Skip to main content

Study on Water Quality Prediction of Urban Reservoir by Coupled CEEMDAN Decomposition and LSTM Neural Network Model

Abstract

Urban reservoir is one of the important urban drinking water sources, and it is of important significance to ensuring the safety of urban water supply. The water quality of the reservoir is an important factor affecting the safety of water supply. Timely and accurate water quality prediction is very important for the formulation of a scientific and reasonable reservoir water supply plan. Considering the problem of high requirement of basic data in constructing water quality hydrodynamic physical model, this paper established a new data-driven model of water quality prediction in urban reservoir based on the Long and Short-Term Memory (LSTM) model, and the water quality data’s decomposition is implemented through the Complete Ensemble Empirical Modal Decomposition with Adaptive Noise (CEEMDAN) method. This model can not only realize the water quality prediction during different foreseen periods, but also solve the problem of low prediction accuracy caused by the randomness and large volatility of the measured data. Taking Xili Reservoir in Shenzhen of China as an example, the prediction of water concentration including total nitrogen, ammonia nitrogen, total phosphorus and PH value of Xili reservoir was realized based on historical monitoring data. Through simulation calculation, the prediction results of total nitrogen, ammonia nitrogen, total phosphorus and PH value in the water quality prediction model are highly consistent with the measured results, it is found that the simulation effect is good, and this model can well simulate the reservoir’s water quality concentration change process. For the total nitrogen and ammonia nitrogen, the relative prediction error of the model can be controlled below 10%, which shows the rationality of the built model. The research of this paper can provide an important theoretical and technical support for the water quality prediction and operation plan formulation of Xili Reservoir.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Availability of Data and Materials

Some or all data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request.

References

  • Barzegar R, Aalami MT, Adamowski J (2020) Short-term water quality variable prediction using a hybrid CNN-LSTM deep learning model. Stoch Environ Res Risk Assess 1–19

  • Dogan E, Sengorur B, Koklu R (2009) Modeling biological oxygen demand of the Melen River in Turkey using an artificial neural network technique. J Environ Manage 90:1229–1235

    Article  Google Scholar 

  • Hochreiter S, Schmidhuber J (1997) LSTM can solve hard long time lag problems. Neural Info Process Syst 473–479

  • Hu ZH, Zhang YR, Zhao YC et al (2019) A water quality prediction method based on the deep LSTM network considering correlation in smart mariculture. Sensors 19:1420

    Article  Google Scholar 

  • Huang CJ, Kuo PH (2018) A deep CNN-LSTM model for particulate matter (PM2.5) forecasting in smart cities. Sensors18:2220

  • Kratzert F, Klotz D, Shalev G et al (2019) Towards learning universal, regional, and local hydrological behaviors via machine-learning applied to large-sample datasets. Hydrol Earth Syst Sci 23:5089–5110

    Article  Google Scholar 

  • Liu S, Peng Y, Shao YM et al (2019) Expressway travel time prediction based on gated recurrent unit neural networks. Appl Math Mech 40:1289–1298

    Google Scholar 

  • Sunna (2019) The application of Sunna Machine Learning Theory in runoff intelligent prediction. Huazhong University of Science and Technology

  • Taieb SB, Bontempi G, Atiya AF et al (2011) A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5 forecasting competition. Expert Syst Appl 39:7067–7083

    Article  Google Scholar 

  • Torres ME, Colominas MA, Schlotthuer G et al (2011) A complete ensemble empirical mode decomposition with adaptive noise. Brain Res Bull 125:4144–4147

    Google Scholar 

  • Wang YY, Zhou J, Chen KJ et al (2017) Water quality prediction method based on LSTM neural network, Nanjing, China: 2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE), pp 1–5

  • Wei YZ, Xu XN (2019) ULTRA-short-term wind speed prediction model using LSTM networks. J Electron Measure Instrument 33:64–71

    Google Scholar 

  • Wu ZH, Huang NE (2004) A study of the characteristics of white noise using the empirical model decomposition method. Proc R Soc Lond 460:1597–1611

    Article  Google Scholar 

  • Xiang Z, Yan J, Demir I (2020) A rainfall‐runoff model with LSTM‐based sequence‐to-sequence learning. Water Resour Res 56

  • Yin H, Zhang X, Wang F et al (2021) Rainfall-runoff modeling using LSTM-based multi- state- vector sequence- to- sequence model. J Hydrol 598:126378

    Article  Google Scholar 

  • Zou K, Li Z, Mu X et al (2021) Study on sewage quality prediction model based on LSTM-GRU. Chin Energy Environ Protect 43(12):59–63

    Google Scholar 

Download references

Funding

This study was financially supported by the Natural Science Foundation of China (52179016, 51809098), Natural Science Foundation of Hubei Province (2021CFB597), Natural Science Fund of Anhui Province (grant no. 2008085ME158).

Author information

Authors and Affiliations

Authors

Contributions

Z.L.: data curation, formal analysis, writing – original draft; J.Z.Q.: conceptualization, funding acquisition, methodology, supervision; H.S.S.: validation, software; D.J.F: investigation, visualization; W.P.F.: writing – original draft, writing – review & editing; Z.T.: funding acquisition, methodology.

Corresponding author

Correspondence to Zhiqiang Jiang.

Ethics declarations

Ethics Approval

Not applicable.

Consent to Participate

Not applicable.

Consent to Publish

All authors agree to publish.

Competing Interests

The authors declare no conflicts of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zhang, L., Jiang, Z., He, S. et al. Study on Water Quality Prediction of Urban Reservoir by Coupled CEEMDAN Decomposition and LSTM Neural Network Model. Water Resour Manage (2022). https://doi.org/10.1007/s11269-022-03224-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s11269-022-03224-y

Keywords

  • Water quality
  • Prediction
  • Neural network
  • Long and short-term memory network
  • CEEMDAN decomposition
  • Shenzhen
  • Xili Reservoir