scholarly journals Structural damage detection using hybrid deep learning algorithm

2020 ◽  
Vol 14 (2) ◽  
pp. 53-64
Author(s):  
Dang Viet Hung ◽  
Ha Manh Hung ◽  
Pham Hoang Anh ◽  
Nguyen Truong Thang
Keyword(s):  
Deep Learning ◽  
Damage Detection ◽  
Data Storage ◽  
Structural Damage ◽  
Large Scale ◽  
Short Term Memory ◽  
Learning Algorithm ◽  
Vibration Sensor ◽  
Structural Damage Detection ◽  
Deep Learning Algorithm

Timely monitoring the large-scale civil structure is a tedious task demanding expert experience and significant economic resources. Towards a smart monitoring system, this study proposes a hybrid deep learning algorithm aiming for structural damage detection tasks, which not only reduces required resources, including computational complexity, data storage but also has the capability to deal with different damage levels. The technique combines the ability to capture local connectivity of Convolution Neural Network and the well-known performance in accounting for long-term dependencies of Long-Short Term Memory network, into a single end-to-end architecture using directly raw acceleration time-series without requiring any signal preprocessing step. The proposed approach is applied to a series of experimentally measured vibration data from a three-story frame and successful in providing accurate damage identification results. Furthermore, parametric studies are carried out to demonstrate the robustness of this hybrid deep learning method when facing data corrupted by random noises, which is unavoidable in reality. Keywords: structural damage detection; deep learning algorithm; vibration; sensor; signal processing.

scholarly journals Research on Efficient Deep Learning Algorithm Based on ShuffleGhost in the Field of Virtual Reality

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Bangtong Huang ◽  
Hongquan Zhang ◽  
Zihong Chen ◽  
Lingling Li ◽  
Lihua Shi
Keyword(s):  
Virtual Reality ◽  
Deep Learning ◽  
Large Scale ◽  
Learning Algorithm ◽  
Feature Maps ◽  
Embedded Devices ◽  
Feature Map ◽  
Deep Learning Algorithm ◽  
Proper Design ◽  
The Cost

Deep learning algorithms are facing the limitation in virtual reality application due to the cost of memory, computation, and real-time computation problem. Models with rigorous performance might suffer from enormous parameters and large-scale structure, and it would be hard to replant them onto embedded devices. In this paper, with the inspiration of GhostNet, we proposed an efficient structure ShuffleGhost to make use of the redundancy in feature maps to alleviate the cost of computations, as well as tackling some drawbacks of GhostNet. Since GhostNet suffers from high computation of convolution in Ghost module and shortcut, the restriction of downsampling would make it more difficult to apply Ghost module and Ghost bottleneck to other backbone. This paper proposes three new kinds of ShuffleGhost structure to tackle the drawbacks of GhostNet. The ShuffleGhost module and ShuffleGhost bottlenecks are utilized by the shuffle layer and group convolution from ShuffleNet, and they are designed to redistribute the feature maps concatenated from Ghost Feature Map and Primary Feature Map. Besides, they eliminate the gap of them and extract the features. Then, SENet layer is adopted to reduce the computation cost of group convolution, as well as evaluating the importance of the feature maps which concatenated from Ghost Feature Maps and Primary Feature Maps and giving proper weights for the feature maps. This paper conducted some experiments and proved that the ShuffleGhostV3 has smaller trainable parameters and FLOPs with the ensurance of accuracy. And with proper design, it could be more efficient in both GPU and CPU side.

scholarly journals An Improved Deep Learning Algorithm for Risk Prediction of Corporate Internet Reporting

2020 ◽  
Vol 34 (4) ◽  
pp. 437-444
Author(s):  
Lingyan Ou ◽  
Ling Chen
Keyword(s):  
Deep Learning ◽  
Risk Prediction ◽  
Short Term Memory ◽  
Learning Algorithm ◽  
Moving Average ◽  
Arma Model ◽  
Evaluation Index System ◽  
Online Information ◽  
Deep Learning Algorithm ◽  
Research Findings

Corporate internet reporting (CIR) has such advantages as the strong timeliness, large amount, and wide coverage of financial information. However, the CIR, like any other online information, faces various risks. With the aid of the increasingly sophisticated artificial intelligence (AI) technology, this paper proposes an improved deep learning algorithm for the prediction of CIR risks, aiming to improve the accuracy of CIR risk prediction. After building a reasonable evaluation index system (EIS) for CIR risks, the data involved in risk rating and the prediction of risk transmission effect (RTE) were subject to structured feature extraction and time series construction. Next, a combinatory CIR risk prediction model was established by combining the autoregressive moving average (ARMA) model with long short-term memory (LSTM). The former is good at depicting linear series, and the latter excels in describing nonlinear series. Experimental results demonstrate the effectiveness of the ARMA-LSTM model. The research findings provide a good reference for applying AI technology in risk prediction of other areas.

