Research on soft sensing of nitrite nitrogen in wastewater treatment process based on tree integration models

MA Yapeng; LI Zhuangzhuang; XU Jingsheng; LÜ Lu

doi:10.13205/j.hjgc.202504012

Volume 43 Issue 4

Apr. 2025

Turn off MathJax

Article Contents

Article Navigation > ENVIRONMENTAL ENGINEERING > 2025 > 43(4): 121-131

MA Y P，LI Z Z，XU J S，et al.Research on soft sensing of nitrite nitrogen in wastewater treatment process based on tree integration models［J］.Environmental Engineering，2025，43（4）：121-131. doi: 10.13205/j.hjgc.202504012

Citation:

PDF( 0 KB)

Research on soft sensing of nitrite nitrogen in wastewater treatment process based on tree integration models

doi: 10.13205/j.hjgc.202504012

1. State Key Laboratory of Pollution Control and Resource Reuse, School of Environment, Nanjing University, Nanjing 210046, China;
2. Joint Institute of International Environmental Technology Innovation of Changzhou Hi-Tech District and Nanjing University, Changzhou 213125, China

Received Date: 2023-11-08
Accepted Date: 2024-03-28
Rev Recd Date: 2024-01-18
Publish Date: 2025-04-01

Abstract

Abstract

The application of machine learning in the water quality monitoring of wastewater treatment processes has become a significant focus of contemporary research. Addressing the challenges about time lag and high detection cost in acquiring key water quality indicators of traditional monitoring methods in wastewater treatment processes， this study focused on the prediction of nitrite nitrogen （NO₂^--N） concentration in the shortcut nitrification-denitrification process， proposed a novel soft sensing method for water quality indicators based on the decision tree integration model. This study selected some easily measured and important operating parameters as the input features of the water quality soft sensing method， constructed four categories of tree integration models based on the decision tree to predict the effluent NO₂^--N concentration accurately in the denitrification process. The prediction accuracy and stability of each model were compared to select the best-performing one. The importance level of these models’ input features to the NO₂^--N prediction results was analyzed to interpret the optimal prediction model. Moreover， the study utilized feature selection analysis to further validate the interpretability of these tree integration models’ results. The research results indicated that among these four decision tree integration models， the Adaptive Boosting （AdaBoost） model exhibited the highest predictive accuracy and stability for the effluent NO₂^--N concentration in the denitrification process. In the prediction performance of the AdaBoost model， the determination coefficient （R²）， mean square error （MSE） and mean absolute percentage error （MAPE） was 0.983， 0.015 and 0.126， better than other models. The results of model interpretation revealed that pH， oxidation reduction potential （ORP） and influent chemical oxygen demand （COD） concentration were the most critical parameters for these decision tree integration models. These parameters affected models’ prediction results significantly and exhibited strong correlations with the effluent NO₂^--N concentration in the denitrification process. Further feature selection analysis confirmed that these parameters play crucial roles in improving the prediction accuracy of these models. In this study， the soft sensing method based on decision tree integration models provided a valuable reference for achieving low-cost and real-time prediction of water quality indicators， expanding the availability of effective data to enhancing prediction accuracy for all kinds of prediction models in wastewater treatment processes.
- machine learning,
- tree integration model,
- soft sensing of water quality,
- shortcut nitrification-denitrification,
- model interpretation

FullText(HTML)

References(33)

