梁翠翠,張啟敏
(寧夏大學(xué) 數(shù)學(xué)計(jì)算機(jī)學(xué)院,寧夏 銀川 750021)
*帶分?jǐn)?shù)Brown運(yùn)動(dòng)的非線(xiàn)性隨機(jī)兩種群系統(tǒng)的最優(yōu)控制
梁翠翠,張啟敏
(寧夏大學(xué) 數(shù)學(xué)計(jì)算機(jī)學(xué)院,寧夏 銀川 750021)
討論帶分?jǐn)?shù)Brown運(yùn)動(dòng)的非線(xiàn)性隨機(jī)兩種群系統(tǒng)的最優(yōu)控制問(wèn)題,得到了當(dāng)外界環(huán)境對(duì)種群系統(tǒng)產(chǎn)生影響的條件下,控制為最優(yōu)的必要條件,所得到的結(jié)論是已有確定性種群系統(tǒng)的擴(kuò)展.
非線(xiàn)性;兩種群;最優(yōu)控制;分?jǐn)?shù)布朗運(yùn)動(dòng)
考慮如下隨機(jī)兩種群系統(tǒng)(P)
分?jǐn)?shù)Brown運(yùn)動(dòng)具有自相似性和長(zhǎng)記憶性,由于種群系統(tǒng)也具有長(zhǎng)記憶性,即種群系統(tǒng)將來(lái)的狀態(tài)不僅僅只跟現(xiàn)在的狀態(tài)有關(guān),跟過(guò)去的狀態(tài)也是有聯(lián)系的,不是相互獨(dú)立的.所以考慮帶分?jǐn)?shù)Brown運(yùn)動(dòng)隨機(jī)種群系統(tǒng)的有關(guān)問(wèn)題更符合實(shí)際.關(guān)于分?jǐn)?shù)Brown運(yùn)動(dòng)大量文獻(xiàn)對(duì)其進(jìn)行了研究,例如:文獻(xiàn)[1]討論了分?jǐn)?shù)Brown運(yùn)動(dòng)的定義,自相似性,連續(xù)性,可微性及其應(yīng)用.由于分?jǐn)?shù)Brown運(yùn)動(dòng)不是半鞅,一般的Ito積分理論不再適用,文獻(xiàn)[2]根據(jù)隨機(jī)變分原理給出了帶分?jǐn)?shù)Brown運(yùn)動(dòng)函數(shù)的Ito公式.文獻(xiàn)[3]討論了線(xiàn)性和非線(xiàn)性帶分?jǐn)?shù)Brown運(yùn)動(dòng)倒向隨機(jī)微分方程解的存在性和唯一性.文獻(xiàn)[4]研究了短期分?jǐn)?shù)Brown運(yùn)動(dòng)環(huán)境中最優(yōu)投資組合選擇問(wèn)題,運(yùn)用Hamilton函數(shù)和Lagrange算子將隨機(jī)最優(yōu)控制問(wèn)題轉(zhuǎn)化為非隨機(jī)優(yōu)化問(wèn)題,根據(jù)確定最優(yōu)控制原理得到了最優(yōu)策略的顯示解.文獻(xiàn)[5]討論了分?jǐn)?shù)Brown運(yùn)動(dòng)與Hurst指數(shù)之間的關(guān)系,給出了當(dāng)Hurst指數(shù)取不同值時(shí)分?jǐn)?shù)Brown運(yùn)動(dòng)相應(yīng)的性質(zhì).文獻(xiàn)[6]研究了半線(xiàn)性隨機(jī)加熱方程描述的系統(tǒng)最優(yōu)控制極大值原理的充分性,結(jié)論用于解決最優(yōu)收獲問(wèn)題.文獻(xiàn)[7]討論了介于0<H<1任意參數(shù)分?jǐn)?shù)布朗運(yùn)動(dòng)的Ito公式.文獻(xiàn)[8]根據(jù)偏微分方程和分?jǐn)?shù)Ito公式的解得到了帶分?jǐn)?shù)Brown運(yùn)動(dòng)的線(xiàn)性倒向隨機(jī)微分方程的顯式解.文獻(xiàn)[9]討論了帶分?jǐn)?shù)Brown運(yùn)動(dòng)的隨機(jī)極大值原理.文獻(xiàn)[10]研究了用Riemann-Liouville分?jǐn)?shù)Brown運(yùn)動(dòng)在生物系統(tǒng)考慮隨機(jī)性,討論了帶分?jǐn)?shù)Brown運(yùn)動(dòng)的兩個(gè)種群的最優(yōu)管理,運(yùn)用Lagrange乘子法將隨機(jī)問(wèn)題轉(zhuǎn)化為非隨機(jī)問(wèn)題,根據(jù)確定系統(tǒng)最優(yōu)控制變分原理得到最優(yōu)解.
定義1.1 (Ω,F(xiàn),P)是概率空間,H為 Hurst參數(shù),0<H<1,隨機(jī)過(guò)程{B(t,H),t≥0}定義在該概率空間上為H階分?jǐn)?shù)布朗運(yùn)動(dòng)(f Bm)H具有下列性質(zhì):
(1)P{B(0,H)=H}=1,
(2)對(duì)任意的t≥0,B(t,H)是F可測(cè)隨機(jī)變量且E[B(t,H)]=0,
(3)對(duì)任意的t≥0,
根據(jù)文獻(xiàn)[11-12],對(duì)分?jǐn)?shù)布朗運(yùn)動(dòng)(f Bm)H的進(jìn)一步討論如下.
