PREDICTION OF COAGULANT DOSAGE FOR IN-SITU TURBIDITY CONTROL IN WATER ECOLOGICAL RESTORATION BASED ON BP NEURAL NETWORK OPTIMIZED BY GENETIC ALGORITHM

YU Feng; WANG Kejia; ZHANG Wenlong; LI Yi

doi:10.13205/j.hjgc.202304022

Volume 41 Issue 4

Apr. 2023

Turn off MathJax

Article Contents

Article Navigation > ENVIRONMENTAL ENGINEERING > 2023 > 41(4): 154-163

YU Feng, WANG Kejia, ZHANG Wenlong, LI Yi. PREDICTION OF COAGULANT DOSAGE FOR IN-SITU TURBIDITY CONTROL IN WATER ECOLOGICAL RESTORATION BASED ON BP NEURAL NETWORK OPTIMIZED BY GENETIC ALGORITHM[J]. ENVIRONMENTAL ENGINEERING , 2023, 41(4): 154-163. doi: 10.13205/j.hjgc.202304022

Citation:

YU Feng, WANG Kejia, ZHANG Wenlong, LI Yi. PREDICTION OF COAGULANT DOSAGE FOR IN-SITU TURBIDITY CONTROL IN WATER ECOLOGICAL RESTORATION BASED ON BP NEURAL NETWORK OPTIMIZED BY GENETIC ALGORITHM[J]. ENVIRONMENTAL ENGINEERING , 2023, 41(4): 154-163. doi: 10.13205/j.hjgc.202304022

Citation:

YU Feng, WANG Kejia, ZHANG Wenlong, LI Yi. PREDICTION OF COAGULANT DOSAGE FOR IN-SITU TURBIDITY CONTROL IN WATER ECOLOGICAL RESTORATION BASED ON BP NEURAL NETWORK OPTIMIZED BY GENETIC ALGORITHM[J]. ENVIRONMENTAL ENGINEERING , 2023, 41(4): 154-163. doi: 10.13205/j.hjgc.202304022

PDF( 8206 KB)

PREDICTION OF COAGULANT DOSAGE FOR IN-SITU TURBIDITY CONTROL IN WATER ECOLOGICAL RESTORATION BASED ON BP NEURAL NETWORK OPTIMIZED BY GENETIC ALGORITHM

doi: 10.13205/j.hjgc.202304022

YU Feng^1,2,
WANG Kejia²,
ZHANG Wenlong^{1,2
,
,},
LI Yi^1,2

1. Key Laboratory of Integrated Regulation and Resources Development of Shallow Lakes of Ministry of Education, College of Environment, Hohai University, Nanjing 210098, China;
2. College of Environment, Hohai University, Nanjing 210098, China

Received Date: 2022-02-21
Available Online: 2023-05-26
Publish Date: 2023-04-01

Abstract

Abstract

After the rainstorm, the turbidity of rivers and lakes increased sharply, which seriously interfered with the restoration and reconstruction of submerged plants in the ecological restoration projects of rivers and lakes. For the problem that the selection and dosage of coagulants in the process of in-situ coagulation and turbidity control in water ecological restoration are difficult to determine, in this study, the simulated river and lake turbid water samples were coagulated under laboratory conditions, and the coagulation prediction data set was constructed. BP neural network model was used to predict the dosage of coagulant, and the genetic algorithm was used to optimize the prediction model. Based on the coagulation experiment results, and the coagulation effect and cost, the coagulation data of the coagulant (aluminum sulfate) with better coagulation performance, and the dosage range (0~30 mg/L) with significant differences in the coagulation effect between different dosages were selected to train the coagulation prediction model. The results showed that, 1) the performance of the BP neural network regression model (R² was 0.78) was better than the multivariate nonlinear, multiple linear regression model and BP neural network classification model, and the prediction error of 88.67% of the samples was below 5 mg/L. After optimization by the genetic algorithm, the model R² was improved to 0.86 and the prediction error of more than 95% of the samples was below 5 mg/L, indicating that the genetic algorithm effectively improved the prediction accuracy and prediction stability of the model. 2) in addition to the amount of modelling data, the coagulant dosing gradient was another important factor affecting the performance of the model. In practical application, the amount of modelling data should be increased as much as possible and the dosing gradient should be reduced, to improve the performance of the coagulation dosing prediction model. The research results provide a reliable theoretical basis for the selection of coagulant types and dosage in the process of in-situ coagulation and turbidity control in water ecological restoration.
- BP neural network,
- genetic algorithm,
- water environment restoration,
- in-situ coagulation,
- optimize coagulant dosage

