采用BP-ANN模型的梁柱平齐端板连接节点极限抗弯承载力预测研究
摘要：筛选文献中报道的143组试验数据，采用误差反向传播人工神经网络(BP-ANN)建立和训练了1个2层BP-ANN模型，对梁柱平齐端板连接节点的极限抗弯承载力进行了预测。该模型利用20个组件特征参数作为输入，以极限抗弯承载力作为输出。通过与传统机器学习算法预测结果对比，验证了方法和模型的有效性，并依据模型推导出一个实用简化的极限抗弯承载力数学表达式。统计分析结果显示：经过训练的BP-ANN模型，在测试集上的平均绝对百分比误差(MAPE)为5.28%,均方误差(MSE)为5.79×10-4。另外，对特征参数进行敏感性分析，得到了组件特征对节点极限抗弯承载力的影响程度。研究结果表明：采用BP-ANN模型能够综合考虑组件特征对节点极限抗弯承载力的影响，预测结果较为准确；该模型为梁柱连接性能评估和改进提供了智能化的解决方案，可作为数值模拟和结构试验研究的有力补充。
关键词：人工神经网络；梁柱节点；平齐端板连接；极限抗弯承载力；
尊敬的用户，本篇文章需要2元，点击支付交费后阅读
限时优惠福利：领取VIP会员
全年期刊、VIP视频免费!
全年期刊、VIP视频免费!
参考文献[1] HORNIK K,STINCHCOMBE M,WHITE H.Multilayer feedforward networks are universal approximators[J].Neural Networks,1989,2(5):359-366.
[2] CYBENKO G.Approximation by superpositions of a sigmoidal function[J].Mathematics of Control,Signals and Systems,1992,5(4):455.
[3] PARAL A,SINGHA ROY D K,SAMANTA A K.A deep learning-based approach for condition assessment of semi-rigid joint of steel frame[J].Journal of Building Engineering,2021,34:101946.
[4] CREMONA C,SANTOS J.Structural health monitoring as a big-data problem[J].Structural Engineering International,2018,28(3):243-254.
[5] HARANDIZADEH H,TOUFIGH V.Application of developed new artificial intelligence approaches in civil engineering for ultimate pile bearing capacity prediction in soil based on experimental datasets[J].Iranian Journal of Science and Technology-Transactions of Civil Engineering,2020,44(1):545-559.
[6] 宋楚平.一种改进的BP神经网络深基坑变形预测方法[J].土木工程与管理学报，2019,36(5):45-49,55.
[7] 金浏，赵瑞，杜修力.混凝土抗压强度尺寸效应的神经网络预测模型[J].北京工业大学学报，2021,47(3):260-268.
[8] 季韬，林挺伟，林旭健.基于人工神经网络的混凝土抗压强度预测方法[J].建筑材料学报，2005,8(6):677-681.
[9] 马高，刘康.基于BP神经网络CFRP约束混凝土抗压强度预测[J].湖南大学学报(自然科学版),2021,48(9):88-97.
[10] 程诗焱，韩建平，于晓辉，等.基于BP神经网络的RC框架结构地震易损性曲面分析：考虑地震动强度和持时的影响[J].工程力学，2021,38(12):107-117.
[11] 周轶凡，陆明，郭喆，等.基于遗传算法与神经网络的建筑结构找形优化研究[C]//数智营造：2020年全国建筑院系建筑数字技术教学与研究学术研讨会论文集.长沙，2020:167-170.
[12] DE LIMA L R O,DA S VELLASCO P C G,DE ANDRADE S A L,et al.Neural networks assessment of beam-to-column joints[J].Journal of the Brazilian Society of Mechanical Sciences and Engineering,2005,27(3):314-324.
[13] 沈袁协，梁诗雪.基于BP神经网络的FRP筋混凝土板抗冲切承载力预测模型[J].浙江理工大学学报(自然科学版),2022,47(3):441-451.
[14] 韩小雷，冯润平，季静，等.基于深度学习的RC梁集中塑性铰模型参数研究[J].工程力学，2021,38(11):160-169.
