电子科技大学研究生专项奖学金申请表 姓名罗金南学号 2016112 20108 学生类别 博士 硕士 年级 2016 政治面貌团员导师姓名 田文 洪 专业 软件工程 中国银行帐户 ( 即发助研助学金的帐户 ) 6216633100000328889 申请奖学金类别世强奖学金 ( 特等 ) 个人 总结 本人在读博士研究生期间思想政治上坚定拥护党和国家的路线方针政策, 具有正确的政治方向 ; 学习科研上勤奋刻苦, 学习成绩优异 科研成果显著, 获得电子科技大学学业一等奖学金和电子科技大学 优秀研究生 ; 日常生活中遵守校纪校规, 礼敬师长 友爱同学, 具有良好的道德品质和行为习惯 ; 在校期间积极参加文体活动和其他公益活动, 德智体美劳全面发展 ( 详细说明见下一页 ) 导师 意见 签字 : 年月日 学院 意见 签字 ( 学院主管研究生工作副书记 ): 年月日 研究生 院意见 签章 : 年月日
个人总结详细说明 在 2017 2018 学年中 : 一 思想方面 : 积极进取, 要求上进, 不断地提高自身政治素养, 拥护党的领导 在校期间遵守校纪校规, 尊敬师长, 团结同学, 态度端正 二 学习与科研方面 : 学习上勤奋刻苦, 正在加拿大滑铁卢大学进行为期两年的学习, 科研上积极进取, 攻坚克难, 在本学年以第一作者发表 SCI 期刊论文 3 篇,EI 会议 1 篇, 分别如下 : 1.Non fragile asynchronous event triggered control for uncertain delayed switched neural networks. Nonlinear Analysis: Hybrid Systems. vol. 29, pp. 54 73, 2018.(SCI,IF:4.01) 2.Non fragile asynchronous H control for uncertain stochastic memory systems with Bernoulli distribution. Applied Mathematics and Computation. vol. 312, pp. 109 128, 2017.(SCI,IF:2.3) 3.Improved delay probability dependent results for stochastic neural networks with randomly occurring uncertainties and multiple delays. International Journal of Systems Science. vol. 49, no. 9, pp. 2039 2059, 2018.(SCI,IF:2.185) 4.Novel delay probability distribution dependent mean square stability analysis for stochastic neural networks. Proceedings of the 36th Chinese Control Conference. 2017.(EI) 此外, 作为博士主研人参与了国家自然科学基金项目 Self Adaptive Energy Efficient Scheduling of Virtual Machines In the Cloud, 项目编号 : 61650110513; 整合大规模负载的服务器自适应优化节能调度算法, 项目编号 : 61672136 三 工作方面 : 工作认真, 关心集体, 乐于助人, 帮助身边同学解决生活和学习中遇到的各种问题, 得到老师和同学们的一致好评 当然, 仅这些是不够的, 在博士研究生剩下的时间里我会以更高的标准严格要求自己, 认真完成学校, 导师交付的每一项任务 以上是我个人的基本情况, 敬请各位领导老师审核 批准
Nonlinear Analysis: Hybrid Systems 29 (2018) 54 73 Contents lists available at ScienceDirect Nonlinear Analysis: Hybrid Systems journal homepage: www.elsevier.com/locate/nahs Non-fragile asynchronous event-triggered control for uncertain delayed switched neural networks Jinnan Luo a, Wenhong Tian a,e, Shouming Zhong b, Kaibo Shi c, *, Wenqin Wang d a School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, PR China b School of Mathematics Sciences, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, PR China c School of Information Science and Engineering, Chengdu University, Chengdu, Sichuan 610106, PR China d School of Sciences, Tianjin Polytechnic University, Tianjin 300130, PR China e Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing 400714, PR China a r t i c l e i n f o a b s t r a c t Article history: Received 3 May 2017 Accepted 31 December 2017 Available online 30 January 2018 Keywords: Non-fragile asynchronous control Bernoulli distribution Event-triggered scheme Randomly occurring uncertainties Switched neural networks This study constructs and investigates the non-fragile asynchronous event-triggered control for uncertain switched neural networks with communication delays and Bernoulli distribution. The system involves both the randomly occurring uncertainties of all parameters (ROUAPs) in the switched system model and randomly gain perturbations of the controller. By introducing several stochastic variables, a new model structure obeying Bernoulli distribution is formulated to describe the uncertainties of system and the controller. In addition, an improved non-fragile asynchronous event-triggered control scheme is proposed and a modified Lyapunov Krasovskii function (LKF) is chosen via flexible terminal method (FTM) to deal with such uncertain switched systems with delays. By combining newly bounding inequalities and convex combination method, new results are presented for the uncertain delayed switched error systems in terms of linear matrix inequality (LMI). Furthermore, the non-fragile asynchronous event-triggered controller is designed with less communication burden. Finally, a numerical example is provided to verify the merits of the proposed results. 2018 Elsevier Ltd. All rights reserved. 