An effective data-driven water quality modeling and water quality risk assessment method

Created by W.Langdon from gp-bibliography.bib Revision:1.8414

@Article{Zhao:2024:engappai,

• author = "Zhiyao Zhao and Bing Fan and Yuqin Zhou and Ding Wang",
• title = "An effective data-driven water quality modeling and water quality risk assessment method",
• journal = "Engineering Applications of Artificial Intelligence",
• year = "2024",
• volume = "138",
• pages = "109457",
• keywords = "genetic algorithms, genetic programming, Water quality model, Data-driven water quality modeling, Water quality index, Risk assessment algorithm",
• ISSN = "0952-1976",
• URL = "https://www.sciencedirect.com/science/article/pii/S0952197624016154",
• DOI = "doi:10.1016/j.engappai.2024.109457",
• abstract = "Water quality modelling and assessment are crucial for preventing and controlling water pollution and play a significant role in environmental protection. However, few existing water body modelling methods consider the issue of model complexity and most of these methods are fixed-value evaluation methods, which makes it challenging to describe the uncertainty of the water evolution process. In order to address these issues, this paper introduces a data-driven water quality modelling and water quality risk assessment method. The proposed method is structured into three components. Firstly, a multi-water quality indicators modelling method is developed to account for model complexity, using the multi-gene genetic programming and the artificial bee colony algorithm. This method enables the simultaneous determination of model structure and parameters. Secondly, an adaptive water quality risk assessment method, referred to as WQI-IEW, which is based on the extended Kalman filter and an enhanced entropy weight method, is formulated in recognition of the water environment's complexity. Finally, in consideration of prediction uncertainties, the Markov chain is integrated with the previously developed WQI-IEW method to create the water quality risk prediction and assessment method, referred to as MCWQI-IEW. By testing the proposed enhanced modelling method on an actual water quality dataset and comparing it with relevant conventional methods, improvements were observed across various water quality parameters evaluation indicators, with improvements reaching up to 89percent in certain indicators",

}

Genetic Programming entries for Zhiyao Zhao Bing Fan Yuqin Zhou Ding Wang

Citations