Emission Concentration Prediction of NO x from Waste Incinerator Based on MIC-PCA-LSTM Model

doi:10.12141/j.issn.1000-565X.240519

Abstract

Abstract:

Accurately predicting the emission concentration of NO _x at the outlet of the selective catalytic reduction (SCR) denitrification system in the waste incineration process is of great significance for enhancing data quality and optimizing ammonia injection. However, the waste incineration process exhibits significant nonlinearity, multivariate coupling, and time-series characteristics. These factors pose substantial challenges to achieving accurate prediction of NO _x emissions. To solve this problem, this paper presents a prediction model for the emission concentration of NO _x at the outlet of SCR denitrification system by integrating maximum information coefficient (MIC), principal component analysis (PCA) and long short-term memory (LSTM) neural networks. First, MIC is employed to assess the maximum normalized mutual information values among variables, and the input variables that exhibit the strong-est correlation with NO _x emission concentration are selected while the redundant variables are eliminated based on the principle of maximum redundancy. Then, PCA is utilized to obtain the cumulative contribution rate of the va-riance of each principal component, extract the principal component features, and obtain the optimal input feature variable set. Finally, an emission prediction model of NO _x at the outlet of SCR denitrification system is established based on the LSTM neural network. The results indicate that, as compared with the back propagation neural network model and the support vector machine model, the proposed model exhibits higher accuracy and generalization ability, achieving a mean absolute percentage error of 6.33%, a root mean squared error of 4.71 mg/m³ and a determination coefficient of 0.90. This research lays a theoretical foundation for achieving the intelligent control of SCR denitrification system in the waste incineration process.

Key words: waste incineration, selective catalytic reduction, emission concentration prediction, maximum information coefficient, principal component analysis, long short-term memory neural network

CLC Number:

TM621

YAO Shunchun, LI Longqian, LIU Wen, LI Zhenghui, ZHOU Anli, LI Wenjing, CHEN Jianghong, LU Zhimin. Emission Concentration Prediction of NO _x from Waste Incinerator Based on MIC-PCA-LSTM Model[J]. Journal of South China University of Technology(Natural Science Edition), 2025, 53(7): 1-10.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Table 1

Fig.4

Fig.5

Table 2

Fig.6

Table 3

Fig.7

Table 4

Fig.8

Table 5

References 19

[1]	DIRIK M ．Prediction of NO _x emissions from gas turbines of a combined cycle power plant using an ANFIS model optimized by GA［J］．Fuel，2022，321：124037/1-11．
[2]	卢志民，李博航，唐雯，等．燃煤电厂SCR脱硝系统喷氨优化模拟［J］．华南理工大学学报（自然科学版），2023，51（8）：62-70．
	LU Zhimin， LI Bohang， TANG Wen，et al ．Optimization simulation of ammonia injection in SCR denitrification system of coal-fired power plant［J］．Journal of South China University of Technology （Natural Science Edition），2023，51（8）：62-70．
[3]	ADAMS D，OH D， KIM D，et al ．Prediction of SO _x -NO _x emission from a coal-fired CFB power plant with machine learning：plant data learned by deep neural network and least square support vector machine［J］．Journal of Cleaner Production，2020，270：122310/1-16．
[4]	LI S， MA S， WANG F ．A combined NO _x emission prediction model based on semi-empirical model and black box models［J］．Energy，2023，264：126130/1-9．
[5]	ZHANG M ．Characteristic-particle-tracked modeling for CFB boiler：coal combustion and ultra-low NO emission［J］．Powder Technology，2020，374：632-647．
[6]	WANG X， LIU W， WANG Y，et al ．A hybrid NO _x emission prediction model based on CEEMDAN and AM-LSTM［J］．Fuel，2022，310：122486/1-12．
[7]	LV Y， YANG T， LIU J ．An adaptive least squares support vector machine model with a novel update for NO _x emission prediction［J］．Chemometrics and Intelligent Laboratory Systems，2015，145：103-113．
[8]	WANG G， AWAD O I， LIU S，et al ．NO _x emissions prediction based on mutual information and back propagation neural network using correlation quantitative analysis［J］．Energy，2020，198：117286/1-10．
[9]	RADOJEVIĆ D， ANTANASIJEVIĆ D， PERIĆ-GRUJIĆ A，et al ．The significance of periodic parameters for ANN modeling of daily SO₂ and NO _x concentrations：a case study of Belgrade，Serbia［J］．Atmospheric Pollution Research，2019，10（2）：621-628．
[10]	WANG J， FENG Y， YE S，et al ．NO _x emission prediction of coal-fired power units under uncertain classification of operating conditions［J］．Fuel，2023，343：127840/1-11．
[11]	ARSIE I， CRICCHIO A， DE CESARE M，et al ．Neural network models for virtual sensing of NO _x emissions in automotive diesel engines with least square-based adaptation［J］．Control Engineering Practice，2017，61：11-20．
[12]	HOCHREITER S， SCHMIDHUBER J ．Long short-term memory［J］．Neural Computation，1997，9（8）：1735-1780．
[13]	LI Z， YAO S， CHEN D，et al ．Multi-parameter co-optimization for NO _x emissions control from waste inci-nerators based on data-driven model and improved particle swarm optimization［J］．Energy，2024，306：132477/1-13．
[14]	TANG J， GUO Y， LI M，et al ．A hybrid approach for the dynamic monitoring and forecasting of NO _x emissions in power plants［J］．Sustainable Energy，Grids and Networks，2023，36：101208/1-11．
[15]	XIAN W， WANG Z， SHI L，et al ．Rapid identification of cocrystal components of explosives based on Raman spectroscopy and principal component analysis［J］．Vibrational Spectroscopy，2024，132：103689/1-6．
[16]	PENG C， ZHENG J， CHEN T，et al ．Tool wear feature extraction in BTA deep hole drilling process based on maximum probability multi-synchrosqueezing transform of spindle current signal［J］．Measurement，2025，241：115780/1-14．
[17]	喻成龙，黄碧纯，杨颖欣．分子筛应用于低温NH₃-SCR脱硝催化剂的研究进展［J］．华南理工大学学报（自然科学版），2015，43（3）：143-150．
	YU Cheng-long， HUANG Bi-chun， YANG Ying-xin ．Research progress on the application of molecular sieves in low-temperature NH₃-SCR denitrification catalysts［J］．Journal of South China University of Technology （Natural Science Edition），2015，43（3）：143-150．
[18]	CONTESSI D， VIVERIT L， PEREIRA L N，et al ．Decoding the future： proposing an interpretable machine learning model for hotel occupancy forecasting using principal component analysis［J］．International Journal of Hospitality Management，2024，121：103802/1-17．
[19]	KUMAR VYAS J， PERUMAL M， MORAMARCO T ．Non-contact discharge estimation at a river site by using only the maximum surface flow velocity［J］．Journal of Hydrology，2024，638：131505/1-12．

