摘要
Accurate water quality prediction plays a vital role in sustainable water management. The artificial intelligence models commonly used in water management including artificial neural networks, long short-term memory, support vector machine, adaptive neuro-fuzzy inference are reviewed. The hybrid models, artificial intelligence coupled with decomposition and optimization algorithms, are thoroughly discussed. After a brief introduction of each model, a review of recent publications to understand the potential and application of these models in surface water quality modeling is proposed. Based on the selected 60 papers published over the past decade (2013-2023), the statistical results in terms of data pre-processing, input and output parameters, modeling approaches, and performance evaluation are analyzed to reveal the latest trends. Artificial neural network has flexible variants to suit many scenarios and long short-term memory model has the advantage of processing time series data. Most remarkable performance in the single and hybrid algorithms is found. Hybrid algorithms generate more satisfactory results in predicting accuracy. The application of these models improves the decision-making mechanisms for environmental governance and show immense potential for various applications.
学科与关键词
- A Review of the Artificial Neural Network Models for Water Quality PredictionChen, Yingyi*; Song, Lihua; Liu, Yeqi; Yang, Ling; Li, DaoliangApplied Sciences-Basel, 2020, 10(17): 5776.被引 216
- A survey on river water quality modelling using artificial intelligence models: 2000-2020Tiyasha*; Tran Minh Tung; Yaseen, Zaher Mundher*Journal of Hydrology, 2020, 585: 124670.被引 390
- Application and innovation of artificial intelligence models in wastewater treatmentXu, Wen-Long; Wang, Ya-Jun; Wang, Yi-Tong*; Li, Jun-Guo; Zeng, Ya-Nan; Guo, Hua-Wei; Liu, Huan; Dong, Kai-Li; Zhang, Liang-YiJournal of Contaminant Hydrology, 2024, 267: 104426.被引 1
