A unified deep learning framework for water quality prediction based on time-frequency feature extraction and data feature enhancement

Rui Xu; Shengri Hu; Hang Wan; Yulei Xie; Yanpeng Cai; Jianhui Wen

doi:10.1016/j.jenvman.2023.119894

A unified deep learning framework for water quality prediction based on time-frequency feature extraction and data feature enhancement

J Environ Manage. 2024 Feb:351:119894. doi: 10.1016/j.jenvman.2023.119894. Epub 2023 Dec 27.

Authors

Rui Xu¹, Shengri Hu¹, Hang Wan², Yulei Xie³, Yanpeng Cai³, Jianhui Wen⁴

Affiliations

¹ School of Computer Science and Information Security, Guilin University of Electronic Technology, Guilin, 541004, China.
² Research Centre of Ecology & Environment for Coastal Area and Deep Sea, Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou), Guangzhou, 511458, China; Guangdong Provincial Key Laboratory of Water Quality Improvement and Ecological Restoration for Watersheds, School of Ecology, Environment and Resources, Guangdong University of Technology, Guangzhou, 510006, China. Electronic address: wanhang@gmlab.ac.cn.
³ Research Centre of Ecology & Environment for Coastal Area and Deep Sea, Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou), Guangzhou, 511458, China; Guangdong Provincial Key Laboratory of Water Quality Improvement and Ecological Restoration for Watersheds, School of Ecology, Environment and Resources, Guangdong University of Technology, Guangzhou, 510006, China.
⁴ Ecological and Environmental Monitoring Center of Guangxi, Guilin, 541002, China.

PMID: 38154219
DOI: 10.1016/j.jenvman.2023.119894

Abstract

Deep learning methods exhibited significant advantages in mapping highly nonlinear relationships with acceptable computational speed, and have been widely used to predict water quality. However, various model selection and construction methods resulted in differences in prediction accuracy and performance. Hence, a unified deep learning framework for water quality prediction was established in the paper, including data processing module, feature enhancement module, and data prediction module. In the established model, the data processing module based on wavelet transform method was applied to decomposing complex nonlinear meteorology, hydrology, and water quality data into multiple frequency domain signals for extracting self characteristics of data cyclic and fluctuations. The feature enhancement module based on Informer Encoder was used to enhance feature encoding of time series data in different frequency domains to discover global time dependent features of variables. Finally, the data prediction module based on the stacked bidirectional long and short term memory network (SBiLSTM) method was employed to strengthen the local correlation of feature sequences and predict the water quality. The established model framework was applied in Lijiang River in Guilin, China. The maximum relative errors between the predicted and observed values for dissolved oxygen (DO), chemical oxygen demand (COD_Mn) were 12.4% and 20.7%, suggesting a satisfactory prediction performance of the established model. The validation results showed that the established model was superior to all other models in terms of prediction accuracy with RMSE values 0.329, 0.121, MAE values 0.217, 0.057, SMAPE values 0.022, 0.063 for DO and COD_Mn, respectively. Ablation tests confirmed the necessity and rationality of each module for the established model framework. The established method provided a unified deep learning framework for water quality prediction.

Keywords: Deep learning; Feature enhancement; Informer encoder; Wavelet transform.

MeSH terms

China
Deep Learning*
Hydrology
Meteorology
Oxygen
Water Quality

Substances

Oxygen