Similar articles

Abstract: Optimizing the operation of coal-fired power plants to reduce nitrogen oxide (NO_x) emissions requires accurate modeling of the NO_x emission process. The careful selection of input parameters not only forms the basis of accurate modeling, but can also be used to reduce the complexity of the model. The present study employs the least squares support vector machine-supervised learning method to model NO_x emissions based on historical real time data obtained from a 1000-MW once-through boiler. The initial input parameters are determined by expert knowledge and operational experience, while the final input parameters are obtained by sensitivity analysis, where the variation in model accuracy for a given set of data is analyzed as one or several input parameters are successively omitted from the calculations, while retaining all other parameters. Here, model accuracy is evaluated according to the mean relative error (MRE). This process reduces the parameters required for NO_x emission modeling from an initial number of 33 to 7, while the corresponding MRE is reduced from 3.09% to 2.23%. Moreover, a correlation of 0.9566 between predicted and measured values was obtained by applying the model with just these seven input parameters to a validation dataset. As such, the proposed method for selecting input parameters serves as a reference for related studies.

This paperdescribes the selection procedure of input parameters in the prediction model for NOX emission in coal fired boiler plants. There are more than 33 variables that can be considered as input parameters in the NOx estimation model. Selection of appropriate input parameters is very complicated problem and has been received attention of many researchers. The authors propose a method of selection of most probable input parameters that can be used in the estimation successfully. I couldn't find any originality nor novelty in this paper, but the systematical selection procedure described in this paper deserves to attract attention of related engineers and researchers and may be very useful in the NOx estimation study.

基于七个运行参数建立煤粉锅炉NO_x排放模型

[1]Ahmadi MA, 2015. Connectionist approach estimates gas–oil relative permeability in petroleum reservoirs: application to reservoir simulation. Fuel, 140:429-439.

[2]Ahmadi MA, Ebadi M, 2014. Evolving smart approach for determination dew point pressure through condensate gas reservoirs. Fuel, 117:1074-1084.

[3]Ahmadi MA, Bahadori A, 2015a. A LSSVM approach for determining well placement and conning phenomena in horizontal wells. Fuel, 153:276-283.

[4]Ahmadi MA, Bahadori A, 2015b. Prediction performance of natural gas dehydration units for water removal efficiency using a least square support vector machine. International Journal of Ambient Energy, 37(5):486-494.

[5]Ahmadi MA, Ebadi M, Hosseini SY, 2014a. Prediction breakthrough time of water coning in the fractured reservoirs by implementing low parameter support vector machine approach. Fuel, 117:579-589.

[6]Ahmadi MA, Ebadi M, Marghmaleki PS, et al., 2014b. Evolving predictive model to determine conden sate-to-gas ratio in retrograded condensate gas reservoirs. Fuel, 124:241-257.

[7]Ahmadi MA, Ebadi M, Yazdanpanah A, 2014c. Robust intelligent tool for estimation dew point pressure in retrograded condensate gas reservoirs: application of particle swarm optimization. Journal of Petroleum Science and Engineering, 123:7-19.

[8]Ahmadi MA, Masoumi M, Askarinezhad R, 2014d. Evolving connectionist model to monitor the efficiency of an in situ combustion process: application to heavy oil recovery. Energy Technology, 2(9-10):811-818.

[9]Ahmadi MA, Zahedzadeh M, Shadizadeh SR, et al., 2015a. Connectionist model for predicting minimum gas miscibility pressure: application to gas injection process. Fuel, 148:202-211.

[10]Ahmadi MA, Masoumi M, Askarinezhad R, 2015b. Evolving smart model to predict the combustion front velocity for in situ combustion. Energy Technology, 3(2):128-135.

[11]Ahmadi MA, Hasanvand MZ, Bahadori A, 2015c. A least-squares support vector machine approach to predict temperature drop accompanying a given pressure drop for the natural gas production and processing systems. International Journal of Ambient Energy, 38(2):122-129.

[12]Ahmadi MA, Lee M, Bahadori M, 2015d. Prediction of a solid desiccant dehydrator performance using least squares support vector machines algorithm. Journal of the Taiwan Institute of Chemical Engineers, 50:115-122.

[13]Ahmadi MH, Ahmadi MA, Sadatsakkak SA, et al., 2015. Connectionist intelligent model estimates output power and torque of stirling engine. Renewable and Sustainable Energy Reviews, 50:871-883.

[14]Ahmed F, Cho HJ, Kim JK, et al., 2015. A real-time model based on least squares support vector machines and output bias update for the prediction of NO_x emission from coal-fired power plant. Korean Journal of Chemical Engineering, 32(6):1029-1036.

