LPG/LNG Gemilerinin Ana Makine Güç Tahmini için Yapay Sinir Ağı Modelinin Eğitim Aşamasında GA ve PSO Algoritmalarının Performans Analizi

Öz

Ülkelerin LPG/LNG gibi alternatif enerji kaynaklarına olan talebinin artmasıyla birlikte LPG/LNG gemilerinin üretimi de artmıştır. Sıvılaştırılmış gaz taşıyıcıları olarak bilinen bu gemilerin kendilerine özgü özellikleri ve tasarımları bulunmaktadır. Ayrıca tehlikeli yük taşımaları nedeniyle yüksek güvenlik standartları dikkate alınarak tasarlanmaktadırlar. Ana makine için gereken gücün belirlenmesi, tasarım sürecinin ilk aşamalarında önemli adımlardan biridir. Bu çalışmada, yapay sinir ağları (YSA) kullanılarak LPG/LNG gemileri için ana makine tahmin modeli üretilmiştir. YSA eğitim sürecinde temel geri yayılım algoritması (BP) ve Levenberg–Marquardt (LM) algoritmalarına ek olarak son yıllarda giderek popülerlik kazanan ve çeşitli disiplinlerde başarıyla uygulanan sezgisel algoritmalar da kullanılmıştır. Bu bağlamda en popüler algoritmalar olan Genetik Algoritma (GA) ve Parçacık Sürü Optimizasyonu (PSO) ile de YSA eğitimi gerçekleştirilmiştir. Bu çalışmanın temel amacı, tahmin modeli eğitiminde sezgisel algoritmaların performansını araştırmaktır. Sonuçlar, sezgiler algoritmalar açısından PSO algoritmasının üstünlüğü göstermiştir. PSO algoritması gradyan tabanlı algoritmaları ile karşılaştırıldığında BP algoritmasına göre üstün çıkarken, LM algoritmasından daha kötü bir performans göstermiştir. LM ile eğitilen ANN modeliyle küresel bir çözüm elde edilmiş fakat sonuçların istatistiksel analizi, LM algoritmasının standart sapmasının yüksek olduğunu ortaya koymuştur. Buna karşılık, PSO algoritması tutarlı bir şekilde daha düşük bir standart sapma değeriyle makul sonuçlar üretmiştir. Friedman testi sonuçları da PSO algoritmasının LM ile rekabet edeceğini göstermiştir.

Anahtar Kelimeler

• LPG/LNG gemileri
• Ana makine güç tahmini
• Yapay sinir ağı
• Öğrenme algoritması
• Algoritma performansı

Performance Analysis of GA and PSO Algorithms in Training Phase of Artificial Neural Network Model for Estimating Main Engine Power of LPG/LNG Ships

Abstract

The production of LPG/LNG ships has also increased with the increasing demand for alternative energy sources such as LPG/LNG in countries. These ships, known as liquefied gas carriers, have their own characteristics and designs. In addition, they are designed by taking into account high safety standards because they carry dangerous cargo. Determining the required power for the main engine is one of the important steps in the initial stages of the design process. In this study, a main engine prediction model for LPG/LNG ships was produced using artificial neural networks (ANN). In the ANN training process, in addition to the basic backpropagation algorithm (BP) and Levenberg–Marquardt (LM) algorithms, heuristic algorithms, which have become increasingly popular in recent years and have been successfully applied in various disciplines, were also used. In this context, ANN training was also carried out with the most popular algorithms, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The main purpose of this study is to investigate the performance of heuristic algorithms in prediction model training. The results showed the superiority of the PSO algorithm among the intuitive algorithms. When comparing PSO with gradient-based algorithms, the PSO algorithm was superior to the BP algorithm, but performed worse than the LM algorithm. A global solution was obtained with the ANN model trained with LM, but the statistical analysis of the results revealed that the standard deviation of the LM algorithm was high. In contrast, the PSO algorithm consistently produced reasonable results with a lower standard deviation value. The Friedman test results also showed that the PSO algorithm would compete with LM.

