ANN-Based modeling and performance analysis of pyrolytic oil production system

Abstract

In this study, the modeling of the Pyrolytic oil production system using Artificial Neural Networks (ANNs) has been conducted with oak acorn, which can be considered as non-wood forest product. The parameters used in the pyrolytic oil production system have been determined as reactor temperature, nitrogen gas flow rate, biomass particle size, and heating rate. In experimental studies, the highest pyrolytic oil production has been achieved at 500 °C temperature, 1.5 L/min nitrogen gas flow rate, 5 °C/min heating rate, and 0-2 mm biomass particle size, with a product yield of 17.83%. 164 different Multi-Layer Feed Forward (MLFF) ANN-based network architectures have been trained for 20,000 iterations using the data obtained from the pyrolytic oil production system. In the training process, various network architectures including activation functions such as TanSig, LogSig, and RadBas with one or two hidden layers have been utilized. According to the results obtained from the studies, the Multi-Layer Feed Forward ANN-based Pyrolytic Oil Production System structure, which has a single hidden layer and contains 16 LogSig activation function neurons, has been the network structure with the best performance with the value of 1.08E-15.

Keywords

• Acorn
• Activation functions
• Artificial neural networks
• Pyrolytic oil production

Pirolitik yağ üretim sisteminin YSA-Tabanlı modellenmesi ve performans analizi

Öz

Bu çalışmada, odun dışı orman ürünü olarak değerlendirilebilecek meşe palamudu ile Yapay Sinir Ağları (YSA-Artificial Neural Network) kullanılarak Pirolitik yağ üretim sisteminin modellenmesi gerçekleştirilmiştir. Pirolitik yağ üretim sisteminde kullanılan parametreler; reaktör sıcaklığı, azot gazı akış hızı, biyokütle parçacık boyutu ve ısıtma hızı olarak belirlenmiştir. Deneysel çalışmalarda, 500 °C sıcaklıkta, 1.5 L/dk. azot gazı akış oranı, 5 °C/dk. ısıtma oranı ve 0-2 mm biyokütle tanecik boyutunda en yüksek pirolitik yağ üretimi gerçekleştirilmiş olup ürün verimi %17.83 elde edilmiştir. Pirolitik yağ üretim sisteminden elde edilen veriler kullanılarak 164 farklı Çok Katmanlı İleri Beslemeli (MLFF) ANN-tabanlı ağ mimarisi 20.000 iterasyon için eğitilmiştir. Eğitim sürecinde, bir veya iki gizli katmana sahip TanSig, LogSig ve RadBas transfer fonksiyonları içeren farklı ağ yapıları kullanılmıştır. Çalışmalardan elde edilen sonuçlara göre, tek gizli katmanlı olan ve 16 LogSig aktivasyon fonksiyonlu nöron içeren Çok Katmanlı İleri Beslemeli YSA-tabanlı Pirolitik Yağ Üretim Sistemi yapısı 1.08E-15 değeri ile en iyi performans elde edilen ağ yapısı olmuştur.

Anahtar Kelimeler

• Meşe palamudu
• Aktivasyon fonksiyonları
• Yapay sinir ağları
• Pirolitik yağ üretimi

References

1. [1] Güney B, Aladağ A. “Microstructural analysis of liquefied petroleum gas vehicle emissions, one of the anthropogenic environmental pollutants”. International Journal of Environmental Science and Technology, 19(1), 249-260, 2022.
2. [2] Abdeshahian P, Lim JS, Ho WS, Hashim H, Lee CT. “Potential of biogas production from farm animal waste in Malaysia”. Renewable and Sustainable Energy Reviews, 60(C), 714-723, 2016.
3. [3] Güney B, Öz A. “Microstructure and chemical analysis of NOx and particle emissions of diesel engines”. International Journal of Automotive Engineering and Technologies, 9(2), 105-112, 2020.
4. [4] Güney B, Aladağ A. “Microstructural characterization of particulate matter from gasoline-fuelled vehicle emissions”. Journal of Engineering Research and Reports, 16(1), 29-39, 2020.
5. [5] Bridgwater AV. “Renewable fuels and chemicals by thermal processing of biomass”. Chemical Engineering Journal, 91(2-3), 87-102, 2003.
6. [6] Kan T, Strezov V, Evans TJ. “Lignocellulosic biomass pyrolysis: A review of product properties and effects of pyrolysis parameters”. Renewable and Sustainable Energy Reviews, 57, 1126-1140, 2016.
7. [7] Kar Y. “Catalytic cracking of pyrolytic oil by using bentonite clay for green liquid hydrocarbon fuels production”. Biomass Bioenergy, 119, 473-479, 2018.
8. [8] Callioglu H, Muftu S, Koplay CN. “Comparison of vibration values of rotating discs with variable parameters obtained by finite element analysis modeling with different machine learning algorithms”. Multidiscipline Modeling in Materials and Structures, 21(1), 98-118 2025.

