scholarly journals Real-Time Hybrid Simulation with Deep Learning Computational Substructures: System Validation Using Linear Specimens

2020 ◽  
Vol 2 (4) ◽  
pp. 469-489
Author(s):  
Elif Ecem Bas ◽  
Mohamed A. Moustafa
Keyword(s):  
Finite Element ◽  
Deep Learning ◽  
Real Time ◽  
Experimental Testing ◽  
Hybrid Simulation ◽  
Systems Design ◽  
Frame Structure ◽  
Alternative Methods ◽  
Training Dataset ◽  
Structural Systems

Hybrid simulation (HS) is an advanced simulation method that couples experimental testing and analytical modeling to better understand structural systems and individual components’ behavior under extreme events such as earthquakes. Conducting HS and real-time HS (RTHS) can be challenging with complex analytical substructures due to the nature of direct integration algorithms when the finite element method is employed. Thus, alternative methods such as machine learning (ML) models could help tackle these difficulties. This study aims to investigate the quality of the RTHS tests when a deep learning algorithm is used as a metamodel to represent the dynamic behavior of a nonlinear analytical substructure. The compact HS laboratory at the University of Nevada, Reno was utilized to conduct exclusive RTHS tests. Simulating a braced frame structure, the RTHS tests combined, for the first time, linear brace model specimens (physical substructure) along with nonlinear ML models for the frame (analytical substructure). Deep long short-term memory (Deep-LSTM) networks were employed and trained to develop the metamodels of the analytical substructure using the Python environment. The training dataset was obtained from pure analytical finite element simulations for the complete structure under earthquake excitation. The RTHS evaluations were first conducted for virtual RTHS tests, where substructuring was sought between the LSTM metamodel and virtual experimental substructure. To validate the proposed RTHS testing methodology and full system, several actual RTHS tests were conducted. The results from ML-based RTHS were evaluated for different ML models and compared against results from conventional RTHS with finite element models. The paper demonstrates the potential of conducting successful experimental RTHS using Deep-LSTM models, which could open the door for unparalleled new opportunities in structural systems design and assessment.

scholarly journals Application of Artificial Neural Networks in Hybrid Simulation

2019 ◽  
Vol 9 (21) ◽  
pp. 4495 ◽  
Author(s):  
Mucha
Keyword(s):  
Finite Element ◽  
Real Time ◽  
Degrees Of Freedom ◽  
Computational Cost ◽  
Hybrid Simulation ◽  
Time Algorithm ◽  
High Accuracy ◽  
Small Time ◽  
Fe Model ◽  
Artificial Neural

Hybrid simulation is a technique for testing mechanical systems. It applies to structures with elements hard or impossible to model numerically. These elements are tested experimentally by straining them by means of actuators, while the rest of the system is simulated numerically using a finite element method (FEM). Data is interchanged between experiment and simulation. The simulation is performed in real-time in order to accurately recreate the dynamic behavior in the experiment. FEM is very computationally demanding, and for systems with a great number of degrees of freedom (DOFs), real-time simulation with small-time steps (ensuring high accuracy) may require powerful computing hardware or may even be impossible. The author proposed to swap the finite element (FE) model with an artificial neural network (ANN) to significantly lower the computational cost of the real-time algorithm. The presented examples proved that the computational cost could be reduced by at least one number of magnitude while maintaining high accuracy of the simulation; however, obtaining appropriate ANN was not trivial and might require many attempts.

scholarly journals Application of Deep Learning on UAV-Based Aerial Images for Flood Detection

Smart Cities ◽  
2021 ◽  
Vol 4 (3) ◽  
pp. 1220-1243
Author(s):  
Hafiz Suliman Munawar ◽  
Fahim Ullah ◽  
Siddra Qayyum ◽  
Amirhossein Heravi
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Imaging System ◽  
Learning Algorithm ◽  
Smart Cities ◽  
Aerial Images ◽  
Management Systems ◽  
Training Dataset ◽  
Delayed Response

Floods are one of the most fatal and devastating disasters, instigating an immense loss of human lives and damage to property, infrastructure, and agricultural lands. To cater to this, there is a need to develop and implement real-time flood management systems that could instantly detect flooded regions to initiate relief activities as early as possible. Current imaging systems, relying on satellites, have demonstrated low accuracy and delayed response, making them unreliable and impractical to be used in emergency responses to natural disasters such as flooding. This research employs Unmanned Aerial Vehicles (UAVs) to develop an automated imaging system that can identify inundated areas from aerial images. The Haar cascade classifier was explored in the case study to detect landmarks such as roads and buildings from the aerial images captured by UAVs and identify flooded areas. The extracted landmarks are added to the training dataset that is used to train a deep learning algorithm. Experimental results show that buildings and roads can be detected from the images with 91% and 94% accuracy, respectively. The overall accuracy of 91% is recorded in classifying flooded and non-flooded regions from the input case study images. The system has shown promising results on test images belonging to both pre- and post-flood classes. The flood relief and rescue workers can quickly locate flooded regions and rescue stranded people using this system. Such real-time flood inundation systems will help transform the disaster management systems in line with modern smart cities initiatives.