Keywords

• LPG/LNG ships
• Main engine power prediction
• Artificial neural network
• Learning algorithm
• Algorithm performance

Kaynakça

1. Akyuz E., Celik M. Application of CREAM human reliability model to cargo loading process of LPG tankers. Journal of Loss Prevention in the Process Industries 2015; 34: 39-48.
2. Aljarah I., Faris H., Mirjalili S. Optimizing connection weights in neural networks using the whale optimization algorithm. Soft Computing 2018; 22: 1-15.
3. Ateş KT. Çok katmanlı yapay sinir ağı modeli ve kültürel algoritma modeli kullanılarak geliştirilen melez yöntem ile kısa vadeli fotovoltaik enerji santrali çıkış gücü tahmini. Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi 2022; 5(1): 342-354.
4. Bai X., Lam JSL. An integrated analysis of interrelationships within the very large gas carrier (VLGC) shipping market. Maritime Economics & Logistics 2019; 21(3): 372-389.
5. Cepowski T. Prediction of the main engine power of a new container ship at the preliminary design stage. Management Systems in Production Engineering 2017; 25(2): 97-99.
6. Cepowski T. Regression formulas for the estimation of engine total power for tankers, container ships and bulk carriers on the basis of cargo capacity and design speed. Polish Maritime Research 2019; 26(101): 82-94.
7. Cepowski T. The prediction of ship added resistance at the preliminary design stage by the use of an artificial neural network. Ocean Engineering 2020; 195: 106657.
8. Cepowski T., Chorab P. The use of artificial neural networks to determine the engine power and fuel consumption of modern bulk carriers, tankers and container ships. Energies 2021; 14(16): 4827.