序号	特征变量名称	单位	取值范围
1	空气预热器出口温度	℃	194.14~234.02
2	SCR还原剂流量	L/h	1.16~65.51
3	SCR反应器进口温度	℃	181.17~209.07
4	SCR反应器出口温度	℃	185.73~193.89
5	集汽箱压力	MPa	6.12~6.24
6	CO质量浓度	mg/m³	0.27~47.42
7	炉排3左外速度	cm/min	20.61~112.50
8	炉排4右外速度	cm/min	3.24~82.65
9	二次风流量A	N·m³/h	502.79~11 670.76
10	二次风流量B	N·m³/h	180.16~12 319.38
11	燃尽风流量	N·m³/h	1 341.32~8 901.18
12	炉排二区左侧热一次风量	N·m³/h	3 725.85~8 957.53
13	炉排四区左侧热一次风量	N·m³/h	1 230.76~4 830.93

参数名称	取值
隐藏层数	3
隐藏神经元数	［55，15，30］
遗忘率	［0.5，0.3，0.2］
初始学习率	0.001
学习率下降系数	0.2

模型	MAPE/%	RMSE/（mg·m^-3）	决定系数	耗时/s
MIC-PCA-BPNN	13.27	8.23	0.75	4.23
MIC-PCA-SVM	7.63	5.57	0.86	0.79
MIC-PCA-LSTM	6.33	4.71	0.90	15.91

模型	MAPE/%	RMSE/（mg·m^-3）	决定系数	耗时/s
BPNN	15.21	9.17	0.70	4.80
MIC-BPNN	13.46	8.91	0.74	4.42
SVM	9.03	6.74	0.79	0.89
MIC-SVM	8.05	6.04	0.83	0.87
LSTM	7.49	5.48	0.86	13.72
MIC-LSTM	6.63	4.79	0.89	13.66

模型	MAPE/%	RMSE/（mg·m^-3）	决定系数	耗时/s
MIC-BPNN	13.46	8.91	0.74	4.42
MIC-PCA-BPNN	13.27	8.23	0.75	4.23
MIC-SVM	8.05	6.04	0.83	0.87
MIC-PCA-SVM	7.63	5.57	0.86	0.79
MIC-LSTM	6.63	4.79	0.89	13.66
MIC-PCA-LSTM	6.33	4.71	0.90	15.91