Deep long short-term memory networks for nonlinear structural seismic response prediction

Zhang, Ruiyang; Chen, Zhao; Su, Chen; Zheng, Jingwei; Büyüköztürk, Oral; Sun, Hao

doi:10.1016/j.compstruc.2019.05.006

Cited by 291 publications

(108 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with general neural networks, connections between hidden units can be established in recurrent neural networks (RNNs), allowing the latter to retain memories of recent events. Conventional RNNs exist gradients vanishing and exploding, a long short‐term memory (LSTM) neural network as a variation of RNNs was thus developed to overcome this problem The LSTM algorithm has recently been used in practical engineering with time‐series characteristics such as the prediction of long‐term settlement of structures, hydro‐mechanical responses of multi‐permeability porous media and structural seismic response . Because soil behaviour under cyclic loading is a continuous process, the current stress‐strain status depends on the soil behaviour at previous steps and also affects the soil behaviour at the later steps.…”

Section: Introductionmentioning

confidence: 99%

An AI‐based model for describing cyclic characteristics of granular materials

Zhang

Yin

Jin

et al. 2020

Num Anal Meth Geomechanics

View full text Add to dashboard Cite

Modelling cyclic behaviour of granular soils under both drained and undrained conditions with a good performance is still a challenge. This study presents a new way of modelling the cyclic behaviour of granular materials using deep learning. To capture the continuous cyclic behaviour in time dimension, the long short-term memory (LSTM) neural network is adopted, which is characterised by the prediction of sequential data, meaning that it provides a novel means of predicting the continuous behaviour of soils under various loading paths. Synthetic datasets of cyclic loading under drained and undrained conditions generated by an advanced soil constitutive model are first employed to explore an appropriate framework for the LSTM-based model. Then the LSTM-based model is used to estimate the cyclic behaviour of real sands, ie, the Toyoura sand under the undrained condition and the Fontainebleau sand under both undrained and drained conditions. The estimates are compared with actual experimental results, which indicates that the LSTM-based model can simultaneously simulate the cyclic behaviour of sand under both drained and undrained conditions, ie, (a) the cyclic mobility mechanism, the degradation of effective stress and large deformation under the undrained condition, and (b) shear strain accumulation and densification under the drained condition. K E Y W O R D S constitutive model, cyclic loading, deep learning, liquefaction, neural network, sand

show abstract

Section: Introductionmentioning

confidence: 99%

An AI‐based model for describing cyclic characteristics of granular materials

Zhang

Yin

Jin

et al. 2020

Num Anal Meth Geomechanics

View full text Add to dashboard Cite

show abstract

“…Combined with computer vision technology, the networks of deep learning represented by convolutional neural network (CNN) are developed to conduct the smart detection of the cracking, corrosion, and looseness for various structural components 42–47 . Moreover, scholars provide the comprehensive approach of deep learning to capture the nonlinear behavior of structure and evaluate the complex process of the evolution of structural performance 48–50 . Instead of the manual operation, the abnormal signal in the real‐time monitoring data can be intelligently detected and recovered by the network of deep learning 51,52 .…”

Section: Introductionmentioning

confidence: 99%

Digital modeling on the nonlinear mapping between multi‐source monitoring data of in‐service bridges

Zhao

Wei

et al. 2020

Struct Control Health Monit

View full text Add to dashboard Cite

Nonlinear mapping of the fuzzy relation exists between structural inputs and outputs, as well as between structural global and local response. It is difficult for the numerical simulation to introduce the nonlinear effects of the variability of loading effects and the uneven deterioration of structure. The big monitoring data make it feasible to mine these nonlinear effects, and the network of deep learning is a good tool to establish the nonlinear mapping model between the multi-source monitoring data. Based on the temperature, strain, and dynamic displacement data from the structural health monitoring (SHM) system of an in-service bridge, the deep learning regression network of long short-term memory (LSTM) is designed and trained, which is used for modeling the nonlinear mapping between the structural input-output and the structural global-local response. The digital model of the nonlinear mapping of the temperature to temperature-induced strain, dynamic displacement to vehicleinduced strain, and vehicle-induced strain to dynamic displacement is established by the deep learning of LSTM regression network. The established nonlinear mapping model can be used to recover the abnormal and missing data with multi-source data, and it digitally linked the structural input-output and the structural global-local response. Then, the utilization of SHM data analytics will be enhanced, and the digital modeling for the in-service state of the bridge structure will be fast implemented.

show abstract

“…Recently, deep learning (DL) has attracted great attention because of its powerful capability in recognizing patterns and discovering intricate structures in large data sets [ 24 ]. Considering its strength in learning nonlinear manifolds of data [ 25 ], researchers have explored several security analysis methods of random numbers by DL. In [ 26 , 27 ], the authors implemented feedforward neural network (FNN) structures for detecting hidden patterns among pseudo-random numbers from DRNGs.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Security Verification for a Random Number Generator Using White Chaos

Zhang

Sang

et al. 2020

Entropy

View full text Add to dashboard Cite

In this paper, a deep learning (DL)-based predictive analysis is proposed to analyze the security of a non-deterministic random number generator (NRNG) using white chaos. In particular, the temporal pattern attention (TPA)-based DL model is employed to learn and analyze the data from both stages of the NRNG: the output data of a chaotic external-cavity semiconductor laser (ECL) and the final output data of the NRNG. For the ECL stage, the results show that the model successfully detects inherent correlations caused by the time-delay signature. After optical heterodyning of two chaotic ECLs and minimal post-processing are introduced, the model detects no patterns among corresponding data. It demonstrates that the NRNG has the strong resistance against the predictive model. Prior to these works, the powerful predictive capability of the model is investigated and demonstrated by applying it to a random number generator (RNG) using linear congruential algorithm. Our research shows that the DL-based predictive model is expected to provide an efficient supplement for evaluating the security and quality of RNGs.

show abstract

Deep long short-term memory networks for nonlinear structural seismic response prediction

Cited by 291 publications

References 45 publications

An AI‐based model for describing cyclic characteristics of granular materials

An AI‐based model for describing cyclic characteristics of granular materials

Digital modeling on the nonlinear mapping between multi‐source monitoring data of in‐service bridges

Deep Learning-Based Security Verification for a Random Number Generator Using White Chaos

Contact Info

Product

Resources

About