Seismic Performance of Storage Tank Isolated by Different Isolators Through Real-Time Hybrid Simulation

2021 ◽  
pp. 2150190
Author(s):  
Zhenyun Tang ◽  
Yue Hong ◽  
Liusheng He ◽  
Zhenbao Li
Keyword(s):  
Real Time ◽  
Seismic Performance ◽  
Storage Tank ◽  
Shear Force ◽  
Experimental Testing ◽  
Reduction Rate ◽  
Hybrid Simulation ◽  
Seismic Energy ◽  
Overturning Moment ◽  
Friction Pendulum

There are three kinds of isolators commonly used in storage tank, friction pendulum bearing (FPB), laminated rubber bearing (LRB), and variable curvature friction pendulum bearing (VCFPB), respectively. Real-time hybrid simulation is conducted in this paper to compare the seismic performance of the storage tank isolated by the above three types of bearings. The storage tank is used as the physical substructure for experimental testing, and the isolators are adopted as the numerical substructure for numerical simulation. The isolation performance is estimated by the following perspectives: deformation of the isolator, shear force, overturning moment, and input energy. Test results show that the deformation of LRB is the largest, which can be twice that of FPB, and that larger deformation will enlarge the seismic energy input into the storage tank. Moreover, the low-frequency components of shear force and overturning moment are amplified by LRB. In contrast, the FPB and VCFPB have a good performance on all frequency bands. Particularly, the softening mechanism enables VCFPB to have better seismic performance and have a reduction rate of about twice that of LRB.

Experimental Evaluation of an Adaptive Actuator Control Scheme for Real-Time Tests of Large-Scale Magneto-Rheological Damper Under Variable Current Inputs

2009 ◽  
Author(s):  
Cheng Chen ◽  
James M. Ricles
Keyword(s):  
Real Time ◽  
Hybrid Simulation ◽  
Mr Damper ◽  
Structural Systems ◽  
Adaptive Compensation ◽  
Hydraulic Actuator ◽  
Control Laws ◽  
Mr Dampers ◽  
Magneto Rheological ◽  
Actuator Control

A Magneto-Rheological (MR) fluid damper is a semi-active device for vibration control of engineering structures subjected to dynamic loading. The characteristics of MR dampers vary under different current inputs to achieve optimized vibration control of structural systems. Experimental evaluation of MR dampers under different control laws is necessary before the device can be accepted by the practical design community. Real-time hybrid simulation provides an economical and efficient dynamic testing technique by accounting for the damper rate-dependency and the damper-structure interaction. A successful real-time hybrid simulation requires accurate actuator control to achieve reliable experiment results. A servo-hydraulic actuator usually introduces a time delay due to servo-hydraulic dynamics. The variable current inputs induced by semi-active control laws would pose additional challenges for actuator control by introducing variable delay in a real-time hybrid simulation. In this paper an adaptive compensation technique is experimentally evaluated for real-time hybrid simulation involving an MR damper under variable current inputs. Predefined band-limited white noise is used as the displacement command for the servo-hydraulic actuator and current command for the MR damper. The adaptive compensation scheme is demonstrated to achieve accurate actuator control and therefore shows great potential for real-time hybrid simulation of structural systems with semi-active energy dissipation devices.

Seismic Evaluation of Large Server Computer Structure

2009 ◽  
Author(s):  
Budy D. Notohardjono ◽  
Shawn Canfield ◽  
James A. Cooke
Keyword(s):  
Finite Element ◽  
Finite Element Model ◽  
Experimental Testing ◽  
Element Model ◽  
Frame Structure ◽  
Seismic Evaluation ◽  
Modal Data ◽  
Server Systems ◽  
Computer Structure ◽  
Vibration Tests

This paper discusses the analysis and verification of a finite element model which simulates the robustness of a high end computer server structure during a severe seismic event. The server consists of the frame which is the structure that components are installed into, such as processor units, input-output units and power components. The finite element modeling of this server frame is presented here not only to inform on creating an accurate model for simulation purposes, but also to provide guidelines as to the critical factors in setting up a large assembly finite element model (FEM) and to establish the optimum methodology for modeling this complex assembly with the available analysis software tools. For verification, the simulated modal data is compared to both modal data measured from an instrumented impact hammer, and to measured swept sine data. The simulated results compare favorably with the measured data, and it has been determined that location and integrity of the welded connections are critical for an accurate vibration response of the finite element model. The analysis frame model was subjected to loads and environmental conditions similar to those endured under horizontal table vibration tests and seismic events. The results of the experimental testing and simulations were compared and proved to be in a good correlation. Based on this verified finite element model, any additional redesign of the frame structure and its stiffening members can proceed very efficiently. This design study builds toward the objective of constructing a verified model of the server frame and components which will lead to a guideline for implementing stiffener designs on high-end server systems.

scholarly journals Numerical prediction of mechanical properties of zirconium alloy with hydrides using finite element method

Energetika ◽  
2018 ◽  
Vol 64 (1) ◽  
Author(s):  
Remigijus Janulionis ◽  
Gintautas Dundulis ◽  
Rita Kriūkienė ◽  
Albertas Grybėnas
Keyword(s):  
Finite Element ◽  
Material Properties ◽  
Zirconium Alloy ◽  
Experimental Testing ◽  
Zirconium Alloys ◽  
Numerical Prediction ◽  
Alternative Methods ◽  
Zirconium Hydride ◽  
Fe Method

During nuclear power plant (NPP) operation, degradation effects like ageing, corrosion, fatigue, and others may significantly impact component integrity. One of the degradation mechanisms is hydrogen absorption. High levels of hydrogen in zirconium alloys can lead to the formation of zirconium hydrides and that can influence material properties. Therefore, determination of material properties under different levels of hydrogen concentration in zirconium alloys is important. It is not always possible to conduct an experimental testing. Therefore, there is a need for alternative methods for determination of material properties. This article presents the numerical prediction of material properties of zirconium 2.5% niobium alloy with hydrides. According to the objective of the work, numerical prediction was performed using the finite element (FE) method. This was done by creating a finite element model of zirconium hydride embedded in zirconium alloy. The geometry and size of hydride were measured from a real specimen. The size of zirconium alloy surrounding the hydride was selected in such a way that hydride volume part in the model would match experimental measurements. The prognosis results were compared with the experimental data.

Sign in / Sign up

Export Citation Format

Share Document