[15] GUZELBEY I H,CEVIK A,GÖGÜŞ M T.Prediction of rotation capacity of wide flange beams using neural networks[J].Journal of Constructional Steel Research,2006,62(10):950-961.
[16] RASHID T.Python神经网络编程[M].北京：人民邮电出版社，2018.
[17] 何宇健.Python与神经网络实战[M].北京：电子工业出版社，2018.
[18] GéRON A.机器学习实战：基于Scikit-Learn、Keras和TensorFlow[M].北京：机械工业出版社，2020.
[19] MAK L,ELKADY A.Experimental database for steel flush end-plate connections[J].Journal of Structural Engineering,2021,147(7):04721006.1-04721006.12.
[20] GLOROT X.,BENGIO Y.Understanding the difficulty of training deep feedforward neural networks[J].Journal of Machine Learning Research-Proceedings Track,2010,9:249-256.
[21] PHAM T M,HADI M N S.Predicting stress and strain of FRP-confined square/rectangular columns using artificial neural networks[J].Journal of Composites for Construction,2014,18(6):04014019.1-04014019.9.
[22] 刘龙，潘文超，吴峰崎.基于SVM的半刚性节点特性分析[C]//第25届全国结构工程学术会议论文集(第Ⅰ册).包头，2016:417-420.
[23] 冯德成，吴刚.混凝土结构基本性能的可解释机器学习建模方法[J].建筑结构学报，2022,43(4):228-238.
[24] HEARST M A,DUMAIS S T,OSUNA E,et al.Support vector machines[J].IEEE Intelligent Systems and Their Applications,1998,13(4):18-28.
[25] PEDREGOSA F,VAROQUAUX G,GRAMFORT A,et al.Scikit-iearn:machine learning in python[J].Journal of Machine Learning Research,2011,12:2825-2830.
[26] SAFAVIAN S R,LANDGREBE D.A survey of decision tree classifier methodology[J].IEEE Transactions on Systems,Man,and Cybernetics,1991,21(3):660-674.
[27] MACKAY D J C.Bayesian interpolation[J].Neural Computation,1992,4(3):415-447.
[28] GARSON G.D.Interpreting neural-network connection weights[J].AI Expert,1991,6(4):46-51.
[2] CYBENKO G.Approximation by superpositions of a sigmoidal function[J].Mathematics of Control,Signals and Systems,1992,5(4):455.
[3] PARAL A,SINGHA ROY D K,SAMANTA A K.A deep learning-based approach for condition assessment of semi-rigid joint of steel frame[J].Journal of Building Engineering,2021,34:101946.
[4] CREMONA C,SANTOS J.Structural health monitoring as a big-data problem[J].Structural Engineering International,2018,28(3):243-254.
[5] HARANDIZADEH H,TOUFIGH V.Application of developed new artificial intelligence approaches in civil engineering for ultimate pile bearing capacity prediction in soil based on experimental datasets[J].Iranian Journal of Science and Technology-Transactions of Civil Engineering,2020,44(1):545-559.
[6] 宋楚平.一种改进的BP神经网络深基坑变形预测方法[J].土木工程与管理学报，2019,36(5):45-49,55.
[7] 金浏，赵瑞，杜修力.混凝土抗压强度尺寸效应的神经网络预测模型[J].北京工业大学学报，2021,47(3):260-268.
[8] 季韬，林挺伟，林旭健.基于人工神经网络的混凝土抗压强度预测方法[J].建筑材料学报，2005,8(6):677-681.
[9] 马高，刘康.基于BP神经网络CFRP约束混凝土抗压强度预测[J].湖南大学学报(自然科学版),2021,48(9):88-97.
[10] 程诗焱，韩建平，于晓辉，等.基于BP神经网络的RC框架结构地震易损性曲面分析：考虑地震动强度和持时的影响[J].工程力学，2021,38(12):107-117.
[11] 周轶凡，陆明，郭喆，等.基于遗传算法与神经网络的建筑结构找形优化研究[C]//数智营造：2020年全国建筑院系建筑数字技术教学与研究学术研讨会论文集.长沙，2020:167-170.
[12] DE LIMA L R O,DA S VELLASCO P C G,DE ANDRADE S A L,et al.Neural networks assessment of beam-to-column joints[J].Journal of the Brazilian Society of Mechanical Sciences and Engineering,2005,27(3):314-324.
[13] 沈袁协，梁诗雪.基于BP神经网络的FRP筋混凝土板抗冲切承载力预测模型[J].浙江理工大学学报(自然科学版),2022,47(3):441-451.
[14] 韩小雷，冯润平，季静，等.基于深度学习的RC梁集中塑性铰模型参数研究[J].工程力学，2021,38(11):160-169.
[15] GUZELBEY I H,CEVIK A,GÖGÜŞ M T.Prediction of rotation capacity of wide flange beams using neural networks[J].Journal of Constructional Steel Research,2006,62(10):950-961.
[16] RASHID T.Python神经网络编程[M].北京：人民邮电出版社，2018.
[17] 何宇健.Python与神经网络实战[M].北京：电子工业出版社，2018.
[18] GéRON A.机器学习实战：基于Scikit-Learn、Keras和TensorFlow[M].北京：机械工业出版社，2020.
[19] MAK L,ELKADY A.Experimental database for steel flush end-plate connections[J].Journal of Structural Engineering,2021,147(7):04721006.1-04721006.12.
[20] GLOROT X.,BENGIO Y.Understanding the difficulty of training deep feedforward neural networks[J].Journal of Machine Learning Research-Proceedings Track,2010,9:249-256.
[21] PHAM T M,HADI M N S.Predicting stress and strain of FRP-confined square/rectangular columns using artificial neural networks[J].Journal of Composites for Construction,2014,18(6):04014019.1-04014019.9.
[22] 刘龙，潘文超，吴峰崎.基于SVM的半刚性节点特性分析[C]//第25届全国结构工程学术会议论文集(第Ⅰ册).包头，2016:417-420.
[23] 冯德成，吴刚.混凝土结构基本性能的可解释机器学习建模方法[J].建筑结构学报，2022,43(4):228-238.
[24] HEARST M A,DUMAIS S T,OSUNA E,et al.Support vector machines[J].IEEE Intelligent Systems and Their Applications,1998,13(4):18-28.
[25] PEDREGOSA F,VAROQUAUX G,GRAMFORT A,et al.Scikit-iearn:machine learning in python[J].Journal of Machine Learning Research,2011,12:2825-2830.
[26] SAFAVIAN S R,LANDGREBE D.A survey of decision tree classifier methodology[J].IEEE Transactions on Systems,Man,and Cybernetics,1991,21(3):660-674.
[27] MACKAY D J C.Bayesian interpolation[J].Neural Computation,1992,4(3):415-447.
[28] GARSON G.D.Interpreting neural-network connection weights[J].AI Expert,1991,6(4):46-51.
Research on prediction of ultimate moment capacity of beam-to-column flush end-plate connection joint via BP-ANN model
Abstract: Based on 143 sets of experimental data reported in the literatures, a two-layer back-propagation artificial neural network(BP-ANN) model was established and trained to predict the ultimate moment capacity of beam-to-column flush end-plate connection joint. The model uses 20 component features as the input and the ultimate moment capacity as the output. Compared with the prediction results of traditional machine learning algorithms, the effectiveness of the method and model were verified, and a simplified explicit expression of ultimate moment capacity was derived. The statistical analysis results show that the trained BP-ANN model has a mean absolute percentage error(MAPE) of 5.28 % and a mean square error(MSE) of 5.79×10-4 on the test set. In addition, the influence of component features on the ultimate moment capacity of joint was obtained by sensitivity analysis. The study results show that the BP-ANN model can comprehensively consider the influence of component features on ultimate moment capacity of joint, and the prediction results are more accurate. The model provides an intelligent solution for the performance evaluation and improvement of beam-to-column connections, and it can be used as a powerful supplement to the numerical simulation and structural test research of such connections.
Keywords: artificial neural network; beam-to-column connection joint; flush end-plate connection; ultimate moment capacity
860
0
0