Finite element and neural network approximations to measure forces using six-component wind tunnel balance

2019 ◽  
Author(s):  
M. N. Danilov ◽  
P. P. Bardaev
Keyword(s):  
Neural Network ◽  
Finite Element ◽  
Wind Tunnel ◽  
Wind Tunnel Balance

scholarly journals Numerical Computation and Analysis of Electromagnetic Field in Magnetic Suspension and Balance System

2021 ◽  
Vol 7 (3) ◽  
pp. 33
Author(s):  
Mingda Zhai ◽  
Wentao Xia ◽  
Zhiqiang Long ◽  
Fengshan Dou
Keyword(s):  
Finite Element ◽  
Electromagnetic Field ◽  
Wind Tunnel ◽  
Numerical Computation ◽  
Electromagnetic Force ◽  
Magnetic Suspension ◽  
Key Factors ◽  
New Device ◽  
Wind Tunnel Balance ◽  
Suspension Model

The magnetic suspension wind tunnel balance (MSBS) is an entirely new device for aerodynamic measurement, and it makes the best of the electromagnetic force to suspend the aircraft model in the wind tunnel without contact. Compared with conventional wind tunnel balance, it absolutely abandons the model support and airflow interference. Therefore, the aerodynamic measurement environment is more authentic and the aerodynamic measurement results are more accurate. The electromagnetic field in MSBS plays a major role in bearing the force of wind. The numerical computation and finite element numerical analysis are performed to investigate key factors of electromagnetic force under different conditions. The calculation results based on finite element method (FEM) have revealed that the diameter and the spacing of of the axial coil, the number of segments and the pitch angle of the suspension model are key factors of electromagnetic force. Based on the above key factors, the structure of the magnetic suspension balance is optimized to maximize the electromagnetic force under multiple constraints.

scholarly journals Prediction of Stress in Power Transformer Winding Conductors Using Artificial Neural Networks: Hyperparameter Analysis

Energies ◽  
2021 ◽  
Vol 14 (14) ◽  
pp. 4242
Author(s):  
Fausto Valencia ◽  
Hugo Arcos ◽  
Franklin Quilumba
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Power Transformer ◽  
Network Models ◽  
Activation Function ◽  
Neural Network Models ◽  
Transformer Winding ◽  
Artificial Neural ◽  
Femm Software

The purpose of this research is the evaluation of artificial neural network models in the prediction of stresses in a 400 MVA power transformer winding conductor caused by the circulation of fault currents. The models were compared considering the training, validation, and test data errors’ behavior. Different combinations of hyperparameters were analyzed based on the variation of architectures, optimizers, and activation functions. The data for the process was created from finite element simulations performed in the FEMM software. The design of the Artificial Neural Network was performed using the Keras framework. As a result, a model with one hidden layer was the best suited architecture for the problem at hand, with the optimizer Adam and the activation function ReLU. The final Artificial Neural Network model predictions were compared with the Finite Element Method results, showing good agreement but with a much shorter solution time.

A New Multi-Component-Wind-Tunnel-Balance for Automotive Windtunnels

1995 ◽  
Author(s):  
L. Polansky ◽  
W. Matich ◽  
J. T. Kutney
Keyword(s):  
Wind Tunnel ◽  
Wind Tunnel Balance

An Interlocking Ligamentous Spinal Disk Arthroplasty with Neural Network Infrastructure

2010 ◽  
Vol 7 ◽  
pp. 55-79 ◽  
Author(s):  
Philip Boughton ◽  
James Merhebi ◽  
C. Kim ◽  
G. Roger ◽  
Ashish D. Diwan ◽  
...  
Keyword(s):  
Neural Network ◽  
Finite Element ◽  
Axial Compression ◽  
Wire Array ◽  
Axial Strain ◽  
Compressive Strain ◽  
Niti Wire ◽  
Motion Segment ◽  
Flexion Extension

An elastomeric spinal disk prosthesis design (BioFI™) with vertebral interlocking anchors has been modified using an embedded TiNi wire array. Bioinert styrenic block copolymer (Kraton®) and polycarbonate urethane (Bionate®) thermoplastic elastomer (TPE) matrices were utilized. Fatigue resistant NiTi wire was pretreated to induce superelastic martensitic microstructure. Stent-like helical structures were produced for incorporation within homogenous TPE matrix. Composite prototypes were fabricated in a vacuum hot press using transfer moulding techniques. Implant prototypes were subject to axial compression using a BOSE ® ELF3400. The NiTi reinforced implants exhibited reduction in axial strain, compliance, and creep compared to TPE controls. The axial properties of the NiTi reinforced Bionate® BioFI™ implant best approximated those of a spinal disk followed by Kraton®-NiTi, Bionate® and Kraton® prototypes. An ovine lumbar segment biomechanical model was used to characterize the disk prosthesis prototypes. Specimens were subject to 7.5Nm pure moments in axial rotation, flexion-extension and lateral bending with a custom jig mounted on an Instron® 8874. The motion preserving ligamentous nature of this arthroplasty prototype was not inhibited by NiTi reinforcement. Joint stiffness for all prototypes was significantly less than the intact and discectomy controls. This was due to lack of vertebral anchor rigidity rather than BioFI™ motion segment matrix type or reinforcement. Implant stress profiles for axial compression and axial torsion conditions were obtained using finite element methods. The biomechanical testing and finite element modelling both support existing BioFI™ design specifications for higher modulus vertebral anchors, endplates and motion segment periphery with gradation to a low modulus core within the motion segment. This closer approximation of the native spinal disk form translates to improvements in prosthesis biomechanical fidelity and longevity. Axial compressive strain induced within a TiNi reinforced Kraton® BioFI™ was found to be linearly proportional to the NiTi helical coil electrical resistance. This neural network capability delivers opportunities to monitor and telemeterize in situ multiaxis joint structural performance and in vivo spine biomechanics.

Sign in / Sign up

Export Citation Format

Share Document