FullText(HTML)

References(35)

References

[1]	郭雅倩, 薛建辉, 吴永波, 等. 沉水植物对富营养化水体的净化作用及修复技术研究进展[J]. 植物资源与环境学报, 2020, 29(3): 58-68.
[2]	郭超, 李为, 李诗琦, 等. 盐龙湖沉水植物群落变化规律及其驱动因子研究[J]. 水生态学杂志, 2021, 42(6): 34-40.
[3]	徐德瑞, 周杰, 吴时强, 等. 夏季东太湖光合有效辐射衰减特征及其对沉水植物恢复的指示[J]. 湖泊科学, 2021, 33(1): 111-122.
[4]	FROGNER-KOCKUM P, GÖRANSSON G, HAEGER-EUGENSSON M. Impact of climate change on metal and suspended sediment concentrations in urban waters[J]. Frontiers in Environmental Science, 2020, 8: 588335.
[5]	刘畅, 韩梅, 赵志伟, 等. 混凝投药控制系统的发展现状与趋势[J]. 给水排水, 2021,57(增刊1): 524-530.
[6]	饶小康, 贾宝良, 鲁立. 基于人工神经网络算法的水厂混凝投药控制系统研究与开发[J]. 长江科学院院报, 2017, 34(5): 135-140.
[7]	KUSUMA H S, AMENAGHAWON A N, DARMOKOESOEMO H, et al. Evaluation of extract of Ipomoea batatas leaves as a green coagulant-flocculant for turbid water treatment: parametric modelling and optimization using response surface methodology and artificial neural networks[J]. Environmental Technology & Innovation, 2021, 24: 102005.
[8]	LI L, RONG S M, WANG R, et al. Recent advances in artificial intelligence and machine learning for nonlinear relationship analysis and process control in drinking water treatment: a review[J]. Chemical Engineering Journal, 2021, 405: 126673.
[9]	SUI KIM I T, SETHU V, ARUMUGASAMY S K, et al. Fenugreek seeds and okra for the treatment of palm oil mill effluent (POME)-Characterization studies and modeling with backpropagation feedforward neural network (BFNN)[J]. Journal of Water Process Engineering, 2020, 37: 101500.
[10]	张长胜, 韩涛, 钱斌, 等. 改进BFO算法优化BPNN的自来水混凝加药预测模型[J]. 中国环境科学,2021,41(10): 4616-4623.
[11]	叶伯生, 谢鹏, 张文彬. 基于随机遗传算法优化BP神经网络的工业机器人整机性能评估模型[J]. 中南大学学报(自然科学版), 2021, 52(9): 3204-3211.
[12]	ZHANG Y Y, GAO X, SMITH K, et al. Integrating water quality and operation into prediction of water production in drinking water treatment plants by genetic algorithm enhanced artificial neural network[J]. Water Research, 2019, 164: 114888.
[13]	樊琦. 基于遗传算法和BP神经网络的微涡流混凝投药控制模型研究[D].上海:华东交通大学,2018.
[14]	王新民, 柯愈贤, 张钦礼, 等. 磁化处理全尾砂料浆沉降规律及其参数优化[J]. 中国矿业大学学报, 2017, 46(4): 803-808.
[15]	王晋, 林超, 张毅敏, 等. 水体浊度对沉水植物菹草生长的影响[J]. 生态与农村环境学报, 2015, 31(3): 353-358.
[16]	朱光敏. 水体浊度和低光条件对沉水植物生长的影响[D].南京:南京林业大学,2009.
[17]	张彦, 邹磊, 梁志杰, 等. 暴雨前后河南北部河流水质分异特征及其污染源解析[J]. 环境科学,2022,43(5): 2537-2547.