9. [9] Boğar E, Boğar ZÖ. “Türkiye’nin sektörel CO2 gazı salınımlarının yapay sinir ağları ile tahmini”. Akademia Disiplinlerarası Bilimsel Araştırmalar Dergisi, 3(2), 15-27, 2017.
10. [10] Akçay MŞ, Koyuncu İ, Alcin M, Tuna M. “FPGA tabanlı LogSig ve TanSig transfer fonksiyonlarının IQ-Math sayı standardında tasarımı ve gerçeklenmesi”. Journal of Materials and Mechatronics: A, 3(2), 225-239, 2022.
11. [11] Le QN, Jeon JW. “Neural-Network-Based Low-Speed-damping controller for stepper motor with an FPGA”. IEEE Transactions on Industrial Electronics, 57(9), 3167-3180, 2010.
12. [12] Alcin M, Koyuncu I, Tuna M, Varan M, Pehlivan I. “A novel high speed Artificial Neural Network-based chaotic True Random Number Generator on Field Programmable Gate Array”. International Journal of Circuit Theory and Applications, 47(3), 365-378, 2019.
13. [13] Şahin M, Oğuz Y, Büyüktümtürk F. “ANN-based estimation of time-dependent energy loss in lighting systems”. Energy Build, 116, 455-467, 2016.
14. [14] Ekici S, Jawzal H. “Breast cancer diagnosis using thermography and convolutional neural networks”. Med Hypotheses, 132, 1-14, 2020.
15. [15] Koyuncu I, Yilmaz C, Alcin M, Tuna M. “Design and implementation of hydrogen economy using artificial neural network on field programmable gate array”. International Journal Hydrogen Energy, 45(41), 20709-20720, 2020.
16. [16] Yilmaz C, Koyuncu I, Alcin M, Tuna M. “Artificial neural networks based thermodynamic and economic analysis of a hydrogen production system assisted by geothermal energy on field programmable gate array”. International Journal Hydrogen Energy, 44(33), 17443-17459, 2019.
17. [17] Koyuncu I, Rajagopal K, Alcin M, Karthikeyan A, Tuna M, Varan M. “Control, synchronization with linear quadratic regulator method and FFANN-based PRNG application on FPGA of a novel chaotic system”. European Physical Journal Special Topics, 230(7), 1915-1931, 2021.
18. [18] Tuna M, Karthikeyan A, Rajagopal K, Alçın M, Koyuncu İ. “Hyperjerk multiscroll oscillators with megastability: Analysis, FPGA implementation and A novel ANN-Ring-based true random number generator”. AEU-International Journal of Electronics and Communications, 112, 152941, 2019.
19. [19] Panda AK, Rout SK, Das AK. “Optimization of diesel engine performance and emission using waste plastic pyrolytic oil by ANN and its thermo-economic assessment”. Environmental Science and Pollution Research, 1, 1-15, 2023.
20. [20] Cao H, Xin Y, Yuan Q. “Prediction of biochar yield from cattle manure pyrolysis via least squares support vector machine intelligent approach”. Bioresource Technology, 202, 158-164, 2016.
21. [21] Sun Y, Liu L, Wang Q, Yang X, Tu X. “Pyrolysis products from industrial waste biomass based on a neural network model”. Journal Analytical Applied Pyrolysis, 120, 94-102, 2016.
22. [22] Sunphorka S, Chalermsinsuwan B, Piumsomboon P. “Artificial neural network model for the prediction of kinetic parameters of biomass pyrolysis from its constituents”. Fuel, 193, 142-158, 2017.
23. [23] Selvarajoo A, Muhammad D, Arumugasamy SK. “An experimental and modelling approach to produce biochar from banana peels through pyrolysis as potential renewable energy resources”. Modeling Earth Systems and Environment, 6, 115-128, 2020.
24. [24] Çepelioğullar Ö, Mutlu İ, Yaman S, Haykiri-Acma H. “Activation energy prediction of biomass wastes based on different neural network topologies”. Fuel, 220, 535-545, 2018.
