇幅值的影响比αβμ的影响更大.

  • 图3 船舶横摇系统的随机响应随σ2变化分析,实线是文中的方法,点线是蒙特卡洛模拟结果.各参数:(a)α=0.1,β=0.1,γ=1.0,μ=0.08,E=0.2,Ω=0.8;(b)α=0.1,β=0.05,γ=0.9,μ=0.08,E=0.2,Ω=0.7;(c)α=0.2,β=0.1,γ=1.0,μ=0.05,E=0.25,Ω=0.75;(d)α=0.2,β=0.1,γ=1.0,μ=0.05,E=0.15,Ω=0.75

  • Fig.3 The response statistics densities of the amplitude of ship roll system for different values of σ2; Solid lines: results of the proposed method, and symbols: results of MCS. The other system parameters are: (a) α=0.1, β=0.1, γ=1.0, μ=0.08, E=0.2, Ω=0.8; (b) α=0.1, β=0.05, γ=0.9, μ=0.08, E=0.2, Ω=0.7; (c) α=0.2, β=0.1, γ=1.0, μ=0.05, E=0.25, Ω=0.75; (d) α=0.2, β=0.1, γ=1.0, μ=0.05, E=0.15, Ω=0.75

  • 此外,本文还研究了四组系统参数下频率扰动强度σ2对横摇响应统计量的影响,如图3所示.相关σ2参数根据文献[13]中对船舶横摇和有界噪声的描述选取,中心频率Ω由船舶横摇的共振频率确定.由于随机平均法主要用于求解小阻尼、弱激励问题,因此阻尼和激励相关参数的取值不宜过大.对于不同的海域大小和海面情况,谱峰频率的取值不同,国际水池会议的标准波谱密度对应的谱峰频率在0.1~0.8范围内[33].从图3(a)和图3(b)可知,系数β的变化对概率峰的位置影响较小; γΩ因子对船舶横摇幅值的影响较大,γ的值由1.0变为0.9,Ω的值由0.8变为0.7.由图2可知,这两个参数与船舶横摇运动振幅相关性相反,因此图3(a)和3(b)形状相似.对于图3(c)和图3(d),仅E的大小不同,然而由图(2)可知,激励的参数E越小,概率峰越陡,与图(3)的变化规律一致.参数σ2类似但区别于αβμ,对响应统计密度的影响不大,它的变化几乎对船舶横摇运动的振幅没有影响,只对概率峰值的高度有影响,且σ2的变化对峰值的影响明显大于阻尼系数和非线性刚度系数.

  • 图4 有界噪声激励下横摇角φ对各参数的灵敏度分析

  • Fig.4 Sensitivity analysis of roll angle φ to system parameters under bounded noise excitation

  • 图4给出了有界噪声激励下船舶横摇角φ对参数的灵敏度性分析.图4与图1的区别是,激励由余弦激励变为随机激励,同时图4也考虑了σ2的影响.其他参数对φ的影响与图1相似,易知,在随机激励下作用下,灵敏度的范围增大; 图4与图1中各个系统参数对横摇角的影响趋势类似,即γEΩ因子对船舶横摇幅值的影响比αβμ的影响更大; 此外,由于本文选取的噪声强度较小(σ2 = 0.02),σ的扰动对φ的影响较小.

  • 上述图1、图2和图3与实际情况是相匹配的,激励幅值E的值决定了外载的大小,外激频率的大小Ω靠近系统的共振频率,它们自然对船舶横摇系统产生重大的影响; 线性刚度系数γ与船舶横摇角成反比,这三个参数的影响是明显的.前两个参数与船舶横摇运动振幅正相关,第三个参数负相关.此外,阻尼系数αβ对横摇运动的角速度有影响,但对角位移无直接影响; 非线性恢复力系数μ对横摇运动角的三次方有影响,然而对于低角位移的情形,三次平方根的影响会大大减弱.中心频率扰动参数σ2Ω小,对响应统计密度影响不大.图2和图3的求解结果与图1和图4的灵敏度分析结果完全吻合,且基于椭圆函数的随机平均法得到的概率解与蒙特卡洛模拟结果一致,表明了本文所提出方法的合理性.

