Prediction of Gross Calorific Value of Washed Turkish Lignite Coals with Support Vector Regression

Öz

In this study, the gross calorific value (GCV) of washed Turkish lignite coals was predicted by using dry-basis proximate analysis data of coal samples with machine learning methods. The data set was generated by using ash (A), volatile matter (VM), sulfur (S) and GCV variables obtained from the analysis results. The GCV prediction models were developed by applying Support Vector Regression (SVR) and three different Artificial Neural Networks (ANNs), namely Multi-Layer Perceptron (MLP), General Regression Neural Network (GRNN) and Radial Basis Function Neural Network (RBFN), separately to the data set. The generalization capability of the developed models was ensured by using 10-fold cross-validation, and the prediction accuracy of the models was calculated by using performance metrics Multiple Correlation Coefficient (R), Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). For GCV prediction, the results reveal that the SVR-based model performed slightly better than the ANN-based models and among the ANN-based models, the RBFN-based model performed better than MLP- and GRNN-based models.

Anahtar Kelimeler

• Gross calorific value,support vector regression,multi-layer perceptron,general regression neural network,radial basis function neural network

Yıkanmış Türk Linyit Kömürlerinin Üst Isıl Değerinin Destek Vektör Regresyonu ile Tahmini

Öz

Bu çalışmada yıkanmış Türk linyit kömürlerinin üst ısıl değeri (GCV), makine öğrenmesi yöntemleri ile kömür numunelerinin kuru baz kısa analiz sonuçları kullanılarak tahmin edilmiştir. Laboratuvar kömür analiz sonuçlarından elde edilen kül (A), uçucu madde (VM), kükürt (S) ve GCV değişkenleri kullanılarak veri kümesi oluşturulmuştur. Veri kümesine, Destek Vektör Regresyonu (SVR) ile Çok Katmanlı Algılayıcı (MLP), Genel Regresyon Sinir Ağı (GRNN) ve Radyal Temelli Fonksiyon Sinir Ağı (RBFN) olmak üzere üç farklı Yapay Sinir Ağı (ANN) uygulanarak GCV tahmin modelleri geliştirilmiştir. Geliştirilen modellerin performans genelleştirme kabiliyeti 10-katlı çapraz-doğrulama kullanılarak sağlanmış ve modellerin tahmin doğruluğu, performans ölçütleri Çoklu Korelasyon Katsayısı (R), Kök Ortalama Kare Hatası (RMSE), Ortalama Mutlak Hata (MAE) ve Ortalama Mutlak Yüzde Hata (MAPE) kullanılarak hesaplanmıştır. Sonuçlar, GCV tahmini için, tüm modeller arasında SVR tabanlı modelin ANN tabanlı modellere göre biraz daha iyi, ANN tabanlı modeller arasında ise RBFN tabanlı modelin MLP ve GRNN tabanlı modellere göre daha iyi performans gösterdiğini ortaya koymuştur.

Anahtar Kelimeler

• Üst ısıl değer,destek vektör regresyonu,çok katmanlı algılayıcı,genel regresyon sinir ağı,radyal temelli fonksiyon sinir ağı

Kaynakça

1. Abut, F., Akay, M. F., & George, J. (2016). Developing new VO2max prediction models from maximal, submaximal and questionnaire variables using support vector machines combined with feature selection. Computers in Biology and Medicine, 79(October), 182–192. https://doi.org/10.1016/j.compbiomed.2016.10.018
2. Açıkkar, M., Akay, M. F., Aktürk, E., & Güleç, M. (2013). Intelligent regression techniques for non-exercise prediction of VO2max. 2013 21st Signal Processing and Communications Applications Conference, SIU 2013. https://doi.org/10.1109/SIU.2013.6531534
3. Açıkkar, M., & Sivrikaya, O. (2018a). Artificial neural networks for estimation of the gross calorific value of Turkish lignite coals. 3rd International Mediterranean Science and Engineering Congress (IMSEC 2018), 1075–1079.
4. Açıkkar, M., & Sivrikaya, O. (2018b). Prediction of gross calorific value of coal based on proximate analysis using multiple linear regression and artificial neural networks. Turkish Journal of Electrical Engineering & Computer Sciences, 26(5), 2541–2552. https://doi.org/10.3906/elk-1802-50
5. Akande, K. O., Owolabi, T. O., Twaha, S., & Olatunji, S. O. (2014). Performance Comparison of SVM and ANN in Predicting Compressive Strength of Concrete. In IOSR Journal of Computer Engineering (Vol. 16). https://doi.org/10.9790/0661-16518894
6. Akhtar, J., Sheikh, N., & Munir, S. (2017). Linear regression-based correlations for estimation of high heating values of Pakistani lignite coals. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 39(10), 1063–1070. https://doi.org/10.1080/15567036.2017.1289283
7. Akkaya, A. V. (2009). Proximate analysis based multiple regression models for higher heating value estimation of low rank coals. Fuel Processing Technology, 90(2), 165–170. https://doi.org/10.1016/j.fuproc.2008.08.016
8. Arliansyah, J., & Hartono, Y. (2015). Trip Attraction Model Using Radial Basis Function Neural Networks. Procedia Engineering, 125, 445–451. https://doi.org/10.1016/J.PROENG.2015.11.117