(1)方程(2)可以由下列等式得到:
Γ是gamma函數(shù).
(3)引用 Maruyama記法(文獻(xiàn)[12])有定義1.2 設(shè)f表示連續(xù)函數(shù),則它的H階分?jǐn)?shù)導(dǎo)數(shù)定義為:
證明 根據(jù)文獻(xiàn)[14]知:控制(ˉu,ˉv)∈U為問(wèn)題1.1最優(yōu)控制的充分條件是對(duì)任意的(u,v)∈U下式成立
由于種群系統(tǒng)經(jīng)常受到外部環(huán)境的影響,具有隨機(jī)擾動(dòng)的種群系統(tǒng)模型便成了人們討論的熱點(diǎn).然而,對(duì)于種群系統(tǒng),大量文獻(xiàn)只針對(duì)確定系統(tǒng)進(jìn)行研究,并未考慮隨機(jī)因素的影響.所以本文討論了帶分?jǐn)?shù)Brown運(yùn)動(dòng)的非線(xiàn)性隨機(jī)兩種群系統(tǒng)的最優(yōu)控制問(wèn)題,給出了控制為最優(yōu)的必要條件.本文的結(jié)論是確定性種群系統(tǒng)的擴(kuò)展.
[1] Benoit B.Mandelbrot,John W.Van Ness.Fractional Brownian Motion,F(xiàn)ractional Noises and Applications[J].SIAMReview,1968,10(4):422-437.
[2] Decreeusefond,A SüStunel.Stochastic Analysis of the Fractional Brownian Motion[J].PotentialAnalysis,1990,10:177-214.
[3] Hu Yao-zhong,Peng Shi-ge.Backward Stochastic Differential Equation Driven by Fractional Brownian Motion[J].SIAM JControlOptim,2009,48(3):1675-1700.
[4] Gao Jian-wei.Optimal Portfolio Selection Under the Short-Range Fractional Brownian Motion[C]//International Conference on Computational Intelligence and natural Computing,2009:433-436.
[5] 牛奉高,劉維奇.分?jǐn)?shù)布朗運(yùn)動(dòng)與 Hurst指數(shù)的關(guān)系研究[J].山西大學(xué)學(xué)報(bào):自然科學(xué)版,2010,33(3):380-383.
[6] Bernt Ksendal.Optimal Control of Stochastic Partial Differential Equations[J].Stochastic Analysis and Applications,2005,23:168-179.
[7] Christian Bender.An Ito Formula for Generalized Functional of Fractional Brownian Motions with Arbitrary Hurst Parameter[J].StochasticProcessesandtheirApplications,2003,104:81-106.
[8] Christian Bender.Explicit Solutions of a class of Linear Fractional BSDEs[J].SystemsandControlLetters,2005,54:671-680.
[9] Francesca Biagini,Hu Yao-zhong,Bernt Ksendal,Agues Sulem.A Stochastic Maximum Principle for Processes Driven by Fractional Brownian Motion[J].StochasticProcessesandtheirApplications,2002,100:233-253.
[10] Guy Jumarie.New Stochastic Fractional Models for Malthusian Growth,the Poissonian Birth Process and Optimal Management of Populations[J].MathematicalandComputerModelling,2006,44:231-254.
[11] Guy Jumarie.Fractional Master Equation:non-standard Analysis and Liouville-Riemann Derivative[J].Chaos,Solitons andFractals,2001,12:2577-2587.
[12] Guy Jumarie.On the Representation of Fractional Brownian Motion as an Integral with Respect to(dt)a[J].Applied MathematicsLetters,2005,18:739-748.
[13] Guy Jumarie.Fractional Brownian Motions Via Random Walk in the Complex Plane and Via Fractional Derivative.Comparison and Further Results on Their Fokker-Planck Equations[J].Chaos,SolitonsandFractals,2004,22:907-925.
[14] Abel Cadenils.A Stochastic Maximum Principle for Systems with Jumps with Applications to Finance[J].Systemsand ControlLetters,2002,47:433-444.
Optimal Control of Nonlinear Stochastic Two
Population Systems with Fractional Brownian Motion
LIANG Cui-cui,ZHANG Qi-min
(SchoolofMathematicsandComputer,NingxiaUniversity,Yinchuan750021,China)
The optimal control of nonlinear stochastic two population systems driven by fractional brownian motion is discussed.When the external environment affects the system,the necessary conditions for the optimization was made clear,obtained the conclusion was the expansion of definite population system.
nonlinear;two systems;optimal control;fractional Brownian motion
O175.12
A
0253-2395(2012)04-0595-08*
2011-07-05;
2011-10-26
國(guó)家自然科學(xué)基金(11061024)
梁翠翠(1988-),女,碩士,主要研究方向:運(yùn)籌學(xué)與控制論.E-mail:liangcui4567@163.com