·832· 智能系统学报 第12卷 彼长的辩证统一关系。为该领域的学术研究者提供 pitching robot based on fuzzy PID control[J].CAAI trans- 了理论推导、分析方法和仿真对比实例，为工程应 actions on intelligent systems,2015,10(3):399-406 用领域提供了可以基于经典系统升级改造的、可靠 [14刘经纬，王普，等.专家模糊增量式自适应的参数在线整 的智能控制系统解决方案。 定优化系统及方法P1.201110023946.6.2012-11-07 [15]王普，刘经纬，等.自适应小波神经网络异常检测故障诊 参考文献： 断分类系统及方法P].201110023943.2,2012-08-30. [16]SHARIFIAN M BB,MIRLO A,TAVOOSI J,SABAHI [1]VOLODYMYR M,KORAY K,DAVID S.Human-level M.Self-adaptive RBF neural network PID controller in lin- control through deep reinforcement learning[J].Nature, ear elevator[Cl//International Conference on Electrical Ma- 2015.518:529-533. chines and Systems.Beijing,China,2011:1-4. [2]RIEDMILLER M,GABEL T,HAFNER R,LANGE S.Re- [17]NIE Yanmin,HE Zhiqiang.Optimization of the main inforcement learning for robot soccer[J].Robots,2009.27: 55-73. steam temperature PID parameters based on improve BP [3]DIUK C,COHENA,LITTMAN,M L.An object-oriented neural network[C]//International Conference on Simula- tion and Modeling Methodologies,Technologies and Ap- representation for efficient reinforcement learning[J].Mach learn,2008(1):240-247. plications.Rome,Italy,2015:113-116. [4]LECUN Y,BENGIO Y,HINTON G.Deep learning[J]. [18]CHEN Zhe,FENG Tianjin,et al.The application of wave- Nature,2016,12(21:436-444. let neural network for time series predictionand system [5]HINTON G.Learning multiple layers of representation[J]. modeling based on multiresolution learning[C]//Interna- Trends in cognitive sciences,2007,11(5):428-434. tional Conference on System,Man and Cybernetics. [6]KRIZHEVSKY A,SUTSKEVER I,HINTON GE.Image- Tokyo.Japan.1999(1)425-430. net classification with deep convolutional neural networks [19]LOUSSIFI H,NOURI K.BRAIEK N B.A new efficient [J].Neural information processing systems foundation, hybrid intelligent method for nonlinear dynamical systems 2012,47:777-780 identification:the wavelet kernel fuzzy neural network[J]. 7]FARABET C,COUPRIE C,NAJMAN L,LECUN Y. Communications in nonlinear science and numerical simu- Learning hierarchical features for scene labeling[J].IEEE lation,2016,32:10-30 transactions on pattern analysis and machine intelligence, [20]SILVA G J,DATTA A,BHATTACHARYYA S P.New 2013,35(8):1915-1929 results on the synthesis of PID controllers[J].IEEE trans on 8]SUSANTO-LEE R.FERNANDO T.SREERAM V.Simu- automatic sontrol,2002,47(2):241-252. lation of fuzzy-modified expert PID algorithms for blood 作者简介： glucose control[C]//10th International Conference on Con- trol,Automation,Robotics and Vision.Hanoi,Vietnam, 刘经纬，男，1982年生，副教授 2008:1583-1589 博士，主要研究方向为智能控制与智 9]XUE Ping,WANG Haichao,HOU Juanjuan.Based on the 能系统。发表学术论文、专利20余 篇，研究成果获国家级科技竞赛3项 fuzzy PID brushless DC motor control system design[C]// 奖励。参加多项国家级、省部级自然 International Conference on Measurement,Information and 科学基金项目。 Control.Harbin,China,2012:703-706. [10]OU Kai,WANG Yaxiong,LI Zhenzhe.Feedforward fuzzy-PID control for air flow regulation of PEM fuel cell 赵辉，男，1988年生，助理研究 system [J].International journal of hydrogen energy,2015, 员，博士后，主要研究方向为人工智能 40(35):11686-11695. 轨道交通智能控制。发表学术论文 [11]SHI Hongbo,HUANG Chuang.A BP wavelet neural net- 10余篇。 work structure for process monitoring and fault detection [C]//The Sixth World Congress on Intelligent Control and Automation.Dalian,China,2006(2):5675-5681 [12]SHARIFIAN M BB:MIRLO A,TAVOOSI J,SABAHI 周瑞，女，1983年生，讲师，博士， M.Self-adaptive RBF neural network PID controller in lin- 主要研究方向为过程控制。主持国家 ear elevator[C]//International Conference on Electrical Ma- 基金1项，完成北京市基金2项，作为 chines and Systems.Beijing,China,2011:1-4. 骨干成员参与国家自然科学基金 [13]赵新华，王璞，陈晓红.智能投球机器人模糊PID控制 3项.发表学术论文10余篇。 U.智能系统学报，2015,10(3)：399-406. ZHAO Xinhua,WANG Pu,CHEN Xiaohong.Intelligent彼长的辩证统一关系。为该领域的学术研究者提供 了理论推导、分析方法和仿真对比实例，为工程应 用领域提供了可以基于经典系统升级改造的、可靠 的智能控制系统解决方案。 