Our Aim was to find the stochastic optimal solution for a boiler process in order to maximize steam generation while complying with pollutant emissions regulations. For a simulation base-model, support vector regression (SVR) is employed to present a general discrete-time dynamical system of a boiler, and a stochastic optimization was performed to take inherent process uncertainties into account. To generate a stochastic optimization problem, sample average approximation (SAA) based on Monte-Carlo sampling was introduced due to the properties of a SVR. Moreover, a gradient free-based particle swarm optimization (PSO) technique was applied to find the optimal parameters of SVR and investigate the stochastic optimal solution for the boiler process. The results show that the stochastic optimal solution provides almost 2% more steam generation under uncertainties: this indicates that the stochastic optimal solution provides more realistic results compared to the deterministic approach. The proposed methodology can be applied straightforwardly to black box models, or when the use of gradient-based optimization solvers is restricted.

Stochastic Optimization Support Vector Regression Particle Swarm Optimization Chemical Recovery Boiler Pulp Mill

Rahat AAM, Wang C, Everson RM, Fieldsend JE, Appl. Energy, 29, 446 (2018)
Repo S, Verifying the pulp mill atmospheric emissions from the recovery boiler, Master’s thesis, LUT School of Energy Systems, Lappeenranta University of Technology, 88 (2018).
Ghaffari S, Romagnoli JA, Comput. Aided Chem., 14, 1019 (2003)
Blasiak W, Tao L, Vaclavinek J, Lidegran P, Energy Conv. Manag., 38(10-13), 995 (1997)
Leiviska K, Simulation in pulp and paper industry, Control Engineering Laboratory, University of Oulu, Report A No 2, ISBN 951-42-4374-9, ISSN 1238-9390, 55 (1996).
Perez Manuel Garcia, Vakkilainen Esa, Hyppanen Timo, Fuel, 185, 872 (2016)
Wallinder J, Lindgren K, Samuelsson A, Dynamic modelling of a Kraft pulp mill producing softwood and hardwood pulp in campaigns, Recycling and Sustainability (PEERS) Conference, 1-6 (2018).
Maakala V, Jarvinen M, Vuorinen V, Energy, 160, 361 (2018)
Safdarnejad SM, Tuttle JF, Powell KM, Comput. Chem. Eng., 124, 62 (2019)
Pan H, Zhong W, Wang Z, Wang G, Comput. Chem. Eng., 70, 987 (2018)
Wang CL, Liu Y, Zheng S, Jiang AP, Energy, 153, 149 (2018)
Song JG, Romero CE, Yao Z, He BS, Fuel, 172, 20 (2016)
Zhou H, Zhao JP, Zheng LG, Wang CL, Cen KF, Appl. Artif. Intel., 25, 147 (2012)
Vapnik VN, The nature of statistical learning theory, Wiley, New York (1995).
Vincent B, Duhamel C, Ren L, Tchernev N, IFAC-PapersOn-Line, 51, 1604 (2018)
Lee CJ, Prasad V, Lee JM, Ind. Eng. Chem. Res., 50(7), 3938 (2011)
Yang HJ, Hwang KS, Lee CJ, Ind. Eng. Chem. Res., 57(6), 2200 (2018)
Halemane KP, Grossmann IE, AIChE J., 29, 425 (1983)
Maranas CD, AIChE J., 43(5), 1250 (1997)
Tayal MC, Diwekar UM, AIChE J., 47(3), 609 (2001)
Shastri Y, Diwekar U, Comput. Chem. Eng., 30(5), 864 (2006)
Vakkilainen EK, Kraft recovery boilers: principles and practice, Valopaino Oy, Helsinki (2005).
Gullichsen J, Fogelholm CJ, Papermaking science and technology: book 6B, Tappi Press, Atlanta (1999).
Adams TN, Frederick WJ, Grace TM, Hupa M, Lisa K, Jones AJ, Tran H, Kraft Recovery Boilers, Tappi Press, Norcross (1997).
Nandi S, Badhe Y, Lonari J, Sridevi U, Rao BS, Tambe SS, Kulkarni BD, Chem. Eng. J., 97(2-3), 115 (2004)
Chiang M, IEEE J. Sel. Areas Commun., 23, 104 (2005)
Kennedy J, Eberhart R, Particle swarm optimization, in: Proceedings of IEEE International Conference on Neural Networks, 1942 (1995).
Chen WN, Zhang J, Lin Y, Chen N, Zhan ZH, Chung HS, Li Y, Shi YH, IEEE Trans. Evol. Comput., 17, 241 (2013)
Schwaab M, Biscaia EC, Monteiro JL, Pinto JC, Chem. Eng. Sci., 63(6), 1542 (2008)
Montgomery DC, Runger GC, Applied statistics and probability for engineers, 5th Ed., Wiley, New York (2010).