基于双向门控循环单元的地表水氨氮预测

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (1346 KB) HTML (0 KB)
输出: BibTeX | EndNote (RIS)

摘要为提高水环境中NH₄⁺-N的预测精度，提出了一种互补完全集合经验模式分解（CCEEMDAN）和双向门控循环单元（BiGRU）神经网络的混合预测模型（CCB）.首先，通过CCEEMDAN将NH₄⁺-N数据分解成一系列较为简单的模态成份；然后利用BiGRU神经网络对各成份进行预测，将所有分解成份的预测结果相加即可获得最终预测结果.以2017年6月~2020年2月鄱阳湖的NH₄⁺-N数据进行模型性能验证.结果表明，利用CCB模型在1d后的NH₄⁺-N预测中平均绝对百分比误差为3.38%，在7d后的NH₄⁺-N预测中平均绝对百分比误差为6.82%，在15d后的NH₄⁺-N预测中平均绝对百分比误差为9.41%，优于本文中参与比较的其他模型.CCB模型在NH₄⁺-N预测方面具有良好的预测性能.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	任永琴
	金柱成
	俞真元
	王晓丽
	彭士涛

关键词 ：鄱阳湖, 氨氮(NH₄⁺-N), 互补完全集合经验模式分解(CCEEMDAN), 双向门控循环单元(BiGRU)

Abstract：For more accurate prediction of NH₄⁺-N, this paper proposes a novel hybrid forecast model (CCB) that uses complementary complete ensemble empirical mode decomposition with adaptive noise (CCEEMDAN) and bidirectional gated recurrent unit (BiGRU) neural network. Firstly, the original NH₄⁺-N data is decomposed into several relatively simple components by CCEEMDAN. Subsequently, BiGRU neural network is employed to predict each component. The final forecast result is obtained by the summation of all the prediction results for the decomposed components. NH₄⁺-N data of Poyang Lake that was monitored from June, 2017 to February, 2020 is used to evaluate the proposed forecast model. Mean absolute percentage error (MAPE) of the forecast result by our model is 3.38% for 1day ahead forecast, 6.82% for 7days ahead forecast and 9.41% for 15days ahead forecast. Moreover, CCB model shows better forecast performance than the competitor models. Results demonstrate that CCB model has a powerful forecast capacity, and it can be effectively used for the analysis and decision-making in water resource management.

Key words： Poyang Lake ammonia nitrogen(NH₄⁺-N) complementary complete ensemble empirical mode decomposition with adaptive noise (CCEEMDAN) bidirectional gated recurrent unit (BiGRU)

收稿日期: 2021-07-10

PACS:

X524

基金资助:中央级公益性科研院所基本科研业务费专项资金资助项目（TKS190202，TKS20200405）；天津市科技计划项目（20JCQNJC00100）

通讯作者: 王晓丽,教授,tjutwxl@163.com E-mail: tjutwxl@163.com

作者简介: 任永琴(1995-),女,山西吕梁人,天津理工大学硕士研究生,主要研究方向为环境安全监测与评价.

引用本文:

任永琴, 金柱成, 俞真元, 王晓丽, 彭士涛. 基于双向门控循环单元的地表水氨氮预测[J]. 中国环境科学, 2022, 42(2): 672-679. REN Yong-qin, KIM Ju-song, YU Jin-won, WANG Xiao-li, Peng Shi-tao. Ammonia nitrogen prediction in surface water based on bidirectional gated recurrent unit. CHINA ENVIRONMENTAL SCIENCECE, 2022, 42(2): 672-679.

链接本文:

