Artificial neural network model of molten carbonate fuel cells: Validation on experimental data

2019 ◽  
Author(s):  
Jarosław Milewski ◽  
Arkadiusz Szczȩśniak ◽  
Łukasz Szabłowski ◽  
Olaf Dybiński ◽  
Andrzej Miller
Keyword(s):  
Neural Network ◽  
Experimental Data ◽  
Artificial Neural Network ◽  
Fuel Cells ◽  
Network Model ◽  
Neural Network Model ◽  
Artificial Neural Network Model ◽  
Molten Carbonate ◽  
Molten Carbonate Fuel Cells ◽  
Artificial Neural

Artificial Neural Network Applied to Predicting the Surface Tension of Acoustically Levitated Droplets of Supercooling Nanofluids

NANO ◽  
2021 ◽  
pp. 2150108
Author(s):  
Baohui Wu ◽  
Yudong LIU ◽  
Dengshi Wang ◽  
Nan Jiang ◽  
Jie Zhang ◽  
...  
Keyword(s):  
Neural Network ◽  
Experimental Data ◽  
Surface Tension ◽  
Artificial Neural Network ◽  
Graphene Oxide ◽  
Network Model ◽  
Neural Network Model ◽  
Artificial Neural Network Model ◽  
Deionized Water ◽  
Artificial Neural

Droplet oscillation method is a noncontact experimental approach, which can be used to measure the surface tension of acoustically levitated droplet. In this paper, we obtained huge amounts of experimental data of deionized water and water-based graphene oxide nanofluids within the temperature range of [Formula: see text]8.2–[Formula: see text]C. Based on the experimental data, we analyzed the influence of droplet’s deformation and frequency shift phenomenon on the surface tension of levitated droplet. Eight parameters that strongly correlate with surface tension were found and used as input neurons of artificial neural network model to predict the surface tension of supercooling graphene oxide nanofluids. The experimental data of nonsupercooling graphene oxide nanofluids were used as training set to optimize artificial neural network model, and that of deionized water were served as validation set, which was used to verify the predictive ability of artificial neural network model. The root mean square error of the optimized artificial neural network model to validation set is only 0.2558[Formula: see text]mN/m, and the prediction values of the surface tension of supercooling deionized water were in good agreement with the theoretical values calculated by Vargaftik equation, which indicates that artificial neural network model can deal well with the complex nonlinear relationship. Afterwards, we successfully predicted the surface tension of supercooling nanofluids by means of the optimized artificial neural network model and obviously reduced the dispersion and deviation caused by droplet deformation and other problems during oscillation process.

scholarly journals Using Artificial Neural Networks to Predict Influences of Heterogeneity on Rock Strength at Different Strain Rates

Materials ◽  
2021 ◽  
Vol 14 (11) ◽  
pp. 3042
Author(s):  
Sheng Jiang ◽  
Mansour Sharafisafa ◽  
Luming Shen
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Strain Rate ◽  
Network Model ◽  
Neural Network Model ◽  
Artificial Neural Network Model ◽  
Rock Strength ◽  
Strain Rates ◽  
Filling Material ◽  
Artificial Neural

Pre-existing cracks and associated filling materials cause the significant heterogeneity of natural rocks and rock masses. The induced heterogeneity changes the rock properties. This paper targets the gap in the existing literature regarding the adopting of artificial neural network approaches to efficiently and accurately predict the influences of heterogeneity on the strength of 3D-printed rocks at different strain rates. Herein, rock heterogeneity is reflected by different pre-existing crack and filling material configurations, quantitatively defined by the crack number, initial crack orientation with loading axis, crack tip distance, and crack offset distance. The artificial neural network model can be trained, validated, and tested by finite 42 quasi-static and 42 dynamic Brazilian disc experimental tests to establish the relationship between the rock strength and heterogeneous parameters at different strain rates. The artificial neural network architecture, including the hidden layer number and transfer functions, is optimized by the corresponding parametric study. Once trained, the proposed artificial neural network model generates an excellent prediction accuracy for influences of high dimensional heterogeneous parameters and strain rate on rock strength. The sensitivity analysis indicates that strain rate is the most important physical quantity affecting the strength of heterogeneous rock.

Sign in / Sign up

Export Citation Format

Share Document