[15]Cen KF, Yao Q, Luo ZY, et al., 2003. Advanced Combustion. Zhejiang University Press, Hangzhou, China (in Chinese).

[16]Choi CR, Kim CN, 2009. Numerical investigation on the flow, combustion and NO_x emission characteristics in a 500 MW_e tangentially fired pulverized-coal boiler. Fuel, 88(9):1720-1731.

[17]Choi GP, Yoo KHY, Back JH, et al., 2017. Estimation of LOCA break size using cascaded fuzzy neural networks. Nuclear Engineering and Technology, 49(3):495-503.

[18]Eberhart RC, Shi Y, Kennedy J, 2001. Swarm Intelligence. Morgan Kaufmann, San Francisco, USA.

[19]Fan WD, Lin ZC, Li YY, et al., 2010. Experimental flow field characteristics of OFA for large-angle counter flow of fuel-rich jet combustion technology. Applied Energy, 87(8):2737-2745.

[20]Fletcher R, 1987. Practical Methods of Optimization. John Wiley & Son, Chichester and New York.

[21]Gu YP, Zhao WJ, Wu ZS, 2011. Online adaptive least squares support vector machine and its application in utility boiler combustion optimization systems. Journal of Process Control, 21(7):1040-1048.

[22]Hattori Y, Otsuka M, 2013. Modeling of feed-forward control using the partial least squares regression method in the tablet compression process. International Journal of Pharmaceutics, 524(1-2):407-413.

[23]Hill SC, Smoot LD, 2000. Modeling of nitrogen oxides formation and destruction in combustion systems. Progress in Energy and Combustion Science, 26(4-6):417-458.

[24]IEA (International Energy Agency), 2015. Key World Energy Statistic. IEA.

[25]Ilamathi P, Selladurai V, Balamurugan K, 2013. Modeling and optimization of unburned carbon in coal-fired boiler using artificial neural network and genetic algorithm. Journal of Energy Resources Technology, 135(3):032201.

[26]Junhom C, Weerapreeyakul N, Tanthanuch W, et al., 2017. Partial least squares regression and Fourier transform infrared (FTIR) microspectroscopy for prediction of resistance in hepatocellular carcinoma HepG2 cells. Experimental Cell Research, 351(1):82-90.

[27]Kennedy J, Eberhart R, 1995. Particle swarm optimization. Proceedings of ICNN’95–International Conference on Neural Networks, p.1942-1948.

[28]Khajehsharifi H, Eskandari Z, Sareban N, 2017. Using partial least squares and principal component regression in simultaneous spectrophotometric analysis of pyrimidine bases. Arabian Journal of Chemistry, 10(S1):S141-S147.

[29]Khajehzadeh M, Taha MR, El-shafie A, et al., 2011. Modified particle swarm optimization for optimum design of spread footing and retaining wall. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering), 12(6):415-427.

[30]Kuang M, Li ZQ, Wang ZH, et al., 2014. Combustion and NO_x emission characteristics with respect to staged-air damper opening in a 600 MW_e down-fired pulverized coal furnace under deep-air-staging conditions. Environmental Science & Technology, 48(1):837-844.

[31]Laurain V, Tóth R, Piga D, et al., 2015. An instrumental least squares support vector machine for nonlinear system identification. Automatica, 54:340-347.

[32]Li ZQ, Liu ZC, Chen ZC, et al., 2013. Effect of angle of arch-supplied overfire air on flow, combustion characteristics and NO_x emissions of a down-fired utility boiler. Energy, 59:377-386.

[33]Liu FD, He H, Zhang CB, et al., 2011. Mechanism of the selective catalytic reduction of NO_x with NH₃ over environmental-friendly iron titanate catalyst. Catalysis Today, 175(1):18-25.

[34]Lv Y, Liu JZ, Yang TT, et al., 2013. A novel least squares support vector machine ensemble model for NO_x emission prediction of a coal-fired boiler. Energy, 55:319-329.

[35]Lv Y, Yang TT, Liu JZ, 2015. An adaptive least squares support vector machine model with a novel update for NO_x emission prediction. Chemometrics and Intelligent Laboratory Systems, 145:103-113.

[36]Malegori C, Malegori EJN, de Freitas ST, et al., 2017. Comparing the analytical performances of micro-NIR and FT-NIR spectrometers in the evaluation of acerola fruit quality, using PLS and SVM regression algorithms. Talanta, 165:112-116.

[37]Modlinski N, 2015. Numerical simulation of SNCR (selective non-catalytic reduction) process in coal fired grate boiler. Energy, 92:67-76.

[38]Rezaei H, Nazir R, Momeni E, 2016. Bearing capacity of thin-walled shallow foundations: an experimental and artificial intelligence-based study. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering), 17(4):273-285.

