Research on FBG-Based CFRP Structural Damage Identification Using BP Neural Network

Geng, Xu; Lu, Shizeng; Jiang, Mingshun; Sui, Qingmei; Lv, Shanshan; Xiao, Hang; Jia, Yuxi; Jia, Lei

doi:10.1007/s13320-018-0466-0

Cited by 55 publications

(23 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An 8-14-1 three-layer topology was used to construct a neural network model for predicting the drilling quality, as shown in Figure 6. The topology of the BP neural network model usually includes an input layer, hidden layer, and output layer [17]. The input layer mainly receives external data and information, the hidden layer is calculated using various functional relationships, and the output layer is mainly used to carry out the calculation conclusions and give prediction results.…”

Section: Establishment Of Neural Network Modelmentioning

confidence: 99%

Prediction of Manufacturing Quality of Holes Based on a BP Neural Network

et al. 2020

View full text Add to dashboard Cite

In order to improve the manufacturing quality of holes (Φ3–Φ8 mm) and to optimize the hole drilling process in T300 carbon fiber-reinforced plastic (CFRP) and 7050-T7 Al alloy stacks, a prediction model of multiple objective parameter optimization was proposed based on a back propagation (BP) neural network algorithm. Four parameters of feed rate, spindle speed, drilling diameter, and cushion plate were taken as the input layer parameters to study the manufacturing quality of holes in four stack types: CFRP/Al, Al/CFRP, Al/CFRP/Al, and CFRP/Al/CFRP. Delamination and tearing defects often appear in the drilling process; thus, a certain degree of defects in CFRP was selected as the output parameter, in an effort to build a prediction model of drilling quality. After the neural network model of the optimized hole-making process of an 8–14–1 three-layer topology was corrected by 170 steps, the error was reduced to 0.00016882, the regression fitting was 0.99978, and the fitting error of training samples was 10−2~10−5. The prediction model of the number of defective holes provided basically similar results to the experimental data. This indicates that the prediction model based on a BP neural network has good prediction ability. Based on the prediction of parameters, verification tests were performed, and the number of defective holes in CFRP was reduced while the manufacturing quality of the holes was improved significantly; the qualified rate of manufactured holes reached 97%.

show abstract

Section: Establishment Of Neural Network Modelmentioning

confidence: 99%

Prediction of Manufacturing Quality of Holes Based on a BP Neural Network

et al. 2020

View full text Add to dashboard Cite

show abstract

“…, 36. The Levenberg-Marquardt (LM) algorithm [43] was used to adjust the weights and thresholds of the network to minimize the network error. Assuming that the expected output of the output layer was T, the output error E was as follows:…”

Section: Appendix Amentioning

confidence: 99%

Modal Strain Energy-Based Structural Damage Detection Using Convolutional Neural Networks

Chen

Liu

et al. 2019

Applied Sciences

View full text Add to dashboard Cite

In this paper, a convolutional neural network (CNN) was used to extract the damage features of a steel frame structure. As structural damage could induce changes of the modal parameters of the structure, the convolution operation was used to extract the features of modal parameters, and a classification algorithm was used to judge the damage state of the structure. The finite element method was applied to analyze the free vibration of the steel frame and obtain the first-order modal strain energy for various damage scenarios, which was used as the CNN training sample. Then vibration experiments were carried out, and modal parameters were obtained from the modal analysis of the vibration signals. The experimental data were inputted into the CNN to verify its damage detection capability. The result showed that the CNN was effective in detecting the intact structure, single damage, and multi damages with an accuracy of 100%. For comparison, the same samples were also applied to the traditional back propagation (BP) neural network, which failed to detect the intact structure and multiple-damage cases. It was found that: (1) The proposed CNN could be trained from finite element simulation data and used in real frame structure damage detection, and it performed better in structural damage detection than BP neural networks; (2) the measured data of a real structure could be supplemented by numerical simulation data, and satisfactory results have been demonstrated.

show abstract

“…The topology used in this manuscript is shown in Fig. 2 [24]. x Figure 3 shows the training and testing phases of the BP neural network to obtain the shifted section(s) information of the Brillouin sensor.…”

Section: Bp Neural Networkmentioning

confidence: 99%

Back propagation neutral network based signal acquisition for Brillouin distributed optical fiber sensors

Cao¹,

Guo²,

Li³

et al. 2019

Opt. Express

View full text Add to dashboard Cite

This manuscript proposes a method based on back propagation (BP) neural network and the spectral subtraction method to quickly obtain sensing information in Brillouin fiber optics sensors. BP neural network's characteristics which can realize any complex nonlinear mapping help to determine the frequency shift section(s) information. The training function, transfer function and number of hidden layer nodes of BP neural network are determined with experimental data. The experimental results show that comparing with traditional Lorentz fitting algorithm and edge detection with Sobel operator, the BP neural network is about 1/12 in terms of time complexity with the Lorentz algorithm, about 1/9 with the edge detection based on Sobel operator; while the respective accuracy on determine the frequency shifted section(s) has improved by 79.4% and 27.9%.

show abstract

Research on FBG-Based CFRP Structural Damage Identification Using BP Neural Network

Cited by 55 publications

References 12 publications

Prediction of Manufacturing Quality of Holes Based on a BP Neural Network

Prediction of Manufacturing Quality of Holes Based on a BP Neural Network

Modal Strain Energy-Based Structural Damage Detection Using Convolutional Neural Networks

Back propagation neutral network based signal acquisition for Brillouin distributed optical fiber sensors

Contact Info

Product

Resources

About