References

[1]	LEWIS W，WURTSBAUGH W，PAERL H，et al. Rationale for control of anthropogenic nitrogen and phosphorus to reduce eutrophication of inland waters［J］. Environmental Science& Technology，2011，45（24）:10300-10305.
[2]	MORRIS L，COLOMBO V，HASSELL K，et al. Municipal wastewater effluent licensing:a global perspective and recommendations for best practice［J］. Science of the Total Environment，2017，580:1327-1339.
[3]	XUE T L，ZHAO D H，HAN F. Svr water quality prediction model based on GA optimization［J］. Environment Engineering，2020，38（3）:123-127. 薛同来，赵冬晖，韩菲. 基于GA优化的SVR水质预测模型研究［J］. 环境工程，2020，38（3）:123-127.
[4]	ZHONG S，ZHANG K，BAGHERI M，et al. Machine learning:new ideas and tools in environmental science and engineering［J］. Environmental Science& Technology，2021，55（19）:12741-12754.
[5]	LI K，DUAN H，LIU L，et al. An integrated first principal and deep learning approach for modeling nitrous oxide emissions from wastewater treatment plants［J］. Environmental Science& Technology，2022，56（4）:2816-2826.
[6]	WANG G，JIA Q，ZHOU M，et al. Artificial neural networks for water quality soft-sensing in wastewater treatment:a review［J］. Artificial Intelligence Review. 2022； 55（1）:565-587.
[7]	LI Y Y，LIU H L. Prediction of total phosphorus in rivers based on attention mechanism of temporal convolutional networks［J］. Environment Engineering，2023，41（5）:163-171. 黎园园，刘海隆. 基于注意力机制的时间卷积网络河流总磷预测［J］. 环境工程，2023，41（5）:163-171.
[8]	DU R，PENG YZ. Technical revolution of biological nitrogen removal from municipal wastewater:recent advances in anammox research and application［J］. Scientia Sinica Technologica，2022，52（3）:389-402. 杜睿，彭永臻. 城市污水生物脱氮技术变革:厌氧氨氧化的研究与实践新进展［J］. 中国科学:技术科学，2022，52（3）:389-402.
[9]	ZHANG L，JIANG L，ZHANG J，et al. Enhancing nitrogen removal through directly integrating anammox into mainstream wastewater treatment:Advantageous，issues and future study［J］. Bioresour Technol，2022，362:127827.
[10]	LI J，WU L N，YAN Z B，et al. Research on control strategy of nitrogen removal functional bacteria in mainstream partial nitrification-anammox process［J］. Environment Engineering，2021，39（8）:45-54，61. 李进，吴莉娜，闫志斌，等. 主流部分亚硝化-厌氧氨氧化工艺脱氮功能菌群的控制策略研究［J］. 环境工程，2021，39（8），45-54，61.
[11]	WU P，CHEN J，GARLAPATI V，et al. Novel insights into Anammox-based processes:A critical review［J］. Chemical Engineering Journal，2022，13（2）:444.
[12]	AGRAWAL S，SEUNTJENS D，COCKER P，et al. Success of mainstream partial nitritation/anammox demands integration of engineering，microbiome and modeling insights［J］. Current Opinion in Biotechnology，2018，50:214-221.
[13]	GUO L J，LI B L. Research on soft measurement of effluent total nitrogen based on GNFA-SVR［J］. Industrial Water Treatment，2022，42（10）:111-117. 郭利进，李博仑. 基于GNFA-SVR污水出水总氮软测量研究［J］. 工业水处理，2022，42（10）:111-117.
[14]	ZHANG M，YANG Y，GUO W，et al. Electrochemical sensor for sensitive nitrite and sulfite detection in milk based on acid-treated Fe₃O₄@SiO₂ nanoparticles［J］. Food Chemistry，2024，35（6）:430.
[15]	LI X M，LUO X K，LI W，et al. Design of a nitrite sensor based on colorimetry［J］. Journal of Beijing University of Chemical Technology（Natural Science Edition），2020，47（6）:79-84. 李新民，罗学科，李文，等. 基于颜色检测的亚硝酸盐传感器设计［J］. 北京化工大学学报（自然科学版），2020，47（6）:79-84.
[16]	WANG C Z，WANG X C，NIU Q，et al. Assessment on measurement uncertainty of nitrite nitrogen in water by n-（1-naphthyl）-ethylenediamine spectrometry［J］. Environment Engineering，2013，31（5）:147-150. 王存政，王兴春，牛倩，等. N-（1-萘基）-乙二胺光度法测定水中亚硝酸盐氮的测量不确定度评定［J］. 环境工程，2013，31（5）:147-150.
[17]	MENG X，ZHANG Y，QIAO J，et al. An adaptive task-oriented RBF network for key water quality parameters prediction in wastewater treatment process［J］. Neural Computing& Applications，2021，33（17）:11401-11414.
[18]	FOSCHI J，TUROLLA A，ANTONELLI M，et al. Soft sensor predictor of E. coli concentration based on conventional monitoring parameters for wastewater disinfection control［J］. Water Research，2021，14（9）:191.
[19]	ZHU J，JIANG Z，FENG L，et al. Improved neural network with least square support vector machine for wastewater treatment process［J］. Chemosphere，2022，15（1）:308.
[20]	GORGAN F，RAJAEE T，ZOUNEMAT M，et al. Investigating machine learning models in predicting lake water quality parameters as a 3-year moving average［J］. Environmental Science and Pollution Research，2023，30（23）:63839-63863.
[21]	QUINLAN J. Improved use of continuous attributes in C4.5［J］. Journal of Artificial Intelligence Research，1996，4:77-90.
[22]	RUDIN C. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature Machine Intelligence［J］. 2019，1（5）:206-215.
[23]	WANG D，THUNELL S，LINDBERG U，et al. A machine learning framework to improve effluent quality control in wastewater treatment plants［J］. Science of the Total Environment，2021，3（9）:784.
[24]	HE X，JI J，LIU K，et al. Soft sensing of silicon content via bagging local semi-supervised models［J］. Sensors，2019，19（17）:36-45.
[25]	WANG R，YU Y，CHEN Y，et al. Model construction and application for effluent prediction in wastewater treatment plant:Data processing method optimization and process parameters integration［J］. Journal of Environmental Management，2022，9（14）:302-308.
[26]	LI X，YI X，LIU Z，et al. Application of novel hybrid deep leaning model for cleaner production in a paper industrial wastewater treatment system［J］. Journal of Cleaner Production，2021，14（6）:294-301.
[27]	LIU X，SHANG J，SHI H，et al. Research on Real-Time Data Processing Method of GNSS-RTK Deformation Monitoring［J］. Chinese Journal of Sensors and Actuators，2020，33（8）:1190-1196.
[28]	DOU J，YUNUS A，BUI D，et al. Improved landslide assessment using support vector machine with bagging，boosting，and stacking ensemble machine learning framework in a mountainous watershed，Japan［J］. Landslides，2020，17（3）:641-658.
[29]	BREIMAN L. Pasting small votes for classification in large databases and on-line［J］. Machine Learning，1999，36（1/2）:85-103.
[30]	CHENG W，YUAN D，XIONG P，et al. Construction and evaluation of city water quality index prediction model based on multiple machine learning algorithms［J］. Acta Scientiae Circumstantiae，2023，43（11）:144-152.
[31]	BIAN L，QIN X，ZHANG C，et al. Application，interpretability and prediction of machine learning method combined with LSTM and LightGBM-a case study for runoff simulation in an arid area［J］. Journal of Hydrology，2023，4（13）:625.
[32]	PARK J，LEE W，KIM K，et al. Interpretation of ensemble learning to predict water quality using explainable artificial intelligence［J］. Science of the Total Environment，2022，19（7）:832-839.
[33]	LI Y，WANG Y，DONG F，et al. Controlling carbon dioxide-to-hydrogen ratio to improve hydrogen utilization and denitrification rates of hydrogenotrophic autotrophic denitrification through homoacetogenesis-heterotrophic denitrification pathway［J］. Bioresource Technology，2024，16（7）:393-400.