  • 4 结论

  • 本文提出了一种基于Jacobi椭圆函数的随机平均法,用于求解有界噪声激励下船舶横摇运动的随机响应.通过局部灵敏度分析和与蒙特卡洛模拟验证该方法的可行性,得到以下结论:

  • (1)对谐和及有界噪声激励下的船舶横摇模型进行局部灵敏度分析,研究了系统中主要参数对横摇角的影响,结果表明不同参数对横摇角的作用效应不同.参数ΩγE的变化对船舶横摇角度有明显的影响,而参数μαβ的变化对其影响较小.对于有界噪声激励,中心频率扰动参数σ2Ω小,对响应统计密度影响不大.

  • (2)应用基于Jacobi椭圆函数的随机平均法求解单自由度横摇运动方程,得到系统的概率解.结果表明,αβμ三个参数的变化影响概率密度峰值的高度,对响应统计密度的影响不大,即对概率密度在横坐标上的移动影响不大.然而EΩγ三个参数的变化影响概率峰的高度,随着EΩ增加或γ减小,概率的峰值向右移动,曲线变得更加平缓.对于随机噪声强度σ2,其类似于αβμ,但与这三个参数的影响略有不同的是,其几乎完全不影响船舶横摇运动的振幅; 同时σ2对峰值的影响明显大于阻尼系数和非线性刚度系数.这与灵敏度分析的结果完全吻合,同时各个参数对船舶横摇角度的影响与灵敏度分析一致.

  • (3)与蒙特卡洛模拟结果进行比较,两种方法得到的结果基本一致.因此,本文提出的基于Jacobi椭圆函数的随机平均法,是求解有界噪声激励下船舶横摇运动问题的一种可靠方法.

  • 参考文献

    • [1] LLOYD A.Seakeeping:ship behavior in rough weather [M].Horwood,1989.

    • [2] SHINOZUKA M,DEODATIS G.Simulation of stochastic processes by spectral representation [J].Applied Mechanics Reviews,1991,44(4):191-203.

    • [3] KIM D S,LEE W K.Analysis of random ship rolling using partial stochastic linearization [J].Journal of the Society of Naval Architects of Korea,1995.

    • [4] GU J Y.Probability analysis of ship nonlinear roll-motion excited by white noise [J].Journal of Ship Mechanics,2006,10(3):17-25.

    • [5] CHAI W,NAESS A,LEIRA B J.Stochastic nonlinear ship rolling in random beam seas by the path integration method[J].Probabilistic Engineering Mechanics,2016,44(apr):43-52.

    • [6] 刘坤峰,靳艳飞.相关白噪声激励下双稳态 Duffing Van der Pol 系统的随机分岔[J].动力学与控制学报,2020,18(4):12-18.LIU K F,JIN Y F.Stochastic bifurcation in bistable Duffing Van der Pol system driven by correlated white noises[J].Journal of Dynamics and Control,2020,18(4):12-18.(in Chinese)

    • [7] COTTONE G,MD PAOLA,IBRAHIM R,et al.Stochastic ship roll motion via path integral method[J].International Journal of Naval Architecture and Ocean Engineering,2010,2(3):119-126.

    • [8] VAIPAN I U.Probability density function solution to nonlinear ship roll motion excited by external Poisson white noise [J].Science China Technological Sciences,2011,(5):1121-1125.

    • [9] MAMIS K I,ATHANASSOULIS G A,PAPADOPOULOS K E.Generalized FPK equations corresponding to systems of nonlinear random differential equations excited by colored noise,revisitation and new directions [J].Procedia Computer Science,2018,136:164-173.

    • [10] ATHANASSOULIS G A,KAPELONIS Z G,MAMIS K.Numerical solution of generalized fpk equations corresponding to random differential equations under colored noise excitation.The transient case [C].8th Conference on Computational Stochastic Mechanics(CSM8),2018.

