Recurrent neural network model with Bayesian training and mutual information for response prediction of large buildings

Perez-Ramirez, Carlos A.; Amezquita-Sanchez, Juan P.; Valtierra-Rodríguez, Martín; Adeli, Hojjat; Domínguez-González, Aurelio; Romero‐Troncoso, Rene J.

doi:10.1016/j.engstruct.2018.10.065

Cited by 195 publications

(114 citation statements)

References 54 publications

Supporting

Mentioning

113

Contrasting

Order By: Relevance

“…For some of the proposed methods, the selection methods of the seismic loads used for verification of the trained NNs were inappropriate. In some studies, the same EQs as those used for training were taken as new EQs for verification, with only the intensities being changed . Thus, the factors affecting the response amplifications of the building structures subjected to the EQs, such as their duration and dominant period, were unchanged.…”

Section: Introductionmentioning

confidence: 99%

“…In some studies, the same EQs as those used for training were taken as new EQs for verification, with only the intensities being changed. 13,16,17 Thus, the factors affecting the response amplifications of the building structures subjected to the EQs, such as their duration and dominant period, were unchanged. Strictly speaking, use of the same EQs as those employed for training with different intensities only cannot be regarded as use of completely new EQs.…”

Section: Introductionmentioning

confidence: 99%

“…They verified the proposed method through a numerical study with several multi‐degrees of freedom. Perez‐Ramirez et al introduced a recurrent NN to predict the seismic responses of large buildings. The NN was developed by setting the time history of the acceleration responses measured on specific floors in a building as input and that of the acceleration responses measured on another floor as output.…”

Section: Introductionmentioning

confidence: 99%

“…In the aforementioned literature, the following broad limitations exist: The NN construction is limited to an FEM model that cannot reflect the uncertainties of actual structures; no emphasis is placed on displacements that are more meaningful responses from a seismic performance evaluation perspective; and measured displacement responses to EQs are required for application of the trained NN . For some of the proposed methods, the selection methods of the seismic loads used for verification of the trained NNs were inappropriate.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Seismic response prediction method for building structures using convolutional neural network

Park

2020

Struct Control Health Monit

View full text Add to dashboard Cite

Summary In this study, a method of predicting the seismic responses of building structures based on a convolutional neural network (CNN) is proposed. In the method, the time histories of acceleration responses previously measured in a building during earthquakes are used in the CNN input layer, with the corresponding time histories of the displacement responses being used in the CNN output layer. The correlations between the features automatically extracted from the acceleration responses by the convolution and pooling operations in the various CNN layers and the displacement responses in the CNN output layer are established through CNN training. The trained CNN predicts the displacement responses for a future earthquake event using only the measured acceleration responses. To verify the validity of the proposed method, a numerical study on the ASCE benchmark model and an experimental study on a reinforced concrete frame structure are conducted. In the numerical study, the structural responses of the ASCE model subject to artificial earthquakes are utilized for CNN training, and the performance of the trained CNN is verified through seismic response prediction. In the experimental study, shaking table tests are performed for various earthquakes to measure the seismic responses of the reinforced concrete test structure. The seismic response prediction performance of the proposed method is examined using CNN trained with the measured seismic responses. Furthermore, the performances of CNNs trained with different numbers of datasets generated by the proposed data generation method based on data overlapping with the same data pool are discussed with related to the number of CNN training iteration number.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Seismic response prediction method for building structures using convolutional neural network

Park

2020

Struct Control Health Monit

View full text Add to dashboard Cite

show abstract

“…In the last decade, vision‐based structural impairment detection methods have received significant attention in monitoring civil infrastructure including bridges, roadways and railways, tunnels, and pavements (Gao & Mosalam, ; Liang, ; Lin, Nie, & Ma, ; Perez‐Ramirez et al., ; Rafiei & Adeli, , ; Rafiei, Khushefati, Demirboga, & Adeli, ; Xue & Li, ; Yeum, Choi, & Dyke, ). Autonomous crack detection systems aid short‐term and long‐term inspections in terms of decreasing human involvement during their operation, resulting in lower cost, higher reliability, and system efficiency.…”

Section: Introductionmentioning

confidence: 99%

Concrete crack detection using context‐aware deep semantic segmentation network

Zhang

Rajan

Story

2019

Computer aided Civil Eng

161

View full text Add to dashboard Cite

Computer‐vision and deep‐learning techniques are being increasingly applied to inspect, monitor, and assess infrastructure conditions including detection of cracks. Traditional vision‐based methods to detect cracks lack accuracy and generalization to work on complicated infrastructural conditions. This paper presents a novel context‐aware deep convolutional semantic segmentation network to effectively detect cracks in structural infrastructure under various conditions. The proposed method applies a pixel‐wise deep semantic segmentation network to segment the cracks on images with arbitrary sizes without retraining the prediction network. Meanwhile, a context‐aware fusion algorithm that leverages local cross‐state and cross‐space constraints is proposed to fuse the predictions of image patches. This method is evaluated on three datasets: CrackForest Dataset (CFD) and Tomorrows Road Infrastructure Monitoring, Management Dataset (TRIMMD) and a Customized Field Test Dataset (CFTD) and achieves Boundary F1 (BF) score of 0.8234, 0.8252, and 0.7937 under 2‐pixel error tolerance margin in CFD, TRIMMD, and CFTD, respectively. The proposed method advances the state‐of‐the‐art performance of BF score by approximately 2.71% in CFD, 1.47% in TRIMMD, and 4.14% in CFTD. Moreover, the averaged processing time of the proposed system is 0.7 s on a typical desktop with Intel® Quad‐Core™ i7‐7700 CPU@3.6 GHz Processor, 16GB RAM and NVIDIA GeForce GTX 1060 6GB GPU for an image of size 256 × 256 pixels.

show abstract

Experimental verification for a mechanics‐based digital cloning method predicting structural performance and damage

Rabiepour

Zhou

Chase

2023

Earthq Engng Struct Dyn

View full text Add to dashboard Cite

This paper introduces a novel digital twin modelling framework combining accurate structural health monitoring (SHM) results delivered by hysteresis loop analysis (HLA) with the basis function modelling to predict structural damage and responses of earthquake-affected structures under future seismic events. The measured data sets of a 12-storey 1/10 scale reinforced concrete (RC) structure shaken by three damaging base excitations (Events 1, 2 and 3) using a shake table has been employed to validate the performance of the introduced predictive framework. The results show that the digital twin created under the first base excitation (Event 1) predicts the inter-storey displacement responses of the test structure due to the second and third excitations (Events 2 and 3) with an average correlation coefficient of ∼0.88 and ∼0.94, respectively. The inter-storey drift ratio (IDR) values predicted by the digital twin also result in the same seismic performance delivered by the measured values, making the introduced framework a promising tool for assessing structural collapse and subsequent financial risk using incremental dynamic analysis (IDA). Moreover, the maximum difference of 5% between the identified and predicted stiffness values illustrates the robust and accurate performance of the proposed framework in diagnosing structural damage in highly nonlinear and damaged structures.

show abstract

Recurrent neural network model with Bayesian training and mutual information for response prediction of large buildings

Cited by 195 publications

References 54 publications

Seismic response prediction method for building structures using convolutional neural network

Seismic response prediction method for building structures using convolutional neural network

Concrete crack detection using context‐aware deep semantic segmentation network

Experimental verification for a mechanics‐based digital cloning method predicting structural performance and damage

Contact Info

Product

Resources

About