Inadequate hydrodynamics can cause deterioration of the water environment within rivers. Regulating water conservancy projects can effectively improve the state of the water environment of rivers and promote sustainable regional development. The river plain in Zhejiang Province, China, suffers from severe hydrodynamic deficiencies, which have a significant impact on the state of the regional water environment. To investigate the changing state of the water environment of the river plain under water conservancy project control, in this study we constructed a two-dimensional hydrodynamic–water quality model based on the AD and ECOLAB modules in MIKE 21 software 2014 Edition. Combined with conservative tracers, the changes in the water displacement rate, flow rate, and water environment quality of the river plain were simulated under different regulation schemes over 7 days. A quantitative analysis of the effects of improving the state of the water environment in the river plain was carried out using a cluster analysis and a cloud model. (1) The water replacement rate of the outer river reached 90% after 3 days and approached 100% after 7 days. The water replacement rates of the inner pond were 51.2, 49.6, and 55.8%. This indicated that the engineering control measures effectively improved the replacement capacity of the river. (2) The contents of DO and BOD5 in the river have increased from class V to above class 3. The overall water quality is in the range of classes 2 to 3, and in some parts it can reach class 1. This indicates that the regulatory plan played a certain role in improving the river water environment. (3) The water pollution in the study area showed a fluctuating and decreasing trend over 7 days. There was a positive correlation between the flow velocity, water replacement rate, DO, and BOD5.
In plain river networks, a sluggish flow due to the flat terrain and hydraulic structures significantly reduces water’s capacity for self-purification, leading to persistent water pollution that threatens aquatic ecosystems and human health. Despite being critical, effective water quality prediction proves challenging in such regions, with current models lacking either physical interpretability or temporal accuracy. To address this gap, both a process-based model (MIKE 21) and a deep learning model (CNN-LSTM-Attention) were developed in this study to predict key water quality indicators—dissolved oxygen (DO), total nitrogen (TN), and total phosphorus (TP)—in a typical river network area in Jiaxing, China. This site was selected for its representative complexity and acute pollution challenges. The MIKE 21 model demonstrated strong performance, with R2 values above 0.88 for all indicators, offering high spatial resolution and mechanistic insight. The CNN-LSTM-Attention model excelled in capturing temporal dynamics, achieving an R2 of 0.9934 for DO. The results indicate the complementary nature of these two approaches: while MIKE 21 supports scenario-based planning, the deep learning model enables highly accurate real-time forecasting. The findings are transferable to similar river network systems, providing a robust reference for selecting modeling frameworks in the design of water pollution control strategies.