Prediction of Natural Gas Wall-Impingement Spray Characteristics by ANN Model 2023-32-0011
In this study, the effect of injection pressure, impingement distance and angle, wall temperature on the macroscopic of wall impingement were investigated experimentally, predicted by using deep neural network in the MATLAB environment. With respect to obtaining data from experiments, input factors affecting impingement phenomena are trained, validated to develop model, which was applied to estimate output such as spray tip penetration and height. According to the results, the estimate parameters by coefficient of determination, root mean square error between 0.998 and 0.029. The ANN_GA model is found to be an effective tool to predict spray behaviors output with minimal experimentation.
Author(s):
Quangkhai Pham, Byungchul Choi, Suhan Park
Affiliated:
Department of Mechanical Engineering, Graduate School of Cho, School of Mechanical Engineering, Chonnam National Universit, School of Mechanical and Aerospace Engineering, Konkuk Unive
Pages: 6
Event:
2023 JSAE/SAE Powertrains, Energy and Lubricants International Meeting
ISSN:
0148-7191
e-ISSN:
2688-3627
Related Topics:
Neural networks
Natural gas
Nozzles
Computer simulation
SAE MOBILUS
Subscribers can view annotate, and download all of SAE's content.
Learn More »
