摘要
为减小对文物本体的破坏,本文基于新疆某土遗址加固保护中碳纤维楠竹锚杆锚固力原位测试试验,考虑锚杆直径、长度、倾斜角以及灌浆体强度、孔径、碳纤维缠绕间距等锚固力影响因素,利用人工神经网络(artificial neural network,ANN)的误差反向传播(back propagation,BP)算法及MATLAB人工神经网络工具箱,建立了锚固力预测的智能模型;并以原位测试所得的数据为学习样本和检验样本,验证了该方法的适用性和可行性.将训练好的网络模型进行扩展计算,基于L25(56)正交表试验理论分析了锚固力对各影响因素的敏感性,为同类加固工程的实际应用提供参考依据.
To protect against the destruction of cultural relics the artificial neural network and the toolbox of MATLAB artificial neural network （ANN） are applied to set up prediction with consideration of the bolt diameter, the bolt length, an intelligent model of the anchoring force the angle of inclination, the grouting body intensity, the aperture and the carbon fiber wrapped spacing, based on the in-situ test of the anchorage force of the CFRP （carbon fibre reinforced plastics）-bamboo bolt used in the protection of a certain earth site in Xinjiang. And by learning the samples from the in-situ test, the applicability and the feasibility of the method are checked. Based on the calculation results, the sensitivitY of influencing factors on the anchorage force of the CFRP-bamboo bolt is analyzed by using L25（5^6）orthogonal table, which may provide a reference for similar reinforcement engineering practical applications.
出处
《力学与实践》
北大核心
2013年第2期40-45,共6页
Mechanics in Engineering
基金
陕西省自然基金(2011JQ1013)
西安建筑科技大学青年科技基金(QN1239)资助项目
关键词
碳纤维楠竹锚杆
锚固力
人工神经网络
BP模型
正交试验设计
敏感性
CFRP-bamboo bolt
anchorage force
artificial neural network
BP model
orthogonal experiment design
sensitivity
