A feature learning-based method for impact load reconstruction and localization of the plate-rib assembled structure

2021 ◽  
pp. 147592172110380
Author(s):  
Tao Chen ◽  
Liang Guo ◽  
Andongzhe Duan ◽  
Hongli Gao ◽  
Tingting Feng ◽  
...  
Keyword(s):  
Neural Network ◽  
Short Term Memory ◽  
Impact Load ◽  
Feature Learning ◽  
Time History ◽  
Great Ability ◽  
Feature Representations ◽  
New Feature ◽  
Formula Derivation ◽  
The Impact

Impact load is the load that machines frequently experienced in engineering applications. Its time-history reconstruction and localization are crucial for structural health monitoring and reliability analysis. However, when identifying random impact loads, conventional inversion methods usually do not perform well because of complex formula derivation, infeasibility of nonlinear structure, and ill-posed problem. Deep learning methods have great ability of feature learning and nonlinear representation as well as comprehensive regularization mechanism. Therefore, a new feature learning-based method is proposed to conduct impact load reconstruction and localization. The proposed method mainly includes two parts. The first part is designed to reconstruct impact load, named convolutional-recurrent encoder–decoder neural network (ED-CRNN). The other part is constructed to localize impact load, called deep convolutional-recurrent neural network (DCRNN). The ED-CRNN utilizes the one-dimensional (1-D) convolutional encoder–decoder to obtain low-dimension feature representations of input signals. Two long short-term memory (LSTM) layers and a bidirectional LSTM (BiLSTM) layer are uniformly distributed in this network to learn the relationship between input features and the output load in time steps. The DCRNN is constructed mainly by two 1-D convolutional neural network (CNN) layers and two BiLSTM layers to learn high-hidden-level spatial as well as temporal features. The fully connected layers are placed at the end to localize an impact load. The effectiveness of the proposed method was demonstrated by two numerical studies and two experiments. The results reveal that the proposed method has the ability to accurately and quickly reconstruct and localize the impact load of complex assembled structure. Furthermore, the performance of the DCRNN is related to the number of sensors and the architecture of the network. Meanwhile, the strategy of alternating layout is proposed to reduce the number of training locations.

Production Forecasting with the Interwell Interference by Integrating Graph Convolutional and Long Short-Term Memory Neural Network

2021 ◽  
pp. 1-17
Author(s):  
Enda Du ◽  
Yuetian Liu ◽  
Ziyan Cheng ◽  
Liang Xue ◽  
Jing Ma ◽  
...  
Keyword(s):  
Neural Network ◽  
Short Term Memory ◽  
Short Term ◽  
Term Memory ◽  
Production Forecasting ◽  
Temporal Correlations ◽  
Proposed Model ◽  
The Mean ◽  
Long Short Term Memory ◽  
The Impact

Summary Accurate production forecasting is an essential task and accompanies the entire process of reservoir development. With the limitation of prediction principles and processes, the traditional approaches are difficult to make rapid predictions. With the development of artificial intelligence, the data-driven model provides an alternative approach for production forecasting. To fully take the impact of interwell interference on production into account, this paper proposes a deep learning-based hybrid model (GCN-LSTM), where graph convolutional network (GCN) is used to capture complicated spatial patterns between each well, and long short-term memory (LSTM) neural network is adopted to extract intricate temporal correlations from historical production data. To implement the proposed model more efficiently, two data preprocessing procedures are performed: Outliers in the data set are removed by using a box plot visualization, and measurement noise is reduced by a wavelet transform. The robustness and applicability of the proposed model are evaluated in two scenarios of different data types with the root mean square error (RMSE), the mean absolute error (MAE), and the mean absolute percentage error (MAPE). The results show that the proposed model can effectively capture spatial and temporal correlations to make a rapid and accurate oil production forecast.

