基于自注意力机制的污水处理厂精确加药模型预测控制

裴力锋; 陈伟杰; 徐敬生; 吕路

doi:10.13205/j.hjgc.202311015

基于自注意力机制的污水处理厂精确加药模型预测控制

doi: 10.13205/j.hjgc.202311015

裴力锋¹,
陈伟杰¹,
徐敬生²,
吕路^1,2, ,

1. 南京大学环境学院污染控制与资源化研究国家重点实验室, 南京 210046;
2. 南京大学常高新国际环保产业技术研究院, 江苏常州 213125

基金项目:

国家水体污染控制与治理科技重大专项（2017ZX07204）

详细信息

作者简介:
裴力锋(1997-),男,硕士研究生,主要研究方向为污水处理厂智慧化。lifengpei@smail.nju.edu.cn

通讯作者:
吕路(1975-),男,教授,主要研究方向为大数据与工业污染减排过程控制。esellu@nju.edu.cn

计量
- 文章访问数: 183
- HTML全文浏览量: 15
- PDF下载量: 10
- 被引次数: 0
出版历程
- 收稿日期: 2022-12-08
- 网络出版日期: 2023-12-25

MODEL PREDICTIVE CONTROL FOR ACCURATE DOSING IN WASTEWATER TREATMENT PLANTS BASED ON SELF-ATTENTION MECHANISM

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

摘要

摘要: 污水处理过程精确控制是解决污水处理厂运行成本高和出水水质不达标问题的有效途径，但由于污水处理过程具有非线性、时变性和时滞性等特点，常规的控制方法难以满足需求。提出了一种模型预测控制方法，以实现污水处理厂药剂的精确投加。所构建的模型基于自注意力机制提取输入序列的有效信息，对参数之间复杂的非线性关系进行建模，从而提高预测准确度。使用江苏省某污水处理厂的反硝化滤池数据进行模型训练和测试，并建立与卷积神经网络（CNN）、长短期记忆网络（LSTM）的对比实验。结果表明，对于出水TN、COD和NH₄⁺-N的预测，自注意力模型的平均误差率（MER）、均方误差（MSE）指标均为最好，实现了较为准确的预测。采用粒子群优化算法计算加药量，并将预测控制模型应用于该污水处理厂。2022年1，2月的模型应用结果表明，出水TN的达标率达到95%以上，优化后的加药量比污水处理厂的月平均加药量分别减少28.72%和21.78%。
- 深度学习 /
- 模型预测控制 /
- 可解释机器学习 /
- 自注意力机制 /
- 粒子群优化 /
- 精确加药
Abstract: Precise control is an effective way to solve the problems of high operating costs and substandard effluent quality of wastewater treatment plants, but due to the non-linear, time-varying, and time-lagging nature of the sewage treatment process, conventional control methods can hardly meet the demand. This study proposed a model predictive control method to achieve accurate dosing of chemicals in sewage treatment plants. The constructed model was based on the self-attention mechanism to extract valid information from the input sequence and to model the complex non-linear relationships between features, thus improving the prediction accuracy. Data from a denitrification filter at a wastewater treatment plant in Jiangsu Province were used for training and testing. Compared with CNN and LSTM, the results showed that the MER and MSE metrics of the self-attention model is the best for the prediction of effluent TN, COD and NH₄⁺-N, achieving more accurate prediction. The predictive control model using a particle swarm optimization algorithm to calculate the dosage was applied to this wastewater treatment plant. The results showed that the effluent achieved more than 95% compliance for TN and the optimized dosage was 28.72% and 21.78% lower than the monthly average dosage of the sewage treatment plant respectively.
- deep learning /
- model predictive control /
- interpretable machine learning /
- self-attention mechanism /
- particle swarm optimization /
- accurate dosing

HTML全文

参考文献(21)

[1]	邱勇, 李冰, 刘垚, 等.污水处理厂化学除磷自动控制系统优化研究[J].给水排水, 2016, 52(7):126-129.
[2]	侯莹.城市污水处理过程多目标优化控制方法及应用研究[D].北京:北京工业大学, 2018.
[3]	杜胜利, 张庆达, 曹博琦, 等.城市污水处理过程模型预测控制研究综述[J].信息与控制, 2022, 51(1):41-53.
[4]	孙培德, 杨朋飞, 楼菊青, 等.全耦合活性污泥模型(FCASM3)在A+A²/O工艺污水处理厂中的数值模拟应用[J].环境科学学报, 2018, 38(9):3561-3572.
[5]	孙月娣.Bardenpho工艺内回流与碳源投加耦合控制动态模拟[J].中国给水排水, 2017, 33(23):66-70.
[6]	吴宇行, 王晓东, 陈宁, 等.典型城镇污水处理厂碳源智能投加控制生产性试验[J].环境工程, 2022, 40(6):212-218 , 271.
[7]	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, 784:147138.
[8]	LI M, HU S, XIA J, et al.Dissolved oxygen model predictive control for activated sludge process model based on the fuzzy C-means cluster algorithm[J].International Journal of Control, Automation and Systems, 2020, 18:2435-2444.
[9]	SUN J, PENG L, YAN X, et al.Reducing aeration energy consumption in a large-scale membrane bioreactor:process simulation and engineering application[J].Water Research, 2016, 93:205-213.
[10]	徐承志, 操家顺, 罗景阳, 等.活性污泥数学模型在污水处理中的研究进展[J].应用化工, 2021, 50(5):66-70.
[11]	WANG D, THUNELL S, LINDBERG U, et al.Towards better process management in wastewater treatment plants:process analytics based on SHAP values for tree-based machine learning methods[J].Journal of Environmental Management, 2022, 301:113941.
[12]	SONG M, CHOI S, BAE W, et al.Identification of primary effecters of N₂O emissions from full-scale biological nitrogen removal systems using random forest approach[J].Water Research, 2020, 184:116144.
[13]	YAQUB M, ASIF H, KIM S, et al.Modeling of a full-scale sewage treatment plant to predict the nutrient removal efficiency using a long short-term memory (LSTM) neural network[J].Journal of Water Process Engineering, 2020, 37:101388.
[14]	NIU G, YI X, CHEN C, et al.A novel effluent quality predicting model based on genetic-deep belief network algorithm for cleaner production in a full-scale paper-making wastewater treatment[J].Journal of Cleaner Production, 2020, 265:121787.
[15]	LECUN Y, BENGIO Y, HINTON G.Deep learning[J].Nature, 2015, 521(7553):436-444.
[16]	朱张莉, 饶元, 吴渊, 等.注意力机制在深度学习中的研究进展[J].中文信息学报, 2019, 33(6):1-11.
[17]	CHAUDHARI S, MITHAL V, POLATKAN G, et al.An attentive survey of attention models[J].ACM Transactions on Intelligent Systems and Technology, 2021, 12(5):1-32.
[18]	FU J, LIU J, TIAN H, et al.Dual Attention Network for Scene Segmentation[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019:3141-3149.
[19]	VASWANI A, SHAZEER N, PARMAR N, et al.Attention Is All You Need[C]//Advances in Neural Information Processing System, 2017.
[20]	XU J, SUN X, ZHANG Z, et al.Understanding and Improving Layer Normalization[C]//Advances in Neural Information Processing System, 2019:4381-4391.
[21]	KENNEDY J, EBERHART R.Particle swarm optimization[C]//1995 IEEE International Conference on Neural Networks Proceedings, 1995:1942-1948.