1. Introduction In the past several decades, neural networks have been successfully applied in many fields, such as computational optimization problems, associative memory, pattern recognition and image processing [1,2]. Considerable attention has been attracted for researching the robust stability, synchronization and parameter uncertainties of different classes of neural networks (see [3 10]). Switched systems, as a significant kind of hybrid system, composed by a family of subsystems, which subsystem is active regulated by a switching law. In fact, many physical or practical systems could be modeled as switched systems [8]. On the This work was supported by the National Natural Science Foundation of China (grant nos. 61703060, 61533006, 61603272, 61273015, 11601474, 11461082, 61672136, 6167060383 and 61650110513), Opening Fund of Geomathematics Key Laboratory of Sichuan Province (scsxdz201704), China Scholarship Council (CSC Grant No. [2017]3109), Xi Bu Zhi Guang (R51A150Z10), the Youth Fund Project of Tianjin Natural Science Foundation (16JCQNJC03900), the Scientific Research Fund of Sichuan Provincial Education Department (17ZB0459), the Research fund for International Young Scientists of National Natural Science Foundation of China (NSFC Grant No. 61550110248). * Corresponding author. E-mail address: skbs111@163.com (K. Shi). https://doi.org/10.1016/j.nahs.2017.12.006 1751-570X/ 2018 Elsevier Ltd. All rights reserved.
Applied Mathematics and Computation 312 (2017) 109 128 Contents lists available at ScienceDirect Applied Mathematics and Computation journal homepage: www.elsevier.com/locate/amc Non-fragile asynchronous H control for uncertain stochastic memory systems with Bernoulli distribution Jinnan Luo a,, Wenhong Tian a,, Shouming Zhong b, Kaibo Shi c, Hao Chen d, Xian-Ming Gu b, Wenqin Wang e a School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, PR China b School of Mathematics Sciences, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, PR China c School of Information Science and Engineering, Chengdu University, Chengdu, Sichuan 610106, PR China d College of Electrical and Information Engineering, Southwest University for Nationalities, Chengdu, Sichuan 610041, PR China e School of Sciences, Tianjin Polytechnic University, Tianjin 300130, PR China a r t i c l e i n f o a b s t r a c t Keywords: Asynchronous control Bernoulli distribution H control Randomly occurring uncertainties Uncertain stochastic memory systems This study constructs and investigates the non-fragile asynchronous H control for uncertain stochastic memory systems with Bernoulli distribution. The system not only contains the randomly occurring uncertainties of all parameters (ROUAPs) and stochastic disturbances in the system model, but also includes randomly gain perturbations in the controller. By introducing the stochastic variables, a new model structure is built obeying Bernoulli distribution to describe the system and the controller. Moreover, a modified Lyapunov Krasovskii function (LKF) is constructed, combining Itô s differential formula and free-weighting matrix method, less conservative results are presented in terms of the linear matrix inequality (LMI). Furthermore, an observer-based non-fragile asynchronous H controller is designed without any limits on the system parameters. Finally, three numerical examples are provided to demonstrate the effectiveness and feasibility of the proposed results. 2017 Elsevier Inc. All rights reserved. 1. Introduction Time-delay systems (TDSs) have been frequently appeared in many fields, such as pattern recognition, optimization problem, associative memory, image processing, etc. [1,2]. Time-delay often causes the performance degradation like instability or oscillations. Therefore, the analysis of delayed systems has received numerous attention. There are abundant results involving both delay-dependent results and delay-independent results for the stability investigations of TDSs, [3,5,7,10]. Delaydependent results make the full use of the delay information, which are less conservative than the delay-independent results. Most of earlier-established results focus on analyzing delay-dependent criteria in the case of the TDSs with control. However, the delay-dependent control results still have a certain extent of conservatism. In order to further reduce the This work was supported by the National Natural Science Foundation of China (grant nos. 61533006, 61603272, 61273015, 11601474, 11461082, 61672136 and 61650110513 ), the Youth Fund Project of Tianjin Natural Science Foundation (16JCQNJC03900), the Scientific Research Fund of Sichuan Provincial Education Department (17ZB0459). Corresponding author. E-mail addresses: jinnanluo@outlook.com (J. Luo), tian_wenhong@uestc.edu.cn (W. Tian). http://dx.doi.org/10.1016/j.amc.2017.05.003 0 096-30 03/ 2017 Elsevier Inc. All rights reserved.
