Journal of the Korean Institute of Gas (한국가스학회지)

The Korean Institute of GAS (한국가스학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Partial Least Squares and Support Vector Machine for the Autoignition Temperature Prediction of Organic Compounds

유기물의 자연발화점 예측을 위한 부분최소자승법과 SVM의 비교

  • Lee, Gi-Baek (Department of Chemical and Biological Engineering, Chungju National University)
  • 이기백 (충주대학교 화공생물공학과)
  • Received : 2011.12.07
  • Accepted : 2012.02.26
  • Published : 2012.02.29
Download PDF
⟨ Previous Next ⟩

Abstract

The autoignition temperature is one of the most important physical properties used to determine the flammability characteristics of chemical substances. Despite the needs of the experimental autoignition temperature data for the design of chemical plants, it is not easy to get the data. This study have built and compared partial least squares (PLS) and support vector machine (SVM) models to predict the autoignition temperatures of 503 organic compounds out of DIPPR 801. As the independent variables of the models, 59 functional groups were chosen based on the group contribution method. The prediction errors calculated from cross-validation were employed to determine the optimal parameters of two models. And, particle swarm optimization was used to get three parameters of SVM model. The PLS and SVM results of the average absolute errors for the whole data range from 58.59K and 29.11K, respectively, indicating that the predictive ability of the SVM is much superior than PLS.

화학물질의 화재위험을 나타내는 가장 중요한 물성의 하나인 자연발화점의 실험 데이터는 그 필요에도 불구하고 데이터를 얻는 것이 어려운 경우가 많다. 이 연구에서는 DIPPR 801에서 얻은 503개 유기물의 자연발화점 실험데이터로부터 자연발화점을 예측하는 부분최소자승법(PLS) 및 support vector machine(SVM) 모델을 만들고 비교하였다. 그룹기여법을 이용하여 59개 작용기가 이 예측모델의 독립변수가 되었다. 두 모델에서 결정해야 할 매개변수는 교차검증으로 계산된 오차를 이용하여 결정되었고, SVM모델은 그 매개변수가 많아 particle swarm optimization을 이용한 최적화를 이용하였다. 전체 데이터에 대해 계산된 평균절대오차는 PLS가 58.59K였고, SVM이 29.11K여서 SVM이 PLS에 비해 매우 우수한 예측성능을 보였다.

Keywords

References

  1. Suzuki, T., "Quantitative Structure- Property Relationships for Auto-ignition Temperatures of Organic Compounds", Fire and Materials, 18, 81-88, (1994) https://doi.org/10.1002/fam.810180204
  2. Tetteh, J., E. Metcalfe and S.L. Howells, "Optimisation of Radial Basis and Backpropagation neural networks for modelling auto-ignition temperature by quantitative-structure property relation- ships", Chemom. Intell. Lab. Syst., 32, 177-191, (1996) https://doi.org/10.1016/0169-7439(95)00088-7
  3. Albahri, T.A. and R.S. George, "Artificial Neural Network Investigation of the Structural Group Contribution Method for Predicting Pure Components Auto Ignition Temperature", Ind. Eng. Chem. Res., 42, 5708-5714, (2003). https://doi.org/10.1021/ie0300373
  4. Pan, Y., J. Jiang, R. Wang and H. Cao, "Advantages of Support Vector Machine in QSPR Studies for Predicting Auto- ignition Temperatures of Organic Compounds", Chemom. Intell. Lab. Syst., 92, 169-178, (2008) https://doi.org/10.1016/j.chemolab.2008.03.002
  5. Pan, Y., J. Jiang, R. Wang, H. Cao and H. Cao, "Predicting the Auto-ignition temperatures of Organic Compounds from Molecular Structure Using Support Vector Machine", J. Hazard. Mater., 164, 1242-1249, (2009) https://doi.org/10.1016/j.jhazmat.2008.09.031
  6. Constantinou, L. and R. Gani, "New Group Contribution Method for Estimating Properties of Pure Comp- ounds," AIChE Jr., 40, 1697-1710, (1994) https://doi.org/10.1002/aic.690401011
  7. Lee, C.J., G. Lee, W. So and E.S. Yoon, "A New Estimation Algorithm of Physical Properties based on a Group Contribution and Support Vector Machine", Korean J. Chem. Eng., 25, 568-574 (2008). https://doi.org/10.1007/s11814-008-0096-0
  8. http://www.aiche.org/dippr/
  9. 이창준, 고재욱, 이기백, "유기물의 인화점 예측을 위한 부분최소자승법과 SVM의 비교", 화학공학, 48, 717-724, (2010)
  10. 이희두, 이무호, 조현우, 한종훈, 장근수, "다변량 통계 분석 방법을 이용한 연속 교반 MMA-VA 공중합 공정 품질 변수 온라인 모니터링", 화학공학, 35, 605-612, (1997)
  11. Vapnik, V.N., The Nature of Statistical Learning Theory, Springer-Verlag, New York, NY(1995)
  12. http://www.csie.ntu.edu.tw/-cjlin/libsvm/
  13. Schwwab, M., E.C. Biscaia, J.L. Monteiro and J.C. Pinto, "Nonlinear Parameter Estimation through Particle Swarm Optimization," Chem. Eng. Sci., 63, 1542-1552, (2008) https://doi.org/10.1016/j.ces.2007.11.024

Cited by

  1. The Development of a SVR-based Empirical Model for the Effect of the Unbalanced Floor Height on MVC of Lifting Task vol.29, pp.4, 2014, https://doi.org/10.14346/JKOSOS.2014.29.4.153
  2. A Study on the Spontaneous Ignition Possibility of Shredded Waste Thermoplastic Elastomer vol.31, pp.1, 2016, https://doi.org/10.14346/JKOSOS.2016.31.1.061