    • [11] MAMIS K I,ATHANASSOULIS G A,KAPELONIS Z G.A systematic path to non-markovian dynamics:new response pdf evolution equations under Gaussian coloured noise excitation [J].The Royal Society,2018,(475):1-18.

    • [12] FALZARANO J M,SU Z,JAMNONGPIPATKUL A,et al.Solving the problem of nonlinear ship roll motion using stochastic dynamics:risk of capsizing [M].2019.

    • [13] ZHU W Q,CAI G Q.On bounded stochastic processes [M].Springer New York,2013:3-24.

    • [14] 朱位秋.非线性随机动力学与控制-Hamilton理论体系框架 [M].北京:科学出版社,2003.ZHU W Q.Nonlinear stochastic dynamics and control-Hamilton theoretical framework [M].Beijing:Science Press,2003.(in Chinese)

    • [15] ROBERTS J B.A stochastic theory for nonlinear ship rolling in irregular seas[J].Journal of Ship Research,1982,26:229-245.

    • [16] 张强,王剑龙,李扬,等.有界噪声激励下系统随机共振的改善 [J].动力学与控制学报,2021,19(6):52-58.ZHANG Q,WANG J L,LI Y,et al.Improvement of stochastic resonance under bounded noise excitation [J].Journal of Dynamics and Control,2021,19(6):52-58.(in Chinese)

    • [17] ROBERTS J B,VASTA M.Markov modeling and stochastic identification for nonlinear ship rolling in random waves [J].Philosophical Transactions of the Royal Society of London A,2000,358:1917-1941.

    • [18] WANG Y G,HUANG Z L,TAN J H.First passage probability of nonlinear ship rolling in random seas [J].Journal of Ship Mechanics,2008,12:870-879.

    • [19] WANG S L,JIN X L,HUANG Z L.Analytical evaluation of first passage failure for nonlinear ship rolling excited by random wave [J].Journal of Vibration Engineering & Technologies,2016,4(4):291-299.

    • [20] WANG L Y,TANG Y G,YAN L I,et al.Studies on stochastic parametric roll of ship with stochastic averaging method [J].China Ocean Engineering,2020,34(2):289-298.

    • [21] CVETICANIN L.Strongly nonlinear oscillators [M].Analytical Solutions,Springer,2014.

    • [22] TIEN W M,NAMACHCHIVAYA N S,COPPOLA V T.Stochastic averaging using elliptic functions to study nonlinear stochastic systems [J].Nonlinear Dynamic,1993,4:373-387.

    • [23] COPPOLA V T,RAND R H.Averaging using elliptic functions:approximation of limit cycles [J].Acta Mechanica,1990,81(3-4):125-142.

    • [24] OKABE T,KONDOU T,OHNISHI J.Elliptic averaging methods using the sum of Jacobian elliptic delta and zeta functions as the generating solution[J].International Journal of Non-Linear Mechanics,2011,46(1):159-169.

    • [25] OKABE T,KONDOU T.Improvement to the averaging method using the Jacobian elliptic function[J].Journal of Sound & Vibration,2009,320(1-2):339-364.

    • [26] RAKARIC Z,KOVACIC I.An elliptic averaging method for harmonically excited oscillators with a purely non-linear non-negative real-power restoring force[J].Communications in Nonlinear Science and Numerical Simulation,2013,18(7):1888-1901.

    • [27] ELÍAS ZÙÑIGA A.Exact solution of the cubic-quintic Duffing oscillator[J].Applied Mathematical Modelling,2013,37(4):2574-2579.

    • [28] 徐文俊,郑丽文,马品奎.一种改进的基于Jacobi椭圆函数的随机平均法[J].振动工程学报,2019,032(003):444-451.XU W J,ZHENG L W,MA P K.An improved stochastic averaging method based on Jacobian elliptic function [J].Journal of Vibration Engineering,2019,032(003):444-451.(in Chinese)

    • [29] WANG S L,XIE Z,ZHONG R,et al.Stochastic analysis of a predator-prey model with modified Leslie-Gower and Holling type Ⅱ schemes [J].Nonlinear Dynamic,2020,101:1245-1262.

