用户名: 密码: 验证码:
地震作用下结构响应的预测及模糊预测控制
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
本文研究了传统预测的方法和局限性,以及模糊神经网络的优良性能。在
    传统预测理论中添加了一种新型模糊Modular神经网络模型,形成了基于模糊
    神经网络的预测方法。它综合考虑了结构各种模糊因素的影响,克服了传统预
    测方法的缺点,具有基于对象的物理特性,而非其数学模型的优点。接着对某
    一建筑结构的地震响应进行了预测。这样的在线实时预测地震下结构各层响应,
    可为建筑结构的主动控制提供较为准确的优化性能指标,因此可为实现在线实
    时控制结构响应提供优良的保证。
     在研究过程中,首先建立模糊神经网络的模型,确定网络的参数,然后将
    用Wilson-θ法计算出的各层前2秒的响应值作为样本输入到网络中训练和学
    习,最后对2秒后的响应进行预测。预测结果显示,本文所用的模糊Modular
    神经网络具有良好的预测功能,预测的精度高,花费时间少。文中还讨论了在
    网络参数不同时预测精度和花费时间的变化。基于本文提出的模糊神经网络预
    测控制系统的初步应用,本文对单自由度的结构进行了预测控制。结果表明,
    该模糊神经预测控制系统具有良好的性能和特点,控制效果令人满意。
In this paper, we discussed method and localization of the traditional prediction, and
     studied the fuzzy neural network excellent capability. A new fuzzy Modular neural
     network is added to the traditional prediction thus the prediction based on the fuzzy neural
     network is formed. It considers the effects of diversified fuzzy factors on the structure
     synthetically, so it can get over the disadvantage of the traditional prediction, and it is based
     on the physics characteristic of the object but not the object mathematics model. At last
     this method is developed for on-line prediction of the response of a certain building. Such
     prediction can provide the standard of more exact optimize capability to the active control
     on the architectural structure.
    
     In researching progress, first the model of fuzzy neural network is formed and its
     parameters are decided. Then the responses of first 2 seconds calculated using Wilson ?0
     method are input and trained in the network, finally the network predicts the rest responses
     after first 2 seconds. The results show that the fuzzy Modular neural network has some
     excellent function of prediction. It has high precision, and it takes short time. In the paper
     we also discussed the variety of the precision and calculating time while the network
     parameters are changed. Finally we use the predictive control system of fuzzy neural
     network from this paper to process active control on some SDQF structures. The last result
     indicates that this predictive control system of fuzzy neural system has a series of good
     capabilities.
引文
1.Kobri, T., Koshika, N., Yamada, K. Ikeda, Y.: Seismic Response Controlled Structure with Active Mass Drive System, Part Ⅰ: Design, Part Ⅱ: Verfication, Earthquake Engrg. and Struct. Dynamics, Vol. 20, 1991, 133—166
    2.Chang, I. C. H., soong, T. T.: Structural Control Using Active Tuned Mass Damper, J. Of Engrg. Mech. ASCE Vol. 106, No. 6, 1980, 1091—1098
    3.Roorda, J.: Tendon Control in Tall Structures, J. Of struct. ASCE Vol. 101, Mar. 1975, 505-521
    4.Chung, L.L., Lin, R.C., Soong, T.T., Reinhorn, A.M.: Experimental Study of Active Control of MDOF Structures under Seismic Excitations, J. Of Engrg. Mech. ASCE Vol. 115, No. 8, 1989, 1609-1627
    5.赵耀江、姚民笃,神经元网络智能预测控制,太原工业大学学报,Vol.26,No.4,1995,62-66
    6.Yang, J.N., Akbarpour, A., Ghaemmaghamic, P.: New Optimal Control Algorithm for Structural Control, J. Of Engrg. Mech. Vol. 113, 1987, 1369-1386
    7.舒迪前,预测控制系统及其应用,机械工业出版社,1996年
    8.王伟、杨建军,广义预测控制:理论、算法和应用,控制理论与应用,第14卷,第6期,777~786,1997年
    9.席裕庚、许晓鸣、张钟俊,预测控制的研究现状和多层智能预测控制,控制理论及应用,第6卷,第2期,1~7页,1989年
    10.席裕庚,预测控制,国防工业出版社,1993年
    11.Adeli H, Yeh C. Perception Learning in Engineering Design. Microcomputers in Civil Engineering, 1989, 4(4): 247-256
    12.Vanluchene D, Sun R. Neural Networks in Structure Engineering. Microcomputers in Civil Engineering, 1990, 5(3): 207-215
    13.Hung S, Adeli H. A Model of Perception Learning with a Hidden Layer for Engineering Design. Neurocomputing, 1991, 3(1): 3-14
    14.Ghaboussi. J, Garrett J. H, Wu X. Knowledge-Based Modeling of Material Behavior and Neural Networks. J. of Engineering Mechance, 1991, 117(1): 132-153
    15.Wu X, Ghaboussi J and Garrett J H. Use of neural Networks in Detection of Structural Damage.Computers and Structures, 1992, 42(4): 649-659
    
