基于新型联想记忆神经网络的非线性系统辨识   总被引：1，自引：1，他引：0  

吴学礼  孟凡华  王永骥  李平  贾辉然  孟华 《地理与地理信息科学》2004,20(4):110-112

Hopfield网络模型具有联想存储器功能,但对系统辨识不适用。具有动态记忆功能的Elman神经网络的泛化能力比较低。该文提出了一种新型联想记忆神经网络结构和学习算法,通过引入联想记忆衰减因子,提高了对非线性系统的辨识能力。通过与Elman动态神经网络辨识方法的仿真比较,说明联想记忆神经网络辨识方法具有很好的动态辨识能力和泛化能力。  相似文献   

Iron abundances from optical Fe iii absorption lines in B-type stellar spectra     

H. M. A. Thompson  F. P. Keenan  P. L. Dufton  C. Trundle  R. S. I. Ryans  P. A. Crowther 《Monthly notices of the Royal Astronomical Society》2008,383(2):729-740

  相似文献   

基于MPSO的RBF耦合算法的桩基动测参数辨识     

郭健  王元汉  苗雨 《岩土力学》2008,29(5):1205-1209

变异粒子群优化算法（MPSO）是一种基于群体智能的改进全局优化技术,其优势在于减小陷入局部极值的机率,增加全局搜索能力。将变异粒子群算法与径向基函数（RBF）神经网络结构进行结合,建立了变异粒子群神经网络（MPSO-RBF）耦合算法,充分发挥了MPSO算法的全局寻优能力和RBF算法的局部搜索优势。数值计算结果表明,所建立的方法能够对桩基动测进行多参数的识别和非线性优化问题的求解,具有良好全局收敛能力,是一种行之有效的智能算法。  相似文献   

Output‐only structural identification in time domain: Numerical and experimental studies     

M. J. Perry  C. G. Koh 《地震工程与结构动力学》2008,37(4):517-533

By identifying changes in stiffness parameters, structural damage can be detected and monitored. Although considerable progress has been made in this research area, many challenges remain in achieving robust structural identification based on incomplete and noisy measurement signals. The identification task is made even more difficult if measurement of input force is to be eliminated. To this end, an output‐only structural identification strategy is proposed to identify unknown stiffness and damping parameters. A non‐classical approach based on genetic algorithms (GAs) is adopted. The proposed strategy makes use of the recently developed GA‐based method of search space reduction, which has shown to be able to accurately and reliably identify structural parameters from measured input and output signals. By modifying the numerical integration scheme, input can be computed as the parameter identification task is in progress, thereby eliminating the need to measure forces. Numerical and experimental results demonstrate the power of the strategy in accurate and efficient identification of structural parameters and damage using only incomplete acceleration measurements. Copyright © 2007 John Wiley & Sons, Ltd.  相似文献   

Reconciling theory with observations: elements of a diagnostic approach to model evaluation   总被引：2，自引：0，他引：2  

Hoshin V. Gupta  Thorsten Wagener  Yuqiong Liu 《水文研究》2008,22(18):3802-3813

This paper discusses the need for a well‐considered approach to reconciling environmental theory with observations that has clear and compelling diagnostic power. This need is well recognized by the scientific community in the context of the ‘Predictions in Ungaged Basins’ initiative and the National Science Foundation sponsored ‘Environmental Observatories’ initiative, among others. It is suggested that many current strategies for confronting environmental process models with observational data are inadequate in the face of the highly complex and high order models becoming central to modern environmental science, and steps are proposed towards the development of a robust and powerful ‘Theory of Evaluation’. This paper presents the concept of a diagnostic evaluation approach rooted in information theory and employing the notion of signature indices that measure theoretically relevant system process behaviours. The signature‐based approach addresses the issue of degree of system complexity resolvable by a model. Further, it can be placed in the context of Bayesian inference to facilitate uncertainty analysis, and can be readily applied to the problem of process evaluation leading to improved predictions in ungaged basins. Copyright © 2008 John Wiley & Sons, Ltd.  相似文献   

小波分析在结构损伤识别中的应用   总被引：1，自引：0，他引：1  

刘涛  李爱群  丁幼亮 《地震工程与工程振动》2008,28(2):29-35

为提高结构损伤识别方法的准确性和适用性,将小波分析引入到结构损伤识别中。本文首先介绍了小波分析的基本原理,然后详细论述了小波分析用于结构损伤识别的三种方法:基于时域响应的方法、基于空间域响应的方法和小波分析与其他方法联合使用,接着分析了各种方法的应用现状、适用范围和存在的问题。通过比较可以看出小波分析用于结构损伤识别有着广阔的应用前景,本文最后针对进一步研究的方向提出了五点建议。  相似文献   

