DOI QR코드

DOI QR Code

Nonlinear Chemical Plant Modeling using Support Vector Machines: pH Neutralization Process is Targeted

SVM을 이용한 비선형 화학공정 모델링: pH 중화공정에의 적용 예

  • 김동원 (고려대학교 공과대학 전기공학부) ;
  • 유아림 (고려대학교 공과대학 화학생명공학과) ;
  • 양대륙 (고려대학교 공과대학 화학생명공학과) ;
  • 박귀태 (고려대학교 공과대학 전기공학부)
  • Published : 2006.12.01

Abstract

This paper is concerned with the modeling and identification of pH neutralization process as nonlinear chemical system. The pH control has been applied to various chemical processes such as wastewater treatment, chemical, and biochemical industries. But the control of the pH is very difficult due to its highly nonlinear nature which is the titration curve with the steepest slope at the neutralization point. We apply SVM which have become an increasingly popular tool for machine teaming tasks such as classification, regression or detection to model pH process which has strong nonlinearities. Linear and radial basis function kernels are employed and each result has been compared. So SVH based on kernel method have been found to work well. Simulations have shown that the SVM based on the kernel substitution including linear and radial basis function kernel provides a promising alternative to model strong nonlinearities of the pH neutralization but also to control the system.

Keywords

References

  1. J. Nie. A. Loh, and C. Hang, 'Modeling pH neutralization processes using fuzzy-neural approaches,' Fuzzy sets and systems, vol. 78, pp. 5-22, 1996 https://doi.org/10.1016/0165-0114(95)00118-2
  2. A. Yoo, Experimental Parameter Identification and Control of pH Neutralization Process Based on an Extended Kalman Filter, Master's thesis, Department of Chemical Engineering, Korea University, 2002
  3. 김동규, 이광순, 양대륙, '최신 제어에의 기술 동향 고찰', Simulation-based dynamic programming' 제어.자동화.시스템공학회지, vol. 10, no. 1, pp. 59-66, 2004
  4. 김동원, 박귀태, '다항식 뉴럴 네트워크의 최적화: 진화론적 방법', 대한전기학회 논문지, vol.52, no. 7, pp. 424-433, 2003
  5. T. J. McAvoy, E. Hsu, and S. Lowenthal, 'Dynamics of pH in controlled stirred tank reactor,' Ind. Engrg. Chem. Process Des. Develop. 11, pp. 68-70,1972 https://doi.org/10.1021/i260041a013
  6. V. Vapnik, The Nature of Statistical Learning Theory, John Wiley, New York, 1995
  7. S. Gunn, 'Support vector machines for classification and regression,' ISIS technical report, Image Speech & Intelligent Systems Group University of Southampton, 1998
  8. K. Kim, 'Financial time series forecasting using support vector machines,' Neurocomputing, vol. 55, pp. 307-319, 2003 https://doi.org/10.1016/S0925-2312(03)00372-2
  9. P. Wolfe, A duality theorem for non-linear programming, Quart. Appl. Math. 19, 1961
  10. W. Wang and Z. Xu, 'A heuristic training for support vector regression,' Neurocomputing, vol. 61, pp. 259-275, 2004 https://doi.org/10.1016/j.neucom.2003.11.012