[39]Samui P, Kim D, Aiyer BG, 2015. Pullout capacity of small ground anchor: a least square support vector machine approach. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering), 16(4):295-301.

[40]Smrekar J, Potočnik P, Senegačnik A, 2013. Multi-step-ahead prediction of NO_x emissions for a coal-based boiler. Applied Energy, 106:89-99.

[41]Song JG, Romero CE, Yao Z, et al., 2017. A globally enhanced general regression neural network for on-line multiple emissions prediction of utility boiler. Knowledge-Based Systems, 118:4-14.

[42]Suykens JAK, Vandewalle J, 1999. Least squares support vector machine classifiers. Neural Processing Letters, 9(3):293-300.

[43]Suykens JAK, Vandewalle J, de Moor B, 2001. Optimal control by least squares support vector machine. Neural Networks, 14(1):23-25.

[44]Suykens JAK, Gestel TV, Brabanter JD, et al., 2002. Least squares support vector machines. Euphytica, 2(2):1599-1604.

[45]Tang Y, Ma X, Lai Z, et al., 2012. NO_x and SO₂ emissions from municipal solid waste (MSW) combustion in CO₂/O₂ atmosphere. Energy, 40(1):300-306.

[46]Thissen U, Pepers M, Üstün B, et al., 2004. Comparing support vector machines to PLS for spectral regression applications. Chemometrics and Intelligent Laboratory Systems, 73(2):169-179.

[47]Wang QC, Zhang JZ, 2011. Wiener model identification and nonlinear model predictive control of a pH neutralization process based on Laguerre filters and least squares support vector machines. Journal of Zhejiang University-SCIENCE C (Computers & Electronics), 12(1):25-35.

[48]Wang ZH, Zhou JH, Zhang YW, et al., 2005. Experiment and mechanism investigation on advanced reburning for NO_x reduction: influence of CO and temperature. Journal of Zhejiang University-SCIENCE, 6B(3):187-194.

[49]Wei ZB, Li XL, Xu LJ, et al., 2013. Comparative study of computational intelligence approaches for NO_x reduction of coal-fired boiler. Energy, 55:683-692.

[50]Wei ZS, Du ZY, Lin ZH, et al., 2007. Removal of NO_x by microwave reactor with ammonium bicarbonate and Ga-A zeolites at low temperature. Energy, 32(8):1455-1459.

[51]Wu F, Zhou H, Ren T, et al., 2009. Combining support vector regression and cellular genetic algorithm for multi-objective optimization of coal-fired utility boilers. Fuel, 88(10):1864-1870.

[52]Xiang J, Wang PY, Su S, et al., 2015. Control of NO and Hg⁰ emissions by SCR catalysts from coal-fired boiler. Fuel Processing Technology, 135:168-173.

[53]Xu YY, Zhang Y, Liu FN, et al., 2014. CFD analysis on the catalyst layer breakage failure of an SCR-DeNO_x system for a 350 MW coal-fired power plant. Computers & Chemical Engineering, 69:119-127.

[54]Yin F, Luo ZH, Li Y, et al., 2017. Coal type identification based on the emission spectra of a furnace flame. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering), 18(2):113-123.

[55]Zhang JH, Liu Y, 2017. Application of complete ensemble intrinsic time-scale decomposition and least-square SVM optimized using hybrid DE and PSO to fault diagnosis of diesel engines. Journal of Zhejiang University-SCIENCE B (Biomedicine & Biotechnology), 18(3):272-286.

[56]Zhang Y, Li H, Wang ZH, et al., 2015. A preliminary study on time series forecast of fair-weather atmospheric electric field with WT-LSSVM method. Journal of Electrostatics, 75:85-89.

[57]Zhou H, Cen KF, Fan JR, 2004. Modeling and optimization of the NO_x emission characteristics of a tangentially fired boiler with artificial neural networks. Energy, 29(1):167-183.

[58]Zhou H, Zhao JP, Zheng LG, et al., 2012. Modeling NO_x emissions from coal-fired utility boilers using support vector regression with ant colony optimization. Engineering Applications of Artificial Intelligence, 25(1):147-158.

[59]Zhou H, Yang Y, Dong K, et al., 2014. Influence of the gas particle flow characteristics of a low-NO_x swirl burner on the formation of high temperature corrosion. Fuel, 134: 595-602.

[60]Zhou H, Yang Y, Wang YY, 2015. Numerical investigation of gas-particle flow in the primary air pipe of low NO_x swirl burner–the DEM-CFD method. Particuology, 19:133-140.

[61]Zhou H, Li Y, Tang Q, et al., 2017. Combining flame monitoring techniques and support vector machine for the online identification of coal blends. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering), 18(9):677-689.