Finite Element Analysis of a Single-Layer Reticulated Dome under Impact Loading

2010 ◽  
Vol 163-167 ◽  
pp. 327-331 ◽  
Author(s):  
Liang Zheng ◽  
Zhi Hua Chen
Keyword(s):  
Finite Element ◽  
Dynamic Response ◽  
Impact Force ◽  
Impact Load ◽  
Residual Deformation ◽  
Time History ◽  
Single Layer ◽  
Element Analysis ◽  
Deformation Stage ◽  
The Impact

Finite element model of both the single-layer Schwedler reticulated dome with the span of 50m and a Cuboid impactor were developed, incorporating ANSYS/LS-DYNA. PLASTIC_KINEMATIC (MAT_003) material model which takes stain rate into account was used to simulate steel under impact load. The automatic point to surface contact (NODES TO SURFACE) was applied between the dome and impact block. Three stages of time history curve of the impact force on the apex of the single-layer Scheduler reticulated dome including the impact stage, stable stalemate stage, the decaying stage were generalized according to its dynamic response. It must be pointed out that the peak of the impact force of the single-layer reticulated dome increase with the increase of the weight and the velocity of the impact block, but the change of the velocity of the impact block is more sensitive than the change of weight of the impact block for the effect of the peak of the impact force, and a platform value of the impact force of the single-layer reticulated dome change near a certain value, and the duration time of the impact gradually increase. Then four stages of time history curve of the impact displacement were proposed according to the dynamic response of impact on the apex of the single-layer reticulated dome based on numerical analysis. Four stages include in elastic deformation stage, plastic deformation stage, elastic rebound stage, free vibration stage in the position of the residual deformation.

Numerical Simulation and Experimental Study on Fluid-Filled Hemispherical Shell under Impact

2013 ◽  
Vol 364 ◽  
pp. 172-176
Author(s):  
Hui Wei Yang ◽  
Bin Qin ◽  
Zhi Jun Han ◽  
Guo Yun Lu
Keyword(s):  
Numerical Simulation ◽  
Dynamic Response ◽  
Impact Load ◽  
Elastic Recovery ◽  
Plastic Hinge ◽  
Time History ◽  
Hemispherical Shell ◽  
History Of ◽  
Local Impact ◽  
The Impact

The dynamic response of fluid-filled hemispherical shell in mass impact is studied by experiment using DHR9401. Combining the time history of impact force with experimental observation of the deformation process, it can be seen that the dynamic response can be divided into four stages: the flattening around the impact point, the forming and expanding outward of shell plastic hinge, the plastic edge region flatten by the punch, and elastic recovery. The experimental results show that: Because the shell filled with liquid, the local impact load that the shell suffered is translated into area load and loads on the inner shell uniformly, so that it has a high carrying capacity. Numerical simulation is used to study the time history of energy absorption of different shell structures. The result shows that the crashworthiness of sandwich fluid-filled shell is improved greatly. Under the certain impact energy, deformation of its inner shell is very small, which can provide effective security space.

ESTIMATION OF IMPACT LOAD IN THICK PLATE BY USING THEORETICAL GREEN FUNCTION AND EXPERIMENTAL MEASUREMENT OF VIBRATION

2008 ◽  
Vol 22 (09n11) ◽  
pp. 1377-1382
Author(s):  
H. W. Kim ◽  
S. K. Lee
Keyword(s):  
Green’S Function ◽  
Green's Function ◽  
Thick Plate ◽  
Impact Load ◽  
Plate Theory ◽  
Time History ◽  
Force Transducer ◽  
Good Prediction ◽  
History Of ◽  
The Impact

The classic plate theory (CPT) as a theoretical solution to an impact load has been used in a thin plate. However, The CPT is not any more useful solution for the impact load in the industrial power plant, which is generally constructed by the thick plate. In this paper a novel and effective approach is developed to determine the time history of the impact load on a thick aluminum plate based on the analysis of the acoustic waveforms measured by a sensor array located on the thick plate surface in combination with the theoretical Green's function for the plate. The Green's functions are derived based on either the exact elastodynamic or theory the approximate shear deformation plate theory (SDPT). If the displacement is measured on the plate, then the time history of impact load can be calculated by deconvolving the measured displacement with the theoretical Green's function. The reconstructed time history for impact load is compared with the time history of the impact load measured by the force transducer. A good prediction is found. This technique presents a valuable method for identification of source and may be applied to in-service structures under impact to signals recorded from acoustic emission of propagating cracks.

scholarly journals Numerical simulation of 2D hydrodynamic impact of wedge and ship section at variable velocity