http://www.zghjkx.com.cn/CN/ 或 http://www.zghjkx.com.cn/CN/Y2022/V42/I2/672

[1]	Lu H, Ma X. Hybrid decision tree-based machine learning models for short-term water quality prediction[J]. Chemosphere, 2020:126169.
[2]	Chen Y Y, Song L H, Liu Y Q, et al. A review of the artificial neural network models for water quality prediction[J]. Applied Sciences-Basel, 2020,10(5776):2-49.
[3]	刘永华.测定地表水中氨氮影响因素的探讨[J].水资源开发与管理, 2021,(3):51-55,59. Liu Y H. Discussion on influencing factors of ammonia nitrogen determination in surface water[J]. Water Resources Development and Management, 2021,(3):51-55,59.
[4]	董怡华,张盛宇,陈峰,等.植物塘-人工湿地复合系统基质与植物筛选及农田退水处理研究[J]. 安全与环境学报, 2021,21(2):771-779. Dong Y H, Zhang S Y, Chen F, et al. Screening of substrates and plant in macrophyte pond-constructed wetland complex system and treatment of farmland drainage[J]. Journal of Safety and Environment, 2021,21(2):771-779.
[5]	石翠翠,刘媛华,陈昕.基于GNIPSO-SVR的水质预测模型研究[J/OL]. 信息与控制, 2021:https://kns.cnki.net/kcms/detail/21.1138.TP.20210617.1737.043.html. Shi C C, Liu Y H, Chen X. Water quality prediction model based on GNIPSO-SVR[J/OL]. Information and Control, 2021:https://kns.cnki.net/kcms/detail/21.1138.TP.20210617.1737.043.html.
[6]	梁玉,卢萍,张晓星,等.鄱阳湖湿地存在的问题及对策[J]. 山东林业科技, 2011,41(5):91-92,76. Liang Y, Lu P, Zhang X X, et al. Problems and countermeasures in Poyang wetland[J]. Journal of Shandong Forestry Science and Technology, 2011,41(5):91-92,76.
[7]	温春云,刘聚涛,胡芳,等.鄱阳湖水质变化特征及水体富营养化评价[J]. 中国农村水利水电, 2020,(11):83-88. Wen C Y, Liu J T, Hu F, et al. Water quality change characteristics and eutrophication assessment of Poyang Lake[J]. China Rural Water and Hydropower, 2020,(11):83-88.
[8]	陈旻坤,徐昔保.近30年来鄱阳湖生态系统服务变化[J]. 湖泊科学, 2021,33(1):309-318. Chen Y K, Xu X B. Lake Poyang ecosystem services changes in the last 30 years[J]. Journal of Lake Sciences, 2021,33(1):309-318.
[9]	熊捷迁,弓晓峰,江良,等.鄱阳湖水体沉积物中Zn-Cd对底栖生物的毒性效应及基准验证[J]. 湖泊科学, 2021,33(6):DOI:10.18307/2021.0607. Xiong J Q, Gong X F, Jiang L, et al. Toxic effects of zinc and cadium on the benthic organisms in sediments of Lake Poyang and verification of quality guideline[J]. Journal of Lake Sciences, 2021,33(6):DOI:10. 18307/2021.0607.
[10]	陈威,陈会娟,戴凡翔,等.基于人工神经网络的污水处理出水水质预测模型[J].给水排水, 2020,56(S1):990-994. Chen W, Chen H J, Dai F X, et al. Effluent water quality prediction model based on artificial neural network[J]. Water & Wastewater Engineering, 2020,56(S1):990-994.
[11]	Wang X Z, Wang K J, Ding J M, et al. Predicting water quality during urbanization based on a causality-based input variable selection method modified back-propagation neural network[J]. Environmental Science and Pollution Research, 2021,28(1):960-973.
[12]	Liang N, Zou Z, Wei Y. Regression models (SVR, EMD and FastICA) in forecasting water quality of the Haihe River of China[J]. Desalination and Water Treatment, 2019,154:147-159.
[13]	王明圆.基于灰色Elman神经网络的湖水水质指标预测研究[D]. 天津:天津大学, 2016. Wang M Y. The research on predicting The water quality of lake with grey elman neural model[D]. Tianjin:Tianjin University, 2016.
[14]	汤斌,赵敬晓,魏彪,等.一种紫外-可见光谱检测水质COD预测模型优化方法[J]. 中国环境科学, 2015,35(2):478-483. Tang B, Zhao J X, Wei B, et al. A method of optimizing the prediction model for the determination of water COD by using UV-visible spectroscopy[J]. China Environmental Science, 2015,35(2):478-483.
[15]	Cao X, Ren N, Tian G, et al. A three-dimensional prediction method of dissolved oxygen in pond culture based on Attention-GRU-GBRT[J]. Computers and Electronics in Agriculture, 2021,181.
[16]	Liang R, Chang X T, Jia P T, et al. Mine gas concentration forecasting model based on an optimized BiGRU Network[J]. Acs. Omega., 2020,5(44):28579-28586.
[17]	童俊.基于SVR-ARIMA组合模型的金泽水库氨氮预测研究[J].微型电脑应用, 2020,26(4):4-6,13. Tong J. Prediction of ammonia nitrogen in Jinze Reservoir based on SVR-ARIMA combination model[J]. Microcomputer Applications, 2020,26(4):4-6,13.
[18]	Cheng Y, Wang Z, Chen B, et al. An improved complementary ensemble empirical mode decomposition with adaptive noise and its application to rolling element bearing fault diagnosis[J]. ISA Trans., 2019,91:218-234.
[19]	袁伟皓,王华,曾一川,等.大型通江湖泊藻类增殖驱动要素时空分异特征[J/OL]. 环境工程, 2021,https://kns.cnki.net/kcms/detail/11.2097.X.20210525.1632.024.html. Yuan W H, Wang H, Zeng Y C, et al. spatiotemporal variation of driving factors of algal proliferation in a large river-connected lake[J/OL]. Environmental Engineering, 2021, https://kns.cnki.net/kcms/detail/11.2097.X.20210525.1632.024.html.
[20]	GB3838-2002地表水环境质量标准[S]. GB3838-2002 Environmental quality standard for surface water[S].
[21]	Zheng J D, Cheng J S, Yang Y. Partly ensemble empirical mode decomposition:An improved noise-assisted method for eliminating mode mixing[J]. Signal Processing, 2014,96:362-374.
[22]	Liu H, Liu Z, Jia W, et al. Remaining useful life prediction using a novel feature-attention-based end-to-end approach[J]. Ieee Transactions on Industrial Informatics, 2021,17(2):1197-1207.
[23]	Huang Y R, Chen J J, Zheng S M, et al. Hierarchical multi-attention networks for document classification[J]. International Journal of Machine Learning and Cybernetics, 2021,12:1639-1647.
[24]	Zhang Z, Dong Z, Lin H, et al. An improved bidirectional gated recurrent unit method for accurate state-of-charge estimation[J]. IEEE Access, 2021,9:11252-11263.
[25]	包智鹏,支永帅,张素民,等.基于BiGRU的多模态驾驶行为及轨迹预测[J]. 大连理工大学学报, 2021,61(3):246-254. Bao Z P, Zhi Y S, Zhang S M, et al. BiGRU based multi-modal maneuvers and trajectory prediction[J]. Journal of Dalian University of Technology, 2021,61(3):246-254.
[26]	Torres M E, Colominas M A, Schlotthauer G, et al. A complete ensemble empirical mode decomposition with adaptive noise[Z]. 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2011:4144-4147.
[27]	Dragomiretskiy K, Zosso D. Variational mode decomposition[J]. Ieee Transactions on Signal Processing, 2014,62(3):531-544.