参考文献： VOLODYMYR M, KORAY K, DAVID S. Human-level control through deep reinforcement learning[J]. Nature, 2015, 518: 529–533. [1] RIEDMILLER M, GABEL T, HAFNER R, LANGE S. Re￾inforcement learning for robot soccer[J]. Robots, 2009, 27: 55–73. [2] DIUK C, COHENA, LITTMAN, M L. An object-oriented representation for efficient reinforcement learning[J]. Mach learn, 2008(1): 240–247. [3] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2016, 12(21): 436–444. [4] HINTON G. Learning multiple layers of representation[J]. Trends in cognitive sciences, 2007, 11(5): 428–434. [5] KRIZHEVSKY A, SUTSKEVER I, HINTON GE. Image￾net classification with deep convolutional neural networks [J]. Neural information processing systems foundation, 2012, 47: 777–780. [6] FARABET C, COUPRIE C, NAJMAN L, LECUN Y. Learning hierarchical features for scene labeling[J]. IEEE transactions on pattern analysis and machine intelligence, 2013, 35(8): 1915–1929. [7] SUSANTO-LEE R, FERNANDO T, SREERAM V. Simu￾lation of fuzzy-modified expert PID algorithms for blood glucose control[C]//10th International Conference on Con￾trol, Automation, Robotics and Vision. Hanoi, Vietnam, 2008: 1583–1589. [8] XUE Ping, WANG Haichao, HOU Juanjuan. Based on the fuzzy PID brushless DC motor control system design[C]// International Conference on Measurement, Information and Control. Harbin, China, 2012: 703–706. [9] OU Kai, WANG Yaxiong, LI Zhenzhe. Feedforward fuzzy-PID control for air flow regulation of PEM fuel cell system [J]. International journal of hydrogen energy, 2015, 40(35): 11686–11695. [10] SHI Hongbo, HUANG Chuang. A BP wavelet neural net￾work structure for process monitoring and fault detection [C]//The Sixth World Congress on Intelligent Control and Automation. Dalian, China, 2006(2): 5675–5681. [11] SHARIFIAN M B B; MIRLO A, TAVOOSI J, SABAHI M. Self-adaptive RBF neural network PID controller in lin￾ear elevator[C]//International Conference on Electrical Ma￾chines and Systems. Beijing, China, 2011: 1–4. [12] 赵新华, 王璞, 陈晓红. 智能投球机器人模糊 PID 控制 [J]. 智能系统学报, 2015, 10(3): 399–406. ZHAO Xinhua, WANG Pu, CHEN Xiaohong. Intelligent [13] pitching robot based on fuzzy PID control[J]. CAAI trans￾actions on intelligent systems, 2015, 10(3): 399–406. 刘经纬, 王普, 等. 专家模糊增量式自适应的参数在线整 定优化系统及方法[P]. 201110023946.6, 2012-11-07. [14] 王普, 刘经纬, 等. 自适应小波神经网络异常检测故障诊 断分类系统及方法[P]. 201110023943.2, 2012-08-30. [15] SHARIFIAN M B B, MIRLO A, TAVOOSI J, SABAHI M. Self-adaptive RBF neural network PID controller in lin￾ear elevator[C]//International Conference on Electrical Ma￾chines and Systems. Beijing, China, 2011: 1–4. [16] NIE Yanmin, HE Zhiqiang. Optimization of the main steam temperature PID parameters based on improve BP neural network[C]//International Conference on Simula￾tion and Modeling Methodologies, Technologies and Ap￾plications. Rome, Italy, 2015: 113–116. [17] CHEN Zhe, FENG Tianjin, et al. The application of wave￾let neural network for time series predictionand system modeling based on multiresolution learning[C]//Interna￾tional Conference on System, Man and Cybernetics. Tokyo, Japan, 1999(1): 425–430. [18] LOUSSIFI H, NOURI K, BRAIEK N B. A new efficient hybrid intelligent method for nonlinear dynamical systems identification: the wavelet kernel fuzzy neural network[J]. Communications in nonlinear science and numerical simu￾lation, 2016, 32: 10–30. [19] SILVA G J, DATTA A, BHATTACHARYYA S P. New results on the synthesis of PID controllers[J]. IEEE trans on automatic sontrol, 2002, 47(2): 241–252. [20] 作者简介： 刘经纬，男，1982 年生，副教授， 博士，主要研究方向为智能控制与智 能系统。发表学术论文、专利 20 余 篇，研究成果获国家级科技竞赛 3 项 奖励。参加多项国家级、省部级自然 科学基金项目。 赵辉，男，1988 年生，助理研究 员，博士后，主要研究方向为人工智能 轨道交通智能控制。发表学术论文 10 余篇。 周瑞，女，1983 年生，讲师，博士， 主要研究方向为过程控制。主持国家 基金 1 项，完成北京市基金 2 项，作为 骨干成员参与国家自然科学基金 3 项，发表学术论文 10 余篇。 ·832· 智 能 系 统 学 报 第 12 卷