International Journal of Systems Science ISSN: 0020-7721 (Print) 1464-5319 (Online) Journal homepage: http://www.tandfonline.com/loi/tsys20 Improved delay-probability-dependent results for stochastic neural networks with randomly occurring uncertainties and multiple delays Jinnan Luo, Wenhong Tian, Shouming Zhong, Kaibo Shi, Xian-Ming Gu & Wenqin Wang To cite this article: Jinnan Luo, Wenhong Tian, Shouming Zhong, Kaibo Shi, Xian-Ming Gu & Wenqin Wang (2018) Improved delay-probability-dependent results for stochastic neural networks with randomly occurring uncertainties and multiple delays, International Journal of Systems Science, 49:9, 2039-2059, DOI: 10.1080/00207721.2018.1483044 To link to this article: https://doi.org/10.1080/00207721.2018.1483044 Published online: 10 Jun 2018. Submit your article to this journal Article views: 51 View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalinformation?journalcode=tsys20
INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE 2018, VOL. 49, NO. 9, 2039 2059 https://doi.org/10.1080/00207721.2018.1483044 Improved delay-probability-dependent results for stochastic neural networks with randomly occurring uncertainties and multiple delays Jinnan Luo a, Wenhong Tian a,b, Shouming Zhong c, Kaibo Shi d, Xian-Ming Gu e and Wenqin Wang f a School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, People s Republic of China; b Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing, People s Republic of China; c School of Mathematics Sciences, University of Electronic Science and Technology of China, Chengdu, People s Republic of China; d School of Information Sciences and Engineering, Chengdu University, Chengdu, People s Republic of China; e School of Economic Mathematics/Institute of Mathematics, Southwestern University of Finance and Economics, Chengdu, People s Republic of China; f School of Sciences, Tianjin Polytechnic University, Tianjin, People s Republic of China ABSTRACT This study seeks to address the delay-probability-dependent stability problem for a new class of stochastic neural networks with randomly occurring uncertainties, neutral type delay, distributed delay and probability-distribution delay. The system not only includes the randomly occurring uncertainties of parameters (ROUPs) but also contains stochastic disturbances, which is not yet investigated in existing papers. First, several stochastic variables which obey Bernoulli distribution are introduced to describe the ROUPs, based on which a new model is built. Second, through fully considering the information on kinds of delays and utilising general delay-partitioning method, an improved Lyapunov Krasovskii function (LKF) is constructed. Combining Itô s differential formula, general bounding, free-weighting matrix and stochastic methods, a new delay-probabilitydependent robustly mean square stable criterion is formulated in terms of linear matrix inequality. Finally, two numerical examples are given to demonstrate the effectiveness of the proposed results. ARTICLE HISTORY Received 6 September 2017 Accepted 25 May 2018 KEYWORDS Probability distribution; stochastic neural networks; randomly occurring uncertainties; convex combination 1. Introduction Neural networks (NNs) have been successfully applied in many fields, such as image processing, pattern recognition, associative memory, optimisation problem and so on (Chen & Fang, 2000;Chua&Yang,1998;Zhang,He, Jiang, & Wu, 2016; Zhang& Yi, 2011). Those applications are tightly related to their dynamics behaviours, particularly stability. However, in the variable measuring, signal transmission and neural processing, it is inevitable to induce time delay (TD) which could cause instability or poor performances of NNs. Therefore, abundant references discussed the stability of the researched system with delays (Meng, Chen, & Wang, 2009; Shi et al.,2018), for instance, Takagi Sugeno fuzzy NNs with interval time-varying delays (TVDs) in Chen, Zhong, Li, Liu, and Adu-Gyamfi (2016) and Wang, Cheng, and Zhan (2017), Hopfield NNs with both discrete and distributedtvdsinli,chen,lin,andzhou(2009), bidirectional associative memory (BAM) NNs with both discrete and distributed TVDs in Wen, Du, Zhong, Xu, and Zhou (2016). However, the delay-dependent results have a certain extent of conservatism, which is usually indexed by the acceptable delay region provided by the corresponding criteria (Zhang, Wang, & Liu, 2014). In order to reduce the conservatism of the delay-dependent stability conditions, many methods are proposed, such as a new Lyapunov Krasovskii function (LKF) constructed in Wang, Zhang, Cheng and Park (2018), Zhang and Yu (2012) and Wang, Cheng, Al-Barakati, and Fardoun (2017), free-weighting matrix in He, Ji, Zhang, and Wu (2016), delay-partitioning techniques in Tian, Xiong, and Xu (2014), Chen et al. (2016) and Shi, Liu,Tang,Zhu,andZhong(2016), and relaxed integral inequalitiesinwang,yan,chengandzhong(2017), Li, Bai, Huang, and Cai (2016) and Zhang, He, Jiang, Wu, andzeng(2016). Although lots of approaches have been utilised for reducing the conservative of the criteria, most of the results only concentrated on the deterministic TDs about the corresponding networks and TD mostly occurs in a CONTACT Jinnan Luo jinnanluo@outlook.com School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, People s Republic of China; Wenhong Tian tianwenhong@cigit.ca.cn School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, People s Republic of China; Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing, People s Republic of China 2018 Informa UK Limited, trading as Taylor & Francis Group