scholarly journals A deep learning view of the census of galaxy clusters in IllustrisTNG

2020 ◽  
Vol 498 (4) ◽  
pp. 5620-5628
Author(s):  
Y Su ◽  
Y Zhang ◽  
G Liang ◽  
J A ZuHone ◽  
D J Barnes ◽  
...  
Keyword(s):  
Deep Learning ◽  
Galaxy Clusters ◽  
Large Scale ◽  
Learning Algorithm ◽  
X Ray ◽  
Deep Learning Algorithm ◽  
Cluster Evolution ◽  
Novel Method ◽  
Massive Clusters ◽  
Activation Mapping

ABSTRACT The origin of the diverse population of galaxy clusters remains an unexplained aspect of large-scale structure formation and cluster evolution. We present a novel method of using X-ray images to identify cool core (CC), weak cool core (WCC), and non-cool core (NCC) clusters of galaxies that are defined by their central cooling times. We employ a convolutional neural network, ResNet-18, which is commonly used for image analysis, to classify clusters. We produce mock Chandra X-ray observations for a sample of 318 massive clusters drawn from the IllustrisTNG simulations. The network is trained and tested with low-resolution mock Chandra images covering a central 1 Mpc square for the clusters in our sample. Without any spectral information, the deep learning algorithm is able to identify CC, WCC, and NCC clusters, achieving balanced accuracies (BAcc) of 92 per cent, 81 per cent, and 83 per cent, respectively. The performance is superior to classification by conventional methods using central gas densities, with an average ${\rm BAcc}=81{{\ \rm per\ cent}}$, or surface brightness concentrations, giving ${\rm BAcc}=73{{\ \rm per\ cent}}$. We use class activation mapping to localize discriminative regions for the classification decision. From this analysis, we observe that the network has utilized regions from cluster centres out to r ≈ 300 kpc and r ≈ 500 kpc to identify CC and NCC clusters, respectively. It may have recognized features in the intracluster medium that are associated with AGN feedback and disruptive major mergers.

scholarly journals NanoReviser: An Error-correction Tool for Nanopore Sequencing Based on a Deep Learning Algorithm

2020 ◽  
Author(s):  
Luotong Wang ◽  
Li Qu ◽  
Longshu Yang ◽  
Yiying Wang ◽  
Huaiqiu Zhu
Keyword(s):  
Deep Learning ◽  
Error Rate ◽  
Short Term Memory ◽  
Learning Algorithm ◽  
Consensus Sequence ◽  
Sequencing Error ◽  
Nanopore Sequencing ◽  
Electrical Signals ◽  
E Coli ◽  
Deep Learning Algorithm

AbstractNanopore sequencing is regarded as one of the most promising third-generation sequencing (TGS) technologies. Since 2014, Oxford Nanopore Technologies (ONT) has developed a series of devices based on nanopore sequencing to produce very long reads, with an expected impact on genomics. However, the nanopore sequencing reads are susceptible to a fairly high error rate owing to the difficulty in identifying the DNA bases from the complex electrical signals. Although several basecalling tools have been developed for nanopore sequencing over the past years, it is still challenging to correct the sequences after applying the basecalling procedure. In this study, we developed an open-source DNA basecalling reviser, NanoReviser, based on a deep learning algorithm to correct the basecalling errors introduced by current basecallers provided by default. In our module, we re-segmented the raw electrical signals based on the basecalled sequences provided by the default basecallers. By employing convolution neural networks (CNNs) and bidirectional long short-term memory (Bi-LSTM) networks, we took advantage of the information from the raw electrical signals and the basecalled sequences from the basecallers. Our results showed NanoReviser, as a post-basecalling reviser, significantly improving the basecalling quality. After being trained on standard ONT sequencing reads from public E. coli and human NA12878 datasets, NanoReviser reduced the sequencing error rate by over 5% for both the E. coli dataset and the human dataset. The performance of NanoReviser was found to be better than those of all current basecalling tools. Furthermore, we analyzed the modified bases of the E. coli dataset and added the methylation information to train our module. With the methylation annotation, NanoReviser reduced the error rate by 7% for the E. coli dataset and specifically reduced the error rate by over 10% for the regions of the sequence rich in methylated bases. To the best of our knowledge, NanoReviser is the first post-processing tool after basecalling to accurately correct the nanopore sequences without the time-consuming procedure of building the consensus sequence. The NanoReviser package is freely available at https://github.com/pkubioinformatics/NanoReviser.