25. [25] Mayol AP, Maningo JMZ, Chua-Unsu AGAY, Felix CB, Rico PI, Chua GS, Manalili EV, Fernandez DD, Cuello JL, Bandala AA, Ubando AT, Madrazo CF, Dadios E, Culaba AB. “Application of artificial neural networks in prediction of pyrolysis behavior for algal mat (LABLAB) biomass”. IEEE 2018 10th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM), Baguio City, Philippines, 29-02 December 2018.
26. [26] Aydinli B, Caglar A, Pekol S, Karaci A. “The prediction of potential energy and matter production from biomass pyrolysis with artificial neural network”. Energy Exploration and Exploitation, 35(6), 698-712, 2017.
27. [27] Bridgwater AV. “Review of fast pyrolysis of biomass and product upgrading”. Biomass and Bioenergy, 38, 68-94, 2012.
28. [28] Xiu S, Shahbazi A. “Bio-oil production and upgrading research: A review”. Renewable and Sustainable Energy Reviews, 16(7), 4406-4414, 2012.
29. [29] Fernandez-Akarregi AR, Makibar J, Lopez G, Amutio M, Olazar M. “Design and operation of a conical spouted bed reactor pilot plant (25 kg/h) for biomass fast pyrolysis”. Fuel Processing Technology, 112, 48-56, 2013.
30. [30] Erdem F. “Parameter estimation in Crystal Sugar production with MLR, ANN and ANFIS”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 28(7), 987-992, 2022.
31. [31] Gürgen S, Altın İ. “Bütanol-Diesel yakıtı kullanılan bir sıkıştırma ateşlemeli motorda motor performansı ve egzoz emisyonlarının yapay sinir ağları ile tahmini”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 24(4), 576-581, 2018.
32. [32] Abdelkareem MA, Soudan B, Mahmoud MS, Sayed ET, AlMallahi MN, Inayat A, Radi MAl, Olabi AG. “Progress of artificial neural networks applications in hydrogen production”. Chemical Engineering Research and Design, 182, 66-86, 2022.
33. [33] Nawaz A, Kumar P. “Optimization of process parameters of Lagerstroemia speciosa seed hull pyrolysis using a combined approach of Response Surface Methodology (RSM) and Artificial Neural Network (ANN) for renewable fuel production”. Bioresource Technology Reports, 18, 101110, 2022.
34. [34] Keskin RSO. “Predicting shear strength of reinforced concrete slender beams without shear reinforcement using artificial neural networks”. Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 23(3), 193-202, 2017.
35. [35 Sahin I, Koyuncu I. “Design and ımplementation of neural networks neurons with RadBas, LogSig, and TanSig Activation Functions on FPGA”. Electronics and Electrical Engineering, 4(120), 51-54, 2012.
36. [36] Bhattacharjee N, Biswas AB. “Pyrolysis of alternanthera philoxeroides (alligator weed): Effect of pyrolysis parameter on product yield and characterization of liquid product and bio char”. Journal of the Energy Institute, 91(4), 605-618, 2018.
37. [37] Santos J, Ouadi M, Jahangiri H, Hornung A. “Valorisation of lignocellulosic biomass investigating different pyrolysis temperatures”. Journal of the Energy Institute, 93(5), 1960-1969, 2020.
38. [38] Kim P, Weaver S, Noh K, Labbé N. “Characteristics of bio-oils produced by an intermediate semipilot scale pyrolysis auger reactor equipped with multistage condensers”. Energy & Fuels, 28(11), 6966-6973, 2014.
39. [39] Tabal A, Belyazid O, Dahman H, Berrich E, Jeguirim M, El Achaby M, Aboulkas A. “Intermediate pyrolysis of Ficus nitida wood in a fixed-bed reactor: effect of pyrolysis parameters on bio-oil and bio-char yields and properties”. Comptes Rendus Chimie, 26(S1), 1-17, 2023.