2013 ◽  
Vol 10 (1) ◽  
pp. 49-58 ◽  
Author(s):  
Md. Mashiur Rahaman ◽  
Hiromichi Akimoto ◽  
Md. Ashim Ali
Keyword(s):  
Impact Load ◽  
Time History ◽  
Free Surface Flow ◽  
Surface Flow ◽  
Variable Velocity ◽  
History Of ◽  
Volume Method ◽  
The Impact ◽  
Simulated Time ◽  
Simulated Flow

A commercial CFD code Fluent 6.3® is used to simulate non-linear free surface flow and compute the impact load during variable velocity water entry of 2D wedge and ship section. The code uses the finite volume method to solve the conservation of mass and momentum equations to obtain simulated flow field. The interface between water and air was modeled using volume of fluid (VOF) method. Wedge section with 30 degree dead-rise angle and a ship section are numerically simulated. Time history of impact force and pressures at distinct locations are predicted; and compared with existing experimental results and other numerical methods. Present numerical results compare well with experimental measurements.DOI: http://dx.doi.org/10.3329/jname.v10i1.14383

scholarly journals Prediction of Building Energy Consumption Based on IPSO-CLSTM Neural Network

2021 ◽  
Vol 3 (2) ◽  
pp. 11
Author(s):  
Qingwu Fan ◽  
Li Shuo ◽  
Xudong Liu
Keyword(s):  
Neural Network ◽  
Prediction Model ◽  
Short Term Memory ◽  
Design Stage ◽  
Cooling Load ◽  
Improve Energy Efficiency ◽  
Heating And Cooling ◽  
Heating And Cooling Load ◽  
Ipso Algorithm ◽  
The Impact

Accurate prediction of building load is essential for energy saving and environmental protection. Exploring the impact of building characteristics on heating and cooling load can improve energy efficiency from the design stage of the building. In this paper, a prediction model of building heating and cooling loads is proposed, which based on Improved Particle Swarm Optimization (IPSO) algorithm and Convolution Long Short-Term Memory (CLSTM) neural network model. Firstly, the characteristic variables are extracted and evaluated by Spearman’s correlation coefficient method; Then the prediction model based on the CLSTM neural network is constructed to predict building heating and cooling load. The IPSO algorithm is adopted to solve the problem that manual work cannot precisely adjust parameters. In this method, the optimization ability of the PSO algorithm is improved by changing the updating rule of inertia weight and learning factors. Finally, the parameters of the neural network are taken as IPSO optimization object to improve the prediction accuracy. In the experimental stage of this paper, a variety of algorithm models are compared, and the results show that IPSO-CLSTM can get the best results in the prediction of heating and cooling load.

scholarly journals A Hybrid Deep Learning Model to Predict the Impact of COVID-19 on Mental Health form Social Media Big Data

2021 ◽  
Author(s):  
Tapotosh Ghosh ◽  
Md. Hasan Al Banna ◽  
Md. Jaber Al Nahian ◽  
Kazi Abu Taher ◽  
M Shamim Kaiser ◽  
...  
Keyword(s):  
Neural Network ◽  
Mental Health ◽  
Social Media ◽  
Big Data ◽  
Language Processing ◽  
Short Term Memory ◽  
Daily Basis ◽  
Significant Rise ◽  
Special Focus ◽  
The Impact

The novel coronavirus disease (COVID-19) pandemic is provoking a prevalent consequence on mental health because of less interaction among people, economic collapse, negativity, fear of losing jobs, and death of the near and dear ones. To express their mental state, people often are using social media as one of the preferred means. Due to reduced outdoor activities, people are spending more time on social media than usual and expressing their emotion of anxiety, fear, and depression. On a daily basis, about 2.5 quintillion bytes of data are generated on social media, analyzing this big data can become an excellent means to evaluate the effect of COVID-19 on mental health. In this work, we have analyzed data from Twitter microblog (tweets) to find out the effect of COVID-19 on peoples mental health with a special focus on depression. We propose a novel pipeline, based on recurrent neural network (in the form of long-short term memory or LSTM) and convolutional neural network, capable of identifying depressive tweets with an accuracy of 99.42%. Preprocessed using various natural language processing techniques, the aim was to find out depressive emotion from these tweets. Analyzing over 571 thousand tweets posted between October 2019 and May 2020 by 482 users, a significant rise in depressing tweets was observed between February and May of 2020, which indicates as an impact of the long ongoing COVID-19 pandemic situation.

