Because the movement of water and sediment is complex, the flow pattern in a forebay can be disordered, which can significantly affect the pumping efficiency of a pump station. Using a pumping station in Ningxia, a multiphase water–sand flow model is built based on field measurements and practical engineering. The vortex flow pattern and the cause of sediment deposition are analyzed by numerical simulations and experiments. Furthermore, 36 rectification measures are proposed to improve the vortex flow pattern and sediment deposition in the forebay. The results show that various rectification measures can correct the flow pattern disorder in the horizontal section of the forebay at different angles. However, the correction effect is not obvious in the slope section. For the first time, a combination plan of adding plates to the back wall of the pumping station is proposed. This strategy not only improves the vortex area of the horizontal section of the forebay but also diverts the vortices of the slope section, reducing the vortex range and the sediment content. At the same time, the flow velocity uniformity of the inlet section of the combined optimal solution increases by 19.22%, and the shear angle decreases by 8.81°. The combination of numerical simulation and experimental analysis demonstrates