    
    16.Masri S F, Chassiakos A G, Ganghey T K. Identification of Nonlinear Dynamic Systems Using Neural Networks. J. Of Applied Mechanics, 1993, 60:123-133
    17.Pandey P C and Barai S V. Nonlinear Analysis of Plates Using Artificial Neural Networks. J. Of Structural Engineering, 1994, 21(1): 65-78
    18.Hegazy T, Fazio P, Moselhi O. Developing Practical Neural Network Application Using Backpropagation. Microcomputers in Civil Engineering, 1994, 9(2): 145-159
    19.Hung S L, Adeli H. Parallel Back-propagation Learning Algorithm on Cray YMP8/864 Supercomputer. Neurocomputing, 1993, 5(5): 287-302
    20.王士同,神经模糊系统及其应用,北京航空航天大学出版社,1998,6
    21.李人厚、张平安等译校,精通MATLAB:综合辅导与指南,西安交通大学出版社,1998,1
    22.徐宜桂、史铁林、杨叔子、周轶尘,BP神经网络及其在结构动力学分析中的应用研究,计算力学学报,Vol.15,No.2,1998,5,210-216
    23.应行仁,采用BP神经网络记忆模糊规则控制,自动化学报,1991
    24.张玲、张钹,神经网络中BP算法的分析,模式识别与人工智能,Vol.7,1994,9,191-195
    25.刘曙光、郑崇勋、刘明远,前馈神经网络中的反向传播算法及其改进:进展与愿望,计算机科学,Vol.23,No.1,1996,76-79
    26.刘增良、刘有才,模糊逻辑与神经网络-理论研究与探索,北京航空航天大学出版社,1996,5
    27.韦岗、贺前华、欧阳景正,关于多层感知器的函数逼近能力,信息与控制,1996,6,321-324
    28.Hong Qin & Zhenya Ha, Variable Step BP Algorithm which prunes away redundant connections dynamically, IJCNN' 92Beijing, Vol. Ⅱ 1992, 441-445
    29.邓志东、孙增圻,利用线性再励的自适应变步长快速BP算法,模式识别与人工智能,Vol.6,No.4,Dec,1993,319-323
    30.工科俊、王克成、李国斌,几种变学习率的快速BP算法较研究,神经网络理论与应用研究,96,223-227
    31.王耀南,智能控制系统:模糊逻辑、专家系统、神经网络控制,湖南大学出版社,1996,10
    32.王士同等,基于模糊Modular神经网络建模研究,舰船科学技术,No.5,1998
    33.王士同等,高木—关野模糊系统,Pi-sigma神经网络及其建模应用,电子科学学刊,No.12,1995
    34.李邦宪等,动态系统预测的多层递阶方法,气象出版社,1996,2
    
    
    35.赵建、邬永革、黄炯等,基于神经网络的股市预测,计算机研究与发展,1996,9,692-697
    36.李文祥,多层递阶在产量预报中的应用,黑龙江气象,No.4,1987
    37.李邦宪,带有周期分量的多层递阶预报模型,成都气象学院学报,No.2,1989
    38.毕义明、李景文、王莲芬,模糊预不测模型,模糊技术与应用选编(1),北京航空航天大学出版社,1997,2,394-398
    39.吴雅、雷鸣、杨叔子,基于参数模型的智能预测系统及其应用,振动工程学报,Vol.6,No.1,1993,3,35-41
    40.舒迪前,预测控制系统及其应用,机械工业出版社,1998,6
    41.王耀南,智能控制系统,湖南大学出版社,1996,10
    42.戴冠中等,一种新的优化搜索算法—遗传算法,控制理论与应用,1995,12,265-271
    43.王凌云、林建华,基于模糊神经网络的结构非模型型主动控制,《计算力学学报》,1998,2:

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700