Chapter 4. Graph Reasoning Networks and Applications

2021 ◽  
Author(s):  
Qingxing Cao ◽  
Wentao Wan ◽  
Xiaodan Liang ◽  
Liang Lin
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Domain Knowledge ◽  
Feature Learning ◽  
External Information ◽  
Feature Representations ◽  
Domain Specific ◽  
The Neural Network ◽  
Structured Knowledge ◽  
Knowledge Graphs

Despite the significant success in various domains, the data-driven deep neural networks compromise the feature interpretability, lack the global reasoning capability, and can’t incorporate external information crucial for complicated real-world tasks. Since the structured knowledge can provide rich cues to record human observations and commonsense, it is thus desirable to bridge symbolic semantics with learned local feature representations. In this chapter, we review works that incorporate different domain knowledge into the intermediate feature representation.These methods firstly construct a domain-specific graph that represents related human knowledge. Then, they characterize node representations with neural network features and perform graph convolution to enhance these symbolic nodes via the graph neural network(GNN).Lastly, they map the enhanced node feature back into the neural network for further propagation or prediction. Through integrating knowledge graphs into neural networks, one can collaborate feature learning and graph reasoning with the same supervised loss function and achieve a more effective and interpretable way to introduce structure constraints.

scholarly journals HYBRID-CNN: An Efficient Scheme for Abnormal Flow Detection in the SDN-Based Smart Grid

2020 ◽  
Vol 2020 ◽  
pp. 1-20
Author(s):  
Pengpeng Ding ◽  
Jinguo Li ◽  
Liangliang Wang ◽  
Mi Wen ◽  
Yuyao Guan
Keyword(s):  
Neural Network ◽  
Smart Grid ◽  
Network Flow ◽  
Short Term Memory ◽  
False Positive Rate ◽  
Flow Analysis ◽  
Feature Learning ◽  
Support Vector ◽  
Global Features ◽  
2D Data

Software-Defined Network (SDN) can improve the performance of the power communication network and better meet the control demand of the Smart Grid for its centralized management. Unfortunately, the SDN controller is vulnerable to many potential network attacks. The accurate detection of abnormal flow is especially important for the security and reliability of the Smart Grid. Prior works were designed based on traditional machine learning methods, such as Support Vector Machine and Naive Bayes. They are simple and shallow feature learning, with low accuracy for large and high-dimensional network flow. Recently, there have been several related works designed based on Long Short-Term Memory (LSTM), and they show excellent ability on network flow analysis. However, these methods cannot get the deep features from network flow, resulting in low accuracy. To address the above problems, we propose a Hybrid Convolutional Neural Network (HYBRID-CNN) method. Specifically, the HYBRID-CNN utilizes a Deep Neural Network (DNN) to effectively memorize global features by one-dimensional (1D) data and utilizes a CNN to generalize local features by two-dimensional (2D) data. Finally, the proposed method is evaluated by experiments on the datasets of UNSW_NB15 and KDDCup 99. The experimental results show that the HYBRID-CNN significantly outperforms existing methods in terms of accuracy and False Positive Rate (FPR), which successfully demonstrates that it can effectively detect abnormal flow in the SDN-based Smart Grid.

Testing and Error Analysis of some Structure’s Impact Load

2010 ◽  
Vol 139-141 ◽  
pp. 2473-2477
Author(s):  
Ji Ping Wu ◽  
You Xing Cai
Keyword(s):  
Impact Load ◽  
Time History ◽  
Load Testing ◽  
Object Based ◽  
Impact Theory ◽  
Finite Element Technology ◽  
Testing Accuracy ◽  
Relationship Of ◽  
The Impact ◽  
The Relationship

The shock response of single freedom system was studied based on impact theory. The shear blade of some merchant shearing machine was selected as the object. The static calibration and impact load testing was carried on the object. Based on the impact theory and finite element technology, the feasibility of the testing scheme and the influence of the shock wave shape on the testing accuracy were discussed. The results show that for the shear blade, when the impact load time history is 50ms and the natural frequency is 2848Hz, there only 0.42% difference between the maximum dynamic response and the static calibrated response. So it is practicable to use the static calibrated results to decide the relationship of the impact load and strain.

Sign in / Sign up

Export Citation Format

Share Document