    • [30] TRIPATHI J P,MEGHWANI S S,THAKUR M,et al.A modified Leslie-Gower predator-prey interaction model and parameter identifiability [J].Communications in Nonlinear Science & Numerical Simulation,2017,54:331-346.

    • [31] SOURAV,KUMAR,SASMAL,et al.Dynamics of a predator-prey system with fear and group defense-ScienceDirect [J].Journal of Mathematical Analysis & Applications,2020,481(1):123471.

    • [32] WANG S L,WANG Z,XU C,et al.Sensitivity analysis and stationary probability distributions of a stochastic two-prey one-predator model[J].Applied Mathematics Letters,2020,116:106996.

    • [33] 朱位秋.随机振动[M].北京;科学出版社,1998.ZHU W Q.Random vibration[M].Beijing:Science Press,1998.(in Chinese)

  • 参考文献

    • [1] LLOYD A.Seakeeping:ship behavior in rough weather [M].Horwood,1989.

    • [2] SHINOZUKA M,DEODATIS G.Simulation of stochastic processes by spectral representation [J].Applied Mechanics Reviews,1991,44(4):191-203.

    • [3] KIM D S,LEE W K.Analysis of random ship rolling using partial stochastic linearization [J].Journal of the Society of Naval Architects of Korea,1995.

    • [4] GU J Y.Probability analysis of ship nonlinear roll-motion excited by white noise [J].Journal of Ship Mechanics,2006,10(3):17-25.

    • [5] CHAI W,NAESS A,LEIRA B J.Stochastic nonlinear ship rolling in random beam seas by the path integration method[J].Probabilistic Engineering Mechanics,2016,44(apr):43-52.

    • [6] 刘坤峰,靳艳飞.相关白噪声激励下双稳态 Duffing Van der Pol 系统的随机分岔[J].动力学与控制学报,2020,18(4):12-18.LIU K F,JIN Y F.Stochastic bifurcation in bistable Duffing Van der Pol system driven by correlated white noises[J].Journal of Dynamics and Control,2020,18(4):12-18.(in Chinese)

    • [7] COTTONE G,MD PAOLA,IBRAHIM R,et al.Stochastic ship roll motion via path integral method[J].International Journal of Naval Architecture and Ocean Engineering,2010,2(3):119-126.

    • [8] VAIPAN I U.Probability density function solution to nonlinear ship roll motion excited by external Poisson white noise [J].Science China Technological Sciences,2011,(5):1121-1125.

    • [9] MAMIS K I,ATHANASSOULIS G A,PAPADOPOULOS K E.Generalized FPK equations corresponding to systems of nonlinear random differential equations excited by colored noise,revisitation and new directions [J].Procedia Computer Science,2018,136:164-173.

    • [10] ATHANASSOULIS G A,KAPELONIS Z G,MAMIS K.Numerical solution of generalized fpk equations corresponding to random differential equations under colored noise excitation.The transient case [C].8th Conference on Computational Stochastic Mechanics(CSM8),2018.

    • [11] MAMIS K I,ATHANASSOULIS G A,KAPELONIS Z G.A systematic path to non-markovian dynamics:new response pdf evolution equations under Gaussian coloured noise excitation [J].The Royal Society,2018,(475):1-18.

    • [12] FALZARANO J M,SU Z,JAMNONGPIPATKUL A,et al.Solving the problem of nonlinear ship roll motion using stochastic dynamics:risk of capsizing [M].2019.

    • [13] ZHU W Q,CAI G Q.On bounded stochastic processes [M].Springer New York,2013:3-24.

    • [14] 朱位秋.非线性随机动力学与控制-Hamilton理论体系框架 [M].北京:科学出版社,2003.ZHU W Q.Nonlinear stochastic dynamics and control-Hamilton theoretical framework [M].Beijing:Science Press,2003.(in Chinese)

    • [15] ROBERTS J B.A stochastic theory for nonlinear ship rolling in irregular seas[J].Journal of Ship Research,1982,26:229-245.

