林清梅,張雪梅,蔣建新
(1.文山學(xué)院 數(shù)學(xué)學(xué)院,云南 文山 663099;2.曲靖師范學(xué)院 數(shù)學(xué)與信息科學(xué)學(xué)院,云南 曲靖 655011)
時標(biāo)上一類時滯BAM神經(jīng)網(wǎng)絡(luò)的偽概周期解
林清梅1,張雪梅2,蔣建新1
(1.文山學(xué)院 數(shù)學(xué)學(xué)院,云南 文山 663099;2.曲靖師范學(xué)院 數(shù)學(xué)與信息科學(xué)學(xué)院,云南 曲靖 655011)
研究了時標(biāo)上一類具有分布型泄露項(xiàng)時滯的中立型BAM神經(jīng)網(wǎng)絡(luò)的偽概周期解的存在性和全局指數(shù)穩(wěn)定性。利用指數(shù)二分性理論、壓縮映射原理、不動點(diǎn)理論和李雅普諾夫函數(shù)法,得到了該系統(tǒng)的偽概周期解的存在性和全局指數(shù)穩(wěn)定性的充分條件。
時標(biāo);偽概周期解;存在性;全局指數(shù)穩(wěn)定性
在自然界中,概周期解和偽概周期解是較周期解更為常見的現(xiàn)象。文獻(xiàn)[1-3]中研究了具連續(xù)的泄露項(xiàng)時滯的神經(jīng)網(wǎng)絡(luò)的概周期解和偽概周期解。自從Li和Wang中提出時標(biāo)上的概周期函數(shù)的概念后[4],許多關(guān)于時標(biāo)上的神經(jīng)網(wǎng)絡(luò)的概周期解的研究陸續(xù)發(fā)表出來[5-7]。而據(jù)我們所知,時標(biāo)上偽概周期解的研究還較少?;谝陨鲜聦?shí),本文研究時標(biāo)上一類具有分布型泄露項(xiàng)時滯的中立型BAM神經(jīng)網(wǎng)絡(luò)的偽概周期解。系統(tǒng)如下:
其中各變量含義詳見文獻(xiàn)[5]。
令 T表示一個概周期時標(biāo)。為了方便起見,記R=(-∞, +∞) 和R+=(0, +∞),設(shè)f∶T→R是偽概周期函數(shù),記。
令X={?=(φ1, φ2, …, φn, ψ1, ψ2, … ψm)T|φi, ψj,∈C1(T, R),φi, ψj是T上的偽概周期函數(shù),i=1, 2,…, n, j=1, 2, …m},其范數(shù)‖?‖=max{|φ|1, |ψ|1},其中,C1(T, R)是T上連續(xù)函數(shù)的集合,其上定義了?導(dǎo)數(shù),顯然X是一個Banach 空間。
系統(tǒng)(1)的初始條件為xi(s)= φi(s),yj(s),ψj(s),s∈(-∞,0]T={t|t∈(-∞,0]∩T},其中φi, ψj∈C1((-∞,0]T, R), i=1, 2, …, n, j=1, 2, …m。假設(shè)以下條件成立:
?*(s)=(φ1*(s), φ2*(s), …, φn*(s), ψ1*(s), ψ2*(s), …的偽概周期解.如果存在λ>0使得對系統(tǒng)(1)具初值?(s)=(φ1(s), φ2(s), …, φn(s), ψ1(s), ψ2(s), …ψm(s))T的任意解
z(t)=(x1(t), x2(t), …xn(t), y1(t), y2(t), …, ym(t))T滿足
那么系統(tǒng)(1)的偽概周期解z*(t)稱為是全局指數(shù)穩(wěn)定的。
在這一節(jié),我們將證明系統(tǒng)(1)的解是全局指數(shù)穩(wěn)定的。為此,我們需進(jìn)一步假設(shè):(H4)對于t∈(0, ∞)T,存在正常數(shù)λ∈Rv+,γi和χj使得
定理二 設(shè)(H1)-(H4)成立,則系統(tǒng)(1)的偽概周期解是全局指數(shù)穩(wěn)定的。
參考文獻(xiàn):
[1] H.Zhang, Existence and stability of almost periodic solutions for CNNs with continuously distributed leakage delays[J]. Neural Computing & Applications,2014(5):1135-1146.
[2] CHANGJIN XU,QIMING ZHANG,YUSEN WU. Existence and stability of pseudo almost periodic solutions for shunting inhibitory cellular neural network with neutral type delays and time-varying leakage delays[J].Computation in Neural Systems,2014(4):168-192.
[3] Bingwen Liu. Pseudo almost periodic solutions for neutral type CNNs with continuously distributed leakage delays[J]. Neurocomputing,2015,148:445-454.
[4] Y.Li, C.Wang, Almost periodic functions on time scales and applications[J]. Discrete Dynamics in Nature & Society, 2011(10):1095-1114.
[5] Hui Zhou,Zongfu Zhou,Wei Jiang. Almost periodic solutions for neutral type BAM neural network with distributed leakage delays on time scales[J].Neurocomputing,2015,157:223-230.
[6] 張雪梅,林清梅.時標(biāo)上一類多重變時滯Lasota-Wazewska模型的概周期解[J].曲靖師范學(xué)院學(xué)報(bào),2015(6):5-9,38.
[7] Pan Wang,Qingmei Lin,and Yongkun Li. Mean-Square Almost Periodic Solutions for Impulsive Stochastic Host-Macroparasite Equation on Time Scales[J].Discrete Dynamics in Nature and Society,2015(1):1-10.
[8] Yongkun Li Chao Wang.Pseudo almost periodic functions and pseudo almost periodic solutions to dynamic equations on time scales[J].Advances in Difference Equations, 2012(1):1-24.
Pseudo Almost Periodic Solutions for a Class of Delayed BAM Neural Networks on Time Scales
LIN Qingmei1, ZHANG Xuemei2, JIANG Jianxin1
(1.School of Mathematics, Wenshan University, Wenshan Yunnan 663099, China; 2.School of Mathematics and Information Science, Qujing Normal University, Qujing Yunnan 655011, China)
In this paper, a class of neutral type BAM neural networks model with distributed leakage delays on time scales is discussed. Some suf fi cient conditions are established for the existence and global exponential stability of the pseudo almost periodic solutions to the considered model using exponential dichotomy theory, contraction mapping principle, fi xed point theorem and Lyapunov functional method.
time scales; pseudo almost periodic solution; existence; global exponential stability
O177
A
1674-9200(2016)06-0048-05
(責(zé)任編輯 劉常福)
2016-09-28
文山學(xué)院科研基金項(xiàng)目“時標(biāo)上時滯神經(jīng)網(wǎng)絡(luò)的概周期解研究”(15WSY13)。
林清梅,女,福建泉州人,文山學(xué)院數(shù)學(xué)學(xué)院講師,碩士,主要從事微分方程研究;張雪梅,女,云南曲靖人,曲靖師范學(xué)院數(shù)學(xué)與信息科學(xué)學(xué)院講師,碩士,主要從事微分方程研究;蔣建新,男,甘肅天水人,文山學(xué)院數(shù)學(xué)學(xué)院講師,碩士,主要從事矩陣?yán)碚摷捌鋺?yīng)用研究。