Applying high-frequency surrogate measurements and a wavelet-ANN model to provide early warnings of rapid surface water quality anomalies

Shi, Bin1; Wang, Peng1,2; Jiang, Jiping1,3; Liu, Rentao1

2018-01

10.1016/j.scitotenv.2017.08.232

SCIENCE OF THE TOTAL ENVIRONMENT   影响因子和分区

0048-9697

1879-1026

It is critical for surface water management systems to provide earlywarnings of abrupt, large variations inwater quality, which likely indicate the occurrence of spill incidents. In this study, a combined approach integrating a wavelet artificial neural network (wavelet-ANN) model and high-frequency surrogate measurements is proposed as a method of water quality anomaly detection and warning provision. High-frequency time series of major water quality indexes (TN, TP, COD, etc.) were produced via a regression-based surrogate model. After wavelet decomposition and denoising, a low-frequency signalwas imported into a back-propagation neural network for one-step prediction to identify the major features of water quality variations. The precisely trained sitespecific wavelet-ANN outputs the time series of residual errors. A warning is triggered when the actual residual error exceeds a given threshold, i.e., baseline pattern, estimated based on long-term water quality variations. A case study based on the monitoring programapplied to the Potomac River Basin in Virginia, USA, was conducted. The integrated approach successfully identified two anomaly events of TP variations at a 15-minute scale from high-frequency online sensors. A storm event and point source inputs likely accounted for these events. The results showthat the wavelet-ANN model is slightly more accurate than the ANN for high-frequency surfacewater quality prediction, and it meets the requirements of anomaly detection. Analyses of the performance at different stations and over different periods illustrated the stability of the proposed method. By combining monitoring instruments and surrogate measures, the presented approach can support timely anomaly identification and be applied to urban aquatic environments for watershed management. (C) 2017 Elsevier B.V. All rights reserved.

Water quality Surrogate parameters Anomaly detection Wavelet denoising Back-propagation neural networks

SCI ; EI

英语

通讯

Southern University of Science and Technology[G01296001]

Environmental Sciences & Ecology

Environmental Sciences

WOS:000411897700142

ELSEVIER SCIENCE BV

20173504096815

Anomaly detection ; Backpropagation ; Neural networks ; Soil conservation ; Time series ; Water conservation ; Water management ; Water quality ; Wavelet decomposition

Water Resources:444 ; Surface Water:444.1 ; Water Analysis:445.2 ; Soils and Soil Mechanics:483.1 ; Artificial Intelligence:723.4 ; Mathematical Transformations:921.3 ; Mathematical Statistics:922.2

ENVIRONMENT/ECOLOGY

Web of Science

1.Harbin Inst Technol, Sch Environm, Harbin 150090, Heilongjiang, Peoples R China
2.Harbin Inst Technol, State Key Lab Urban Water Resource & Environm, Harbin 150090, Heilongjiang, Peoples R China
3.Southern Univ Sci & Technol, Sch Environm Sci & Engn, Shenzhen 518055, Peoples R China

Shi, Bin,Wang, Peng,Jiang, Jiping,et al. Applying high-frequency surrogate measurements and a wavelet-ANN model to provide early warnings of rapid surface water quality anomalies[J]. SCIENCE OF THE TOTAL ENVIRONMENT,2018,610:1390-1399.

Shi, Bin,Wang, Peng,Jiang, Jiping,&Liu, Rentao.(2018).Applying high-frequency surrogate measurements and a wavelet-ANN model to provide early warnings of rapid surface water quality anomalies.SCIENCE OF THE TOTAL ENVIRONMENT,610,1390-1399.

Shi, Bin,et al."Applying high-frequency surrogate measurements and a wavelet-ANN model to provide early warnings of rapid surface water quality anomalies".SCIENCE OF THE TOTAL ENVIRONMENT 610(2018):1390-1399.