9. Chen JF., Do QH., Hsieh HN. Training artificial neural networks by a hybrid PSO-CS algorithm. Algorithms 2015; 8(2): 292-308.
10. Çirçir E., Gürgen S. Artificial neural network approach for main engine power prediction of general cargo vessels. Mersin University Journal of Maritime and Logistics Research 2024; 6(2): 113-129.
11. Eberhart R., Kennedy J. A new optimizer using particle swarm theory. Sixth International Symposium on Micro Machine and Human Science, 04-06 October 1995, 39-43, Nagoya, Japan.
12. Ekinci S., Çelebi UB., Bal M., Amasyali MF., Boyaci UK. Predictions of oil/chemical tanker main design parameters using computational intelligence techniques. Applied Soft Computing 2011; 11(2): 2356-2366. Fam ML., Tay ZY., Konovessis D. An artificial neural network for fuel efficiency analysis for cargo vessel operation. Ocean Engineering 2022; 264: 112437. Fernández IA., Gómez MR., Gómez JR., Insua ÁB. Review of propulsion systems on LNG carriers. Renewable and Sustainable Energy Reviews 2017; 67: 1395-1411. Gonzalez C., Pérez-Labajos C. Typology of the gas carrier fleet. Journal of Maritime Research 2017; 14 (1): 97-103. Göksu B., Erginer K. Prediction of ship main engine failures by artificial neural networks. Journal of ETA Maritime Science 2020; 8(2): 98-113.
13. Guzelbulut C., Badalotti T., Fujita Y., Sugimoto T., Suzuki K. Artificial neural network-based route optimization of a wind-assisted ship. Journal of Marine Science and Engineering 2024; 12(9): 1645.
14. Gülmez B. Market zinciri ürün dağıtımı probleminin farklı genetik algoritma versiyonları ile çözümü ve karşılaştırması. Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi 2023; 6(1): 180-196.
15. Güneş Ü. Estimating bulk carriers’ main engine power and emissions. Brodogradnja: An International Journal of Naval Architecture and Ocean Engineering for Research and Development 2023; 74 (1): 85-98.
16. Güneş Ü., Başhan V., Karakurt AS. Predicting tanker main engine power using regression analysis and artificial neural networks. Sigma Journal of Engineering and Natural Sciences 2023; 41(2): 216-225.
17. Gürgen S. Yapay zeka yaklaşımları ile gemi ana makinesinin belirlenmesi ve optimum organik rankine çevrimli atık ısı geri kazanım sisteminin kurulması. Karadeniz Teknik Üniversitesi, Fen Bilimleri Enstitüsü, Doktora Tezi, Trabzon, Türkiye, 2021.
18. Gürgen S. Artificial neural network-based approach to predict main engine power in reefer ships. 3rd International Mediterranean Congress, 17-18 April 2023, 718-725, Mersin, Türkiye.
19. Gürgen S., Altın I., Ozkok M. Prediction of main particulars of a chemical tanker at preliminary ship design using artificial neural network. Ships and Offshore Structures 2018; 13(5): 459-465.
20. Ho YC., Pepyne DL. Simple explanation of the no-free-lunch theorem and its implications. Journal of Optimization Theory and Applications 2002; 115: 549-570.
21. Holland JH. Adaptation in natural and artificial systems: an introductory analysis with applications to biology, control, and artificial intelligence. MIT press; 1992.
22. Karaboga D., Akay B., Ozturk C. Artificial bee colony (ABC) optimization algorithm for training feed-forward neural networks. Modeling Decisions for Artificial Intelligence 4th International Conference, 16-18 August 2007, 318-329, Kitakyushu, Japan.
23. Korlak PK. Prediction of the very-and ultra-large container ships’ electricity generation capacity at the initial design stage. Naše More International Journal of Maritime Science & Technology 2022; 69(2): 103-113.
24. Koycegiz C., Buyulyildiz M. Estimation of streamflow using different artificial neural network models. Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi 2022; 5(3): 1141-1154.
25. Le LT., Nguyen H., Dou J., Zhou J. A comparative study of PSO-ANN, GA-ANN, ICA-ANN, and ABC-ANN in estimating the heating load of buildings’ energy efficiency for smart city planning. Applied Sciences 2019; 9(13): 2630.
26. Mangudis L. Safety of logistics processes in the supply of liquid natural gas (LNG). European Research Studies Journal 2024; 27(Special A): 828-838.
27. Marques CH., Caprace JD. Exploring various sizes of liquefied gas carriers by an optimisation approach to early-stage project. Applied Ocean Research 2020; 97: 102079.
28. Matulja D., Dejhalla R., Bukovac O. Application of an artificial neural network to the selection of a maximum efficiency ship screw propeller. Journal of Ship Production and Design 2010; 26(03): 199-205.
29. McCulloch WS., Pitts W. A logical calculus of the ideas immanent in nervous activity. The Bulletin of Mathematical Biophysics 1943; 5: 115-133.
30. Miana M., Del Hoyo R., Rodrigálvarez V., Valdés JR., Llorens R. Calculation models for prediction of Liquefied Natural Gas (LNG) ageing during ship transportation. Applied Energy 2010; 87(5): 1687-1700.
31. Michail NA., Melas KD. Geopolitical risk and the LNG-LPG trade. Peace Economics, Peace Science and Public Policy 2022; 28(3): 243-265.
32. Mirjalili S. Evolutionary algorithms and neural networks theory and applications. Switzerland: Springer; 2019. Mirjalili S., Hashim SZM., Sardroudi HM. Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm. Applied Mathematics and Computation 2012; 218(22): 11125-11137.
33. Okumuş F., Ekmekçioğlu A., Kara SS. Modelling ships main and auxiliary engine powers with regression-based machine learning algorithms. Polish Maritime Research 2021; 1: 83-96.
34. Ozsari I. Predicting main engine power and emissions for container, cargo, and tanker ships with artificial neural network analysis. Brodogradnja: An International Journal of Naval Architecture and Ocean Engineering for Research and Development 2023; 74(2): 77-94.
35. Sea-Web Ships. 2021. Data base: https://maritime.ihs.com (accessed at 2021)
36. Shuai Y., Li G., Cheng X., Skulstad R., Xu J., Liu H., Zhang H. An efficient neural-network based approach to automatic ship docking. Ocean Engineering 2019; 191: 106514.
37. Şahin B., Gürgen S., Ünver B., Altin İ. Forecasting the baltic dry index by using an artificial neural network approach. Turkish Journal of Electrical Engineering and Computer Sciences 2018; 26 (3): 1673-1684.
38. Uzlu E., Akpınar A., Özturk HT., Nacar S., Kankal M. Estimates of hydroelectric generation using neural networks with the artificial bee colony algorithm for Turkey. Energy 2014; 69: 638-647.
39. Wolpert DH., Macready WG. No free lunch theorems for optimization. IEEE Transactions on Evolutionary Computation 1997; 1(1): 67-82.