基于在线神经网络算法的结构恢复力预测方法
|
摘要
为实现在线预测非线性结构恢复力,提出一种在线BP神经网络算法。针对两个自由度非线性结构,进行了Bouc-Wen模型的恢复力预测,验证了在线神经网络算法的有效性。分析了样本数量以及目标误差对预测精度和计算耗时的影响。结果表明:与传统离线学习方式算法相比,在线学习算法提高了算法计算效率和预测精度;随着样本数量的减少,算法计算效率增快;随着目标误差减小,算法预测精度增大。
This paper is motivated by the need for on-line prediction of restoring force of nonlinear structure and proposes an online BP neural network algorithm. The algorithm research is comprised of the prediction of the restoring force of the Bouc-Wen model in response to a two-degree-of-freedom nonlinear structure; the validation of the effectiveness of the proposed online neural network algorithm;and the analysis of the influence of sample size and target error on prediction accuracy and time consumption of the algorithm. The results show that the proposed online learning algorithm boasts a greater calculation efficiency and accuracy than conventional off-line learning algorithm; an decrease in sample numbers is followed by an increase in computation efficiency of the algorithm; and a decrease in the target errors is followed by the prediction accuracy of the algorithm.
This paper is motivated by the need for on-line prediction of restoring force of nonlinear structure and proposes an online BP neural network algorithm. The algorithm research is comprised of the prediction of the restoring force of the Bouc-Wen model in response to a two-degree-of-freedom nonlinear structure; the validation of the effectiveness of the proposed online neural network algorithm;and the analysis of the influence of sample size and target error on prediction accuracy and time consumption of the algorithm. The results show that the proposed online learning algorithm boasts a greater calculation efficiency and accuracy than conventional off-line learning algorithm; an decrease in sample numbers is followed by an increase in computation efficiency of the algorithm; and a decrease in the target errors is followed by the prediction accuracy of the algorithm.
引文
[1]YANG W J,NAKANO Y.Substructure online test by using real time hysteresis modeling with a neural network[J].Advances in Experimental Structural Engineering,2004(38):267-274.
[2]YUN G Y,GHABOUSSI J,ELNASHAI A S.A new network-based model for hysteretic behavior of materials[J].International Journal for Numerical Methods in Engineering,2008,73(4):447-469.
[3]张健.自适应子结构拟动力试验方法[D].哈尔滨:哈尔滨工业大学,2010.
[4]王涛.基于模型更新的土木结构混合试验方法[D].哈尔滨:哈尔滨工业大学,2014.
[5]孙浩,杨智春,张玲凌.基于神经网络的振动响应趋势预测研究[J].机械科学与技术,2006,25(17):1455-1518.
[6]李书进,瞿伟廉.基于神经网络的滞变结构在线识别[J].武汉理工大学学报,2003,25(11):91-93.
[7]赵轩.基于神经网络的结构非线性地震反应仿真[D].重庆:重庆大学,2005.
[8]张廆.基于在线学习方法的BP神经网络训练算法研究[D].武汉:华中科技大学,2009.
[2]YUN G Y,GHABOUSSI J,ELNASHAI A S.A new network-based model for hysteretic behavior of materials[J].International Journal for Numerical Methods in Engineering,2008,73(4):447-469.
[3]张健.自适应子结构拟动力试验方法[D].哈尔滨:哈尔滨工业大学,2010.
[4]王涛.基于模型更新的土木结构混合试验方法[D].哈尔滨:哈尔滨工业大学,2014.
[5]孙浩,杨智春,张玲凌.基于神经网络的振动响应趋势预测研究[J].机械科学与技术,2006,25(17):1455-1518.
[6]李书进,瞿伟廉.基于神经网络的滞变结构在线识别[J].武汉理工大学学报,2003,25(11):91-93.
[7]赵轩.基于神经网络的结构非线性地震反应仿真[D].重庆:重庆大学,2005.
[8]张廆.基于在线学习方法的BP神经网络训练算法研究[D].武汉:华中科技大学,2009.