[18]	张帅, 赵志伟, 丁昭霞, 等. 针对高浊山溪水的除浊工艺构建与效能研究[J]. 中国给水排水, 2018, 34(13): 38-42.
[19]	辛苑, 李萍, 吴晋峰, 等. 强降雨对北运河流域沙河水库水质的影响[J]. 环境科学学报, 2021, 41(1): 199-208.
[20]	张淳, 徐东耀, 康赛, 等. 磁混凝预处理小城镇混合污水的效能与混凝机制研究[J]. 环境科学学报,2022,42(7): 268-278.
[21]	李祥林, 钟建东. 响应面法优化聚合氯化铝混凝效果的研究[J]. 中国给水排水, 2015, 31(21): 141-143.
[22]	QIAO Y H, FENG J F, LIU X, et al. Surface water pH variations and trends in China from 2004 to 2014[J]. Environmental Monitoring and Assessment, 2016, 188(7): 443.
[23]	ZHANG W L, SHI M, WANG L Q, et al. New insights into nitrogen removal potential in urban river by revealing the importance of microbial community succession on suspended particulate matter[J]. Environmental Research, 2022, 204: 112371.
[24]	YAN Z G, FAN J T, ZHENG X, et al. Neglect of temperature and ph impact leads to underestimation of seasonal ecological risk of ammonia in chinese surface freshwaters[J]. Journal of Chemistry, 2019: 3051398.
[25]	QIU R J, WANG Y K, WANG D, et al. Water temperature forecasting based on modified artificial neural network methods: two cases of the Yangtze River[J]. Science of the Total Environment, 2020, 737: 139729.
[26]	汪萌. 上海市河流表层水体固氮速率及其影响因子[D].上海:华东师范大学,2017.
[27]	张佩. 都市区小型湖库水体碳形态与甲烷排放的时空特征研究[D].重庆:重庆大学,2020.
[28]	张艳晴, 周东, 周升, 等. 软围隔系统在迎风岸坡植被恢复中的应用研究[J]. 安徽农业科学, 2022, 50(6): 193-197.
[29]	王琦, 韩煜, 史娜娜, 等. 沉水植物群落重构技术在滇池草海水生态修复中的应用[C]//中国环境科学学会. 2020中国环境科学学会科学技术年会论文集(第2卷), 2020: 216-223.
[30]	年跃刚, 宋英伟, 李英杰, 等. 富营养化浅水湖泊稳态转换理论与生态恢复探讨[J]. 环境科学研究, 2006,19(1): 67-70.
[31]	HORNIK K, STINCHCOMBE M, WHITE H. Multilayer feedforward networks are universal approximators[J]. Neural Networks, 1989, 2(5): 359-366.
[32]	GRIFFITHS K A, ANDREWS R C. The application of artificial neural networks for the optimization of coagulant dosage[J]. Water Supply, 2011, 11(5): 605-611.
[33]	ZHANG Q, STANLEY S J. Real-time water treatment process control with artificial neural networks[J]. Journal of Environmental Engineering, 1999, 125(2): 153-160.
[34]	MAIER H. Use of artificial neural networks for predicting optimal alum doses and treated water quality parameters[J]. Environmental Modelling & Software, 2004, 19(5): 485-494.
[35]	VINITHA E V, MANSOOR AHAMMED M, GADEKAR M R. Chemical coagulation of greywater: modelling using artificial neural networks[J]. Water Science and Technology, 2018, 2017(3): 869-877.