scholarly journals Unsupervised Structural Damage Detection Technique Based on a Deep Convolutional Autoencoder

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Zahra Rastin ◽  
Gholamreza Ghodrati Amiri ◽  
Ehsan Darvishan
Keyword(s):  
Deep Learning ◽  
Damage Detection ◽  
Structural Damage ◽  
Numerical Models ◽  
Health State ◽  
Training Phase ◽  
Structural Damage Detection ◽  
Healthy State ◽  
Convolutional Autoencoder ◽  
Unsupervised Deep Learning

Structural health monitoring (SHM) is a hot research topic with the main purpose of damage detection in a structure and assessing its health state. The major focus of SHM studies in recent years has been on developing vibration-based damage detection algorithms and using machine learning, especially deep learning-based approaches. Most of the deep learning-based methods proposed for damage detection in civil structures are based on supervised algorithms that require data from the healthy state and different damaged states of the structure in the training phase. As it is not usually possible to collect data from damaged states of a large civil structure, using such algorithms for these structures may be impractical. This paper proposes a new unsupervised deep learning-based method for structural damage detection based on convolutional autoencoders (CAEs). The main objective of the proposed method is to identify and quantify structural damage using a CAE network that employs raw vibration signals from the structure and is trained by the signals solely acquired from the healthy state of the structure. The CAE is chosen to take advantage of high feature extraction capability of convolution layers and at the same time use the advantages of an autoencoder as an unsupervised algorithm that does not need data from damaged states in the training phase. Applications on the two numerical models of IASC-ASCE benchmark structure and a grid structure located at the University of Central Florida, as well as the full-scale Tianjin Yonghe Bridge, prove the efficiency of the proposed algorithm in assessing the global health state of the structures and quantifying the damage.

scholarly journals Real Time LDR Data Prediction using IoT and Deep Learning Algorithm

2020 ◽  
pp. 158-161
Author(s):  
Chandraprabha S ◽  
Pradeepkumar G ◽  
Dineshkumar Ponnusamy ◽  
Saranya M D ◽  
Satheesh Kumar S ◽  
...  
Keyword(s):  
Deep Learning ◽  
Light Intensity ◽  
Real Time ◽  
Short Term Memory ◽  
Learning Algorithm ◽  
Time Interval ◽  
Deep Learning Algorithm ◽  
Data Prediction ◽  
Solar Plant

This paper outfits artificial intelligence based real time LDR data which is implemented in various applications like indoor lightning, and places where enormous amount of heat is produced, agriculture to increase the crop yield, Solar plant for solar irradiance Tracking. For forecasting the LDR information. The system uses a sensor that can measure the light intensity by means of LDR. The data acquired from sensors are posted in an Adafruit cloud for every two seconds time interval using Node MCU ESP8266 module. The data is also presented on adafruit dashboard for observing sensor variables. A Long short-term memory is used for setting up the deep learning. LSTM module uses the recorded historical data from adafruit cloud which is paired with Node MCU in order to obtain the real-time long-term time series sensor variables that is measured in terms of light intensity. Data is extracted from the cloud for processing the data analytics later the deep learning model is implemented in order to predict future light intensity values.

scholarly journals Deep learning algorithm evaluation of hypertension classification in less photoplethysmography signals conditions

2021 ◽  
Vol 54 (3-4) ◽  
pp. 439-445
Author(s):  
Chih-Ta Yen ◽  
Sheng-Nan Chang ◽  
Cheng-Hong Liao
Keyword(s):  
Deep Learning ◽  
Short Term Memory ◽  
Learning Algorithm ◽  
Training Data ◽  
Short Term ◽  
Deep Learning Algorithm ◽  
Testing Dataset ◽  
Data Points ◽  
Optimal Classification ◽  
Long Short Term Memory

This study used photoplethysmography signals to classify hypertensive into no hypertension, prehypertension, stage I hypertension, and stage II hypertension. There are four deep learning models are compared in the study. The difficulties in the study are how to find the optimal parameters such as kernel, kernel size, and layers in less photoplethysmographyt (PPG) training data condition. PPG signals were used to train deep residual network convolutional neural network (ResNetCNN) and bidirectional long short-term memory (BILSTM) to determine the optimal operating parameters when each dataset consisted of 2100 data points. During the experiment, the proportion of training and testing datasets was 8:2. The model demonstrated an optimal classification accuracy of 76% when the testing dataset was used.

Sign in / Sign up

Export Citation Format

Share Document