Multi-Objective Optimum Design for Wave-Plate Demister Based on NSGA-ΙΙ Algorithm

Abstract:

For optimizing the structure design of the wave-plate demister vanes in wet flue gas desulfurization system (WFGD) of power plants, the characteristics models of removal efficiency and pressure drop were established by using least squares support vector machine (LSSVM) based on numerical simulation results. The highest relative error between the predicted output and measured value is 2%, it proves the modeling is good for the prediction. Based on the characteristics models, a multi-objective optimization model was established. It used the structural parameters as the optimal variables and the demister characteristics as the objective function. This optimization model was solved by non dominated sorting genetic algorithm (NSGA-II). The simulation data show that the Multi-objective optimum method can get more effective results compared to the weight coefficient method.

• Periodical: Advanced Materials Research (Volumes 864-867)
• Pages: 1163-1167
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.864-867.1163
• Online since: December 2013
• Authors: Zong Liang Qiao*, Lei Zhang, Feng Qi Si, Zhi Gao Xu
• Keywords: Demister, LSSVM, NSGA-ΙΙ Algorithm, Pressure Drop, Removal Efficiency

References

[1] Zhichun Sun, Yonghong Guo, Haiping Xiao, et al. Numerical simulation and experimental validation on demisting characteristic of the mist eliminator for Jet Bubble reactor dusulfurization system, Proceedings of the CSEE, 30(8)(2010).

[2] Jie Yao, ZhaoPing Zhong, Shanming Zhou. Numerical simulation on wave-plate demister with hamulus in wet flue gas desulfurization, Proceedings of the CSEE, 30(14) (2010): 61-67(in Chinese).

[3] Zongliang Qiao, Jianxin Zhou, Weiqing Zhou, et al．A method for optimum design of wave-plate demister based on the numerical computation, Journal of Southeast University(Natural Science Edition), 43(1) (2013): 76 - 82 ( in Chinese).

[4] P.W. James, B.J. Azzopardi, Y Wang, J.P. Hughes．A model for liquid film flow and separation in a wave-plate mist eliminator, Chemical Engineering Research and Design, 83(5) (2005): 469-477.

DOI: 10.1205/cherd.03363

[5] Chiara G, Elisabetta B, Leonardo T．A numerical model for gas flow and droplet motion in wave-plate mist eliminators with drainage channels, Chemical Engineering Science, 63(23) (2008): 5639-5652.

DOI: 10.1016/j.ces.2008.08.013

[6] ZhengYang Gao, Zhen Guo, JiaQi Hu, et al. Multi-objective combustion optimization and flame reconstruction for W shaped boiler based on support vector regression and numerical simulation, Proceedings of the CSEE, 31(5) (2011): 13-19(in Chinese).

[7] Suykens J A K. Least Squares Support Vector Machines, Singapore: World Scientific Pubishing Co Pte Ltd, (2002).

[8] Suykens J A K, Vandewalle J．Least Squares Support Vector Machine classifiers, Neural Processing Letters, 9(3) (1999): 293-300.

[9] K. De Brabanter, P. Karsmakers, J. DeBrabanter, et al．Confidence bands for least squares support vector machine classifiers: A regression approach, Pattern Recognition, 2012, 45: 2280-2287.

DOI: 10.1016/j.patcog.2011.11.021

[10] Srinivas N, Deb K. Multi-objective optimization using non dominated sorting in genetic algorithms. Evolutionary Computation, 1994, 2: 221-248.

DOI: 10.1162/evco.1994.2.3.221

[11] Deb K, Agrawal S, Pratap A, Meyarivan T. A fast elitist non-dominated sorting genetic algorithm for multi-objective optimization: NSGA-Ⅱ/Proceedings of the Parallel Problem Solving from NatureⅥ Conference. Berlin, 2000: 849-858.

DOI: 10.1007/3-540-45356-3_83