Proceedings of the 36th Chinese Control Conference July 26-28, 2017, Dalian, China Novel delay-probability-distribution-dependent mean square stability analysis for stochastic neural networks LUO Jinnan 1, TIAN Wenhong 1, ZHONG Shouming 2, SHI Kaibo 3, WANG Wenqin 4 1. School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, P.R. China E-mail: jinnanluo@outlook.com 2. School of Mathematical Sciences, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, P.R. China E-mail: zhongsm@uestc.edu.cn 3. School of Information Science and Engineering, Chengdu University, Chengdu, Sichuan 610106, P.R. China E-mail: skbs111@163.com 4. School of Sciences, Tianjin Polytechnic University, Tianjin 300130, P.R. China E-mail: wenqinwang123@163.com Abstract: In this paper, the delay-probability-distribution-dependent stability problem for stochastic neural networks with various time-varying delays is investigated. By considering a new Lyapunov-Krasovskii functional (LKF) concluding more delaypartitioning term and combining free weight matrices and stochastic processing techniques, an improved delay-probabilitydistribution-dependent condition is built in terms of linear matrix inequality (LMI) so that the system is mean square stable. Finally, a numerical example is given to verify the effectiveness the proposed criterion. Key Words: Stochastic neural networks; Probability-distribution; Multiple delays. 1 Introduction In recent decades, the neural networks have been extensive applied in a variety of areas, such as associative memory, pattern recognition, image processing, optimization problem and many other fields [1,2,14,15], which motivate many researchers paying lots of attention on the neural networks. However, due to the variable measuring not in time, signal transmission delayed and other factors, the time delay is inevitably encountered in the neural networks and it could make the system poor performance or instability. Therefore, the analysis of many aspects for delayed neural networks have been studied and numerous significant results have been published in [3-5]. On the one hand, the synaptic transmission is a noisy process brought on by random fluctuation from the release of neuron transmitters and other probabilistic causes in real nervous systems [6], which also results in the disability or oscillation of the systems. Hence, the stochastic neural networks have drawn considerable attentions [7,8]. On the other hand, time delays largely occur in a random form, which named stochastic delays. In this situation, the time delay is very large and the probabilities of the delays appearing are sufficiently small, if only the deterministic time delays are considered, the results will be more conservative. So, it is essential and important to study the delay-probabilitydistribution-dependent problem for stochastic neural networks. For example, in [9], the delay-distribution-dependent state estimation problem is investigated for the discretetime stochastic neural networks with random delay. In [10], the authors study the delay dependent-probabilitydistribution globally robustly asymptotically stable for neutral type stochastic neural networks by constructing a proper LKF. But, there is still room to further create a certain more efficacious ways and obtain the less conservative results. This paper mainly focuses on studying the problem of mean square stable for stochastic neural networks, which aims at obtaining new less conservative and effective method. By introducing the probability-distribution delay and making the most of the information of the all delays, an appropriate LKF with delay-partition term is constructed. Moreover, an improved delay-probabilitydistribution-dependent condition is built by the Itô s differential formula and the stochastic methods. In addition, more free weight matrices are utilized. Compared with the existing method in [10], the method presented in this paper is less conservative. Finally, the feasibility and merits of the proposed method is shown by the numerical simulation example. Notation: Throughout this paper, R n, are the n- dimensional Euclidean space, Euclidean norm, separately. R m n denotes the set of all m n dimensional matrices. I n, T, stand for the n n identity matrix, transposition, symmetric block, respectively. X>0 (X 0) means that the matrix is symmetric positive definite (positive semi-definite). E{} is the mathematical expectation operator with respect to the given probability measure P and (Θ,F,{F t } t 0,P) denotes a complete probability space with a filtration {F t } t 0 satisfying the usual conditions. He{F } = F + F T and diag{} means the diagonal matrix. 2 Notation and Background In this paper, we consider the following stochastic neural networks: d(x(t)) = d[fx(t ϱ)]+[ Gx(t)+Jg(x(t))+ Kg(x(t τ(t))) + L t t δ g(x(s))ds]dt +[Sx(t)+Ex(t τ(t))]dw(t) x(t) =Φ(t), t [ h, 0], h =2max{ϱ, τ, δ} where x(t) = [x 1 (t),x 2 (t),...,x n (t)] T R n, g(x( )) = [g 1 (x 1 ( )), g 2 (x 2 ( )),..., g n (x n ( ))] T R n (1) 1862