9. Baydaroğlu, Ö., & Koçak, K. (2014). SVR-based prediction of evaporation combined with chaotic approach. Journal of Hydrology, 508, 356–363. https://doi.org/10.1016/J.JHYDROL.2013.11.008
10. Campbell, C. (2002). Kernel methods: a survey of current techniques. In Neurocomputing (Vol. 48). Retrieved from www.elsevier.com/locate/neucom
11. Channiwala, S. A., & Parikh, P. P. (2002). A unified correlation for estimating HHV of solid, liquid and gaseous fuels. Fuel, 81(8), 1051–1063. https://doi.org/10.1016/S0016-2361(01)00131-4
12. Chelgani, S. C., Mesroghli, S., & Hower, J. C. (2010). Simultaneous prediction of coal rank parameters based on ultimate analysis using regression and artificial neural network. International Journal of Coal Geology, 83(1), 31–34. https://doi.org/10.1016/j.coal.2010.03.004
13. Chen, S., Hong, X., & Harris, C. J. (2005). Orthogonal Forward Selection for Constructing the Radial Basis Function Network with Tunable Nodes. https://doi.org/10.1007/11538059_81
14. Chen, W., & Xu, R. (2010). Clean coal technology development in China. Energy Policy, 38(5), 2123–2130. https://doi.org/10.1016/j.enpol.2009.06.003
15. Feng, Q., Zhang, J., Zhang, X., & Wen, S. (2015). Proximate analysis based prediction of gross calorific value of coals: A comparison of support vector machine, alternating conditional expectation and artificial neural network. Fuel Processing Technology, 129, 120–129. https://doi.org/10.1016/j.fuproc.2014.09.001
16. Fu, J. (2016). Application of SVM in the estimation of GCV of coal and a comparison study of the accuracy and robustness of SVM. 2016 International Conference on Management Science and Engineering (ICMSE), 553–560. https://doi.org/10.1109/ICMSE.2016.8365486
17. Hadavandi, E., Hower, J. C., & Chelgani, · S Chehreh. (2017). Modeling of gross calorific value based on coal properties by support vector regression method. Earth Syst. Environ, 3(1), 37. https://doi.org/10.1007/s40808-017-0270-7
18. Heydari, A., Garcia, D. A., Keynia, F., Bisegna, F., & Santoli, L. De. (2019). Renewable Energies Generation and Carbon Dioxide Emission Forecasting in Microgrids and National Grids using GRNN-GWO Methodology. Energy Procedia, 159, 154–159. https://doi.org/10.1016/j.egypro.2018.12.044
19. Hsu, C.-W., Chang, C.-C., & Lin, C.-J. (2003). A Practical Guide to Support Vector Classification. Retrieved from http://www.csie.ntu.edu.tw/~cjlin
20. Huang, X., Liu, X., & Ren, Y. (2018). Enterprise credit risk evaluation based on neural network algorithm. Cognitive Systems Research, 52, 317–324. https://doi.org/10.1016/J.COGSYS.2018.07.023
21. Kavzoglu, T., & Colkesen, I. (2009). A kernel functions analysis for support vector machines for land cover classification. International Journal of Applied Earth Observation and Geoinformation, 11(5), 352–359. https://doi.org/10.1016/J.JAG.2009.06.002
22. Majumder, A. K., Jain, R., Banerjee, P., & Barnwal, J. P. (2008). Development of a new proximate analysis based correlation to predict calorific value of coal. Fuel, 87(13–14), 3077–3081. https://doi.org/10.1016/j.fuel.2008.04.008
23. Matin, S. S., & Chelgani, S. C. (2016). Estimation of coal gross calorific value based on various analyses by random forest method. Fuel, 177, 274–278. https://doi.org/10.1016/j.fuel.2016.03.031
24. Mazumdar, B. K. (2000). Theoretical oxygen requirement for coal combustion: relationship with its calorific value. Fuel, 79(11), 1413–1419. https://doi.org/10.1016/S0016-2361(99)00272-0
25. Mesroghli, S., Jorjani, E., & Chehreh Chelgani, S. (2009). Estimation of gross calorific value based on coal analysis using regression and artificial neural networks. International Journal of Coal Geology, 79(1–2), 49–54. https://doi.org/10.1016/j.coal.2009.04.002