    • [16] 张强,王剑龙,李扬,等.有界噪声激励下系统随机共振的改善 [J].动力学与控制学报,2021,19(6):52-58.ZHANG Q,WANG J L,LI Y,et al.Improvement of stochastic resonance under bounded noise excitation [J].Journal of Dynamics and Control,2021,19(6):52-58.(in Chinese)

    • [17] ROBERTS J B,VASTA M.Markov modeling and stochastic identification for nonlinear ship rolling in random waves [J].Philosophical Transactions of the Royal Society of London A,2000,358:1917-1941.

    • [18] WANG Y G,HUANG Z L,TAN J H.First passage probability of nonlinear ship rolling in random seas [J].Journal of Ship Mechanics,2008,12:870-879.

    • [19] WANG S L,JIN X L,HUANG Z L.Analytical evaluation of first passage failure for nonlinear ship rolling excited by random wave [J].Journal of Vibration Engineering & Technologies,2016,4(4):291-299.

    • [20] WANG L Y,TANG Y G,YAN L I,et al.Studies on stochastic parametric roll of ship with stochastic averaging method [J].China Ocean Engineering,2020,34(2):289-298.

    • [21] CVETICANIN L.Strongly nonlinear oscillators [M].Analytical Solutions,Springer,2014.

    • [22] TIEN W M,NAMACHCHIVAYA N S,COPPOLA V T.Stochastic averaging using elliptic functions to study nonlinear stochastic systems [J].Nonlinear Dynamic,1993,4:373-387.

    • [23] COPPOLA V T,RAND R H.Averaging using elliptic functions:approximation of limit cycles [J].Acta Mechanica,1990,81(3-4):125-142.

    • [24] OKABE T,KONDOU T,OHNISHI J.Elliptic averaging methods using the sum of Jacobian elliptic delta and zeta functions as the generating solution[J].International Journal of Non-Linear Mechanics,2011,46(1):159-169.

    • [25] OKABE T,KONDOU T.Improvement to the averaging method using the Jacobian elliptic function[J].Journal of Sound & Vibration,2009,320(1-2):339-364.

    • [26] RAKARIC Z,KOVACIC I.An elliptic averaging method for harmonically excited oscillators with a purely non-linear non-negative real-power restoring force[J].Communications in Nonlinear Science and Numerical Simulation,2013,18(7):1888-1901.

    • [27] ELÍAS ZÙÑIGA A.Exact solution of the cubic-quintic Duffing oscillator[J].Applied Mathematical Modelling,2013,37(4):2574-2579.

    • [28] 徐文俊,郑丽文,马品奎.一种改进的基于Jacobi椭圆函数的随机平均法[J].振动工程学报,2019,032(003):444-451.XU W J,ZHENG L W,MA P K.An improved stochastic averaging method based on Jacobian elliptic function [J].Journal of Vibration Engineering,2019,032(003):444-451.(in Chinese)

    • [29] WANG S L,XIE Z,ZHONG R,et al.Stochastic analysis of a predator-prey model with modified Leslie-Gower and Holling type Ⅱ schemes [J].Nonlinear Dynamic,2020,101:1245-1262.

    • [30] TRIPATHI J P,MEGHWANI S S,THAKUR M,et al.A modified Leslie-Gower predator-prey interaction model and parameter identifiability [J].Communications in Nonlinear Science & Numerical Simulation,2017,54:331-346.

    • [31] SOURAV,KUMAR,SASMAL,et al.Dynamics of a predator-prey system with fear and group defense-ScienceDirect [J].Journal of Mathematical Analysis & Applications,2020,481(1):123471.

    • [32] WANG S L,WANG Z,XU C,et al.Sensitivity analysis and stationary probability distributions of a stochastic two-prey one-predator model[J].Applied Mathematics Letters,2020,116:106996.

    • [33] 朱位秋.随机振动[M].北京;科学出版社,1998.ZHU W Q.Random vibration[M].Beijing:Science Press,1998.(in Chinese)

  • 微信公众号二维码

    手机版网站二维码