Metaheuristic approach for an artificial neural network: Exergetic sustainability and environmental effect of a business aircraft
In the current study, exergetic metaheuristic design for a business jet aircraft are presented for the prediction of exergetic sustainability index (ESI) and environmental effect factor (EEF) with the aid of artificial neural network (ANN) models at various flight phases. In this respect, real databases of ESI and EEF with regards to several engine parameters achieved by multiple number of runs of a business aircraft engine at various settings have been utilized to develop hybrid genetic algorithm (GA)-ANN models. Adoption of a metaheuristics based optimization on the developed Multilayer perceptron (MLP) ANN models has yielded optimum initial network weights, biases, step-size, and momentum rate for the back-propagation (BP) training algorithm as well as the optimum number of neurons in the hidden layer(s) in terms of network topology design. An error analysis has revealed that linear correlation coefficient values between the reference real data and predicted ESI and EEF values have been attained as 0.999862 and 0.999986, respectively. For both models, more accurate testing results have been achieved for one-hidden-layer networks compared to two-hidden-layer ones. Consequently, optimization of ANN models by GAs has enhanced the time effectiveness and accuracy of the derived models ensuring a drop-off in the testing phase errors.
- Record URL:
- Record URL:
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/issn/13619209
-
Supplemental Notes:
- © 2018 Elsevier Ltd. All rights reserved. Abstract reprinted with permission of Elsevier.
-
Authors:
- Baklacioglu, Tolga
- Turan, Onder
- Aydin, Hakan
- Publication Date: 2018-8
Language
- English
Media Info
- Media Type: Web
- Features: Appendices; Figures; Glossary; References; Tables;
- Pagination: pp 445-465
-
Serial:
- Transportation Research Part D: Transport and Environment
- Volume: 63
- Issue Number: 0
- Publisher: Elsevier
- ISSN: 1361-9209
- Serial URL: http://www.sciencedirect.com/science/journal/13619209
Subject/Index Terms
- TRT Terms: Business aircraft; Environmental impacts; Genetic algorithms; Heuristic methods; Neural networks; Optimization; Sustainable transportation; Vehicle design
- Uncontrolled Terms: Metaheuristics
- Subject Areas: Aviation; Data and Information Technology; Design; Environment;
Filing Info
- Accession Number: 01675376
- Record Type: Publication
- Files: TRIS
- Created Date: Jul 19 2018 2:45PM