26. Nasir, S., Kucerik, J., & Mahmood, Z. (2012). A study on the washability of the Azad Kashmir (Pakistan) coalfield. Fuel Processing Technology, 99, 75–81. https://doi.org/10.1016/j.fuproc.2012.02.003
27. Orr, M. J. L. (1996). Introduction to Radial Basis Function Networks. Centre for Cognitive Science, University of Edinburgh, Scotland.
28. Parikh, J., Channiwala, S. A., & Ghosal, G. K. (2005). A correlation for calculating HHV from proximate analysis of solid fuels. Fuel, 84(5), 487–494. https://doi.org/10.1016/j.fuel.2004.10.010
29. Patel, S. U., Jeevan Kumar, B., Badhe, Y. P., Sharma, B. K., Saha, S., Biswas, S., … Kulkarni, B. D. (2007). Estimation of gross calorific value of coals using artificial neural networks. Fuel, 86(3), 334–344. https://doi.org/10.1016/j.fuel.2006.07.036
30. Qi, M., Luo, H., Wei, P., & Fu, Z. (2019). Estimation of low calorific value of blended coals based on support vector regression and sensitivity analysis in coal-fired power plants. Fuel, 236, 1400–1407. https://doi.org/10.1016/j.fuel.2018.09.117
31. Quej, V. H., Almorox, J., Arnaldo, J. A., & Saito, L. (2017). ANFIS, SVM and ANN soft-computing techniques to estimate daily global solar radiation in a warm sub-humid environment. Journal of Atmospheric and Solar-Terrestrial Physics, 155, 62–70. https://doi.org/10.1016/J.JASTP.2017.02.002
32. Ren, S., & Gao, L. (2011). Combining artificial neural networks with data fusion to analyze overlapping spectra of nitroaniline isomers. Chemometrics and Intelligent Laboratory Systems, 107(2), 276–282. https://doi.org/10.1016/j.chemolab.2011.04.012
33. Sherrod, P. H. (2014). DTREG Predictive Modeling Software. Retrieved from www.dtreg.com
34. Sivrikaya, O. (2014). Cleaning study of a low-rank lignite with DMS, Reichert spiral and flotation. Fuel, 119, 252–258. https://doi.org/10.1016/j.fuel.2013.11.061
35. Specht, D. F. (1991). A General Regression Neural Network. IEEE Transactions on Neural Networks, 2(6), 568–576. https://doi.org/10.1109/72.97934
36. Tan, P., Zhang, C., Xia, J., Fang, Q. Y., & Chen, G. (2015). Estimation of higher heating value of coal based on proximate analysis using support vector regression. Fuel Processing Technology, 138, 298–304. https://doi.org/10.1016/j.fuproc.2015.06.013
37. Tozsin, G., Acar, C., & Sivrikaya, O. (2018). Evaluation of a Turkish Lignite Coal Cleaning by Conventional and Enhanced Gravity Separation Techniques. International Journal of Coal Preparation and Utilization, 38(3), 135–148. https://doi.org/10.1080/19392699.2016.1209191
38. Vapnik, V. (1995). The Nature of Statistical Learning Theory. New York: Springer-Verlag.
39. Vapnik, V., Golowich, S. E., & Smola, A. (1997). Support vector method for function approximation, regression estimation, and signal processing. Annual Conference on Neural Information Processing Systems (NIPS), 281–287. https://doi.org/10.1007/978-3-642-33311-8_5
40. Wen, X., Jian, S., & Wang, J. (2017). Prediction models of calorific value of coal based on wavelet neural networks. Fuel, 199, 512–522. https://doi.org/10.1016/j.fuel.2017.03.012
41. Xia, W., Xie, G., & Peng, Y. (2015, June 1). Recent advances in beneficiation for low rank coals. Powder Technology, Vol. 277, pp. 206–221. https://doi.org/10.1016/j.powtec.2015.03.003
42. Yalçin Erik, N., & Yilmaz, I. (2011). On the Use of Conventional and Soft Computing Models for Prediction of Gross Calorific Value (GCV) of Coal. International Journal of Coal Preparation and Utilization, 31(1), 32–59. https://doi.org/10.1080/19392699.2010.534683
43. Yilmaz, I., Erik, N. Y., & Kaynar, O. (2010). Different types of learning algorithms of artificial neural network (ANN) models for prediction of gross calorific value (GCV) of coals. Scientific Research and Essays, 5(16), 2242–2249.