一种迭代算法在结构物理参数识别中的应用     

王祥建  崔杰 《地震工程与工程振动》2008,28(1):44-50

本文在利用复合反演方法进行剪切型结构物理参数识别过程中,引入一种迭代算法来求解关于参数的病态方程组,探讨了几种比例阻尼形式的比例系数识别的处理方法。以10层剪切型框架结构为例,利用该迭代算法研究了无阻尼、三种比例阻尼形式以及噪信比为5％、10％的高斯白噪声等情形下的参数识别问题。结果表明,该迭代算法能很好地求解参数的病态方程组,且对四种阻尼形式和两种噪信比的噪声有很好的适应性。  相似文献   

基于多重测点数目标函数族的结构损伤识别方法     

张立涛  李兆霞  费庆国 《地震工程与工程振动》2008,28(6)

基于加速度时域测试数据进行结构损伤识别计算时,所用测试数据的点数必须足够多才能够使识别有效,但往往又容易出现收敛到局部极小解的情况。为解决这一问题,本文提出了基于多重测点数目标函数族的结构损伤识别方法;所谓多重测点数目标函数族,即由不同点数的测试数据出发构造一族目标函数,以取代传统的基于单一点数的目标函数;迭代计算时采用了：Tikhonov正则化技术以抑制解答的病态性。以Benchmark结构为算例,分析了两种基于单一点数的损伤识别计算难以收敛到正确解答的原因;并考证了文中提出的方法。计算结果表明：基于多重测点数目标函数族的结构损伤识别方法,使得识别计算易于收敛到正确解答;从而证明了本文提出的方法是有效的。  相似文献   

2007年宁洱6.4级地震研究     

刘翔  吴国华  陈慧  刘学军  付虹  毛慧玲 《地震》2008,28(4):136-144

根据云南宁洱地区地质构造、 历史强震资料、 云南监测中心区域台及现场流动台数字化观测台网测定的2007年宁洱6.4级地震序列资料及1970年以来宁洱地区6级地震序列等, 深入分析研究了宁洱地区中强地震活动规律、 宁洱6.4级地震的余震时空展布以及参数变化等特征, 并利用2007年宁洱6.4级地震地震现场数字化观测流动台测定的余震波资料研究了该序列地震波动力学特征。 研究结果认为, 宁洱6级地震活动在空间上具有较高的原地重复性, 在时间上具有强弱交替活动的韵律; 2007年6月3日宁洱6.4级地震的余震展布方向与穿过震区NW向宁洱断裂的走向大致一致, 宁洱6.4级地震可能是由无量山断裂活动构造加剧所引起;根据2007年6月3日宁洱6.4级地震序列及历次宁洱6级地震序列时序发展特征分析研究得出, 宁洱6.4级地震序列能量主要集中在震后3天内释放, 4.9级以上余震发生在主震当天, 地震序列早期参数和3.0级以上余震的垂直方向P波最大振幅与S波最大振幅非线性关系等显示, 宁洱6.4级地震序列具有较明显的主-余震型特征。  相似文献   

Toxicity of a secondary-treated sewage effluent to marine biota in Bass Strait, Australia: development of action trigger values for a toxicity monitoring program     

Adams MS  Stauber JL  Binet MT  Molloy R  Gregory D 《Marine pollution bulletin》2008,57(6-12):587-598

Melbourne Water's Eastern Treatment Plant (ETP) produces a secondary-treated sewage effluent which is chlorinated and discharged into Bass Strait at Boags Rocks, Victoria, Australia. Disappearance of the sensitive brown seaweed Hormosira banksii from rock platforms immediately adjacent to the shore-line discharge was identified in the early 1990s. Subsequently, Melbourne Water and CSIRO undertook an environmental impact assessment and review of land and marine effluent disposal options, which included ambient water quality monitoring, biological monitoring, bioaccumulation studies and toxicity testing of existing effluent to assess the nature and magnitude of the environmental effects. This paper presents data from the toxicity monitoring programs since 2001. Chronic toxicity testing using macroalgal germination and cell division (H. banksii), microalgal growth rate (Nitzschia closterium) and scallop larval development (Chlamys asperrima), confirmed that ammonia was the major cause of effluent toxicity. Results from this toxicity monitoring program were used to develop action trigger values for toxicity for each species, which were then used in a refined monitoring program in 2005-2007. An upgrade of the ETP is in progress to improve nitrification/denitrification in order to reduce ammonia concentrations and the toxicity of the effluent. Toxicity testing with a simulated upgraded effluent confirmed that ammonia concentrations and toxicity were reduced. Estimated "safe" dilutions of effluent, calculated using species sensitivity distributions, decreased from 1:140-300 for existing ETP effluent to 1:20 for nitrified effluent, further confirming that treatment improvements should reduce the impact on marine biota in the vicinity of the discharge.  相似文献