|Table of Contents|

Reconstruction of Missing Stress Data for Steel Structure Based on Improved BP Neural Network(PDF)

《建筑科学与工程学报》[ISSN:1673-2049/CN:61-1442/TU]

Issue:
2022年04期
Page:
166-173
Research Field:
Publishing date:

Info

Title:
Reconstruction of Missing Stress Data for Steel Structure Based on Improved BP Neural Network
Author(s):
YOU Ying1;  WANG Jian1;  LIU Xue-gang2;  PENG Ning1;  ZHOU Min-feng2;  DENG Zhi-yang1
(1. School of Mechanical Engineering, Hubei University of Technology, Wuhan 430064, Hubei, China; 2. Wuhan Zhongke Kechuang Engineering Inspection Co., Ltd., Wuhan 430077, Hubei, China)
Keywords:
steel structure;  data reconstruction;  improved BP neural network;  structural health monitoring;  particle swarm algorithm;  additional momentum method
PACS:
TU391
DOI:
10.19815/j.jace.2021.11130
Abstract:
The lack of stress data in the process of steel structure health monitoring will interfere with the operation status of each link of monitoring, and cannot guarantee the safety of the construction stage. Therefore, it is very important to solve the problem of data missing. In the stress and strain monitoring of the steel structure during the construction stage, due to the influence of many complex external factors, the monitoring data is inaccurate and missing, and large deviation is between the structural value and the real value. The improved BP neural network was used to reconstruct the monitoring data of the stress and strain of the steel structure of the 300 groups and 30 groups respectively. The applicability of the data reconstruction method of the improved BP neural network was analyzed. The results show that compared with the linear regression method, the improved BP neural network method reduces the average error of discrete missing data reconstruction by 0.7%. Especially for locally missing data, the reconstruction accuracy of the neural network method is higher, and the average local error is reduced by 2.2%. In order to achieve better reconstruction accuracy, when using the improved BP neural network to reconstruct missing data, the missing rate of data should not exceed 20%. The improved BP neural network method can provide technical support for the reconstruction of missing stress data of steel structures and structural health monitoring, which has good practicability.

References:

[1] 罗尧治,梅宇佳,沈雁彬,等.国家体育场钢结构温度与应力实测及分析[J].建筑结构学报,2013,34(11):24-32.
LUO Yao-zhi,MEI Yu-jia,SHEN Yan-bin,et al.Field Measurement of Temperature and Stress on Steel Structure of the National Stadium and Analysis of Temperature Action[J].Journal of Building Structures,2013,34(11):24-32.
[2]LUO Y Z,YANG P C,SHEN Y B,et al.Development of a Dynamic Sensing System for Civil Revolving Structures and Its Field Tests in a Large Revolving Auditorium[J].Smart Structures and Systems,2014,13(6):993-1014.
[3]鲜 勇,杨子成,郭玮林,等.基于BP神经网络的惯导初始对准误差辨识方法[J].飞行力学,2021,39(2):77-82,94.
XIAN Yong,YANG Zi-cheng,GUO Wei-lin,et al.Initial Alignment Error Identification Method of SINS Based on BP Neural Network[J].Flight Dynamics,2021,39(2):77-82,94.
[4]罗尧治,刘 钝,沈雁彬,等.杭州铁路东站站房钢结构施工监测[J].空间结构,2013,19(3):3-8,26.
LUO Yao-zhi,LIU Dun,SHEN Yan-bin,et al.Steel Structure Construction Monitoring of Hangzhou Eastrailway Station Building[J].Spatial Structure,2013,19(3):3-8,26.
[5]赵 昕,贾 京,郑毅敏.基于BP神经网络的大跨高空连廊应变监测数据恢复[J].建筑科学与工程学报,2009,26(1):101-106.
ZHAO Xin,JIA Jing,ZHENG Yi-min.Strain Monitoring Data Restoring of Large-span Steel Skybridge Based on BP Neural Network[J].Journal of Architecture and Civil Engineering,2009,26(1):101-106.
[6]杨 渊,练继建,周观根,等.基于深度学习的随机缺失数据重构和结构损伤识别[J].工业建筑,2021,51(增):401-405.
YANG Yuan,LIAN Ji-jian,ZHOU Guan-gen,et al.Random Missing Data Reconstruction and Structural Damage Identification Based on Deep Learning[J].Industrial Construction,2021,51(S):401-405.
[7]孟 欣.建筑能源监测系统中缺失数据的重构与修补方法研究[D].大连:大连理工大学,2021.
MENG Xin.Research on Reconstruction and Repair of Missing Data in Building Energy Monitoring System[D].Dalian:Dalian University of Technology,2021.
[8]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.
JIN Liu,ZHAO Rui,DU Xiu-li.Neural Network Prediction Model of Concrete Compressive Strength Size Effect[J].Journal of Beijing University of Technology,2021,47(3):260-268.
[9]MANO RAJA PAUL M,KANNAN R,LEEBAN MOSES M,et al.Fault Identification in a Grid Connected Solar PV System Using Back Propagation Neural Network[J].IOP Conference Series:Materials Science and Engineering,2021,1084(1):012109.
[10]ROSA SUSILA J K,AFIT M,LAKSONO P.Implementation of Back Propagation Artificial Neural Network for Heart Disease Abnormality Diagnosis[J].Journal of Physics:Conference Series,2021,1764(1):012165.
[11]PENG H,WU H,WANG J W.Research on the Prediction of the Water Demand of Construction Engineering Based on the BP Neural Network[J].Advances in Civil Engineering,2020,2020:8868817.
[12]NANGLIA P.Lung Cancer Classification Using Feed Forward Back Propagation Neural Network for CT Images[J].International Journal of Medical Engineering and Informatics,2021,13(2):1.
[13]欧青立,张 磊,邓 鹏,等.粒子群优化BP神经网络PID控制注塑机液压系统[J].应用科技,2018,45(4):50-55.
OU Qing-li,ZHANG Lei,DENG Peng,et al.Particle Swarm Optimization BP Neural Network PID Control Hydraulic System of Injection Molding Machine[J].Applied Science and Technology,2018,45(4):50-55.
[14]黄 璇,郭立红,李 姜,等.改进粒子群优化BP神经网络的目标威胁估计[J].吉林大学学报(工学版),2017,47(3):996-1002.
HUANG Xuan,GUO Li-hong,LI Jiang,et al.Target Threat Assessment Based on BP Neural Network Optimized by Modified Particle Swarm Optimization[J].Journal of Jilin University(Engineering and Technology Edition),2017,47(3):996-1002.
[15]王雪松,许爱德,赵中林,等.粒子群优化递归神经网络的SRM磁链观测器[J].电气工程学报,2017,12(10):1-8.
WANG Xue-song,XU Ai-de,ZHAO Zhong-lin,et al.Stator Flux Observer of SRM Based on Particle Swarm Optimized Recurrent Neural Network[J].Journal of Electrical Engineering,2017,12(10):1-8.
[16]KENNEDY J,EBERHART R.Particle Swarm Optimization[C]//IEEE.Proceedings of ICNN'95 — International Conference on Neural Networks.Perth:IEEE,1995:1942-1948.

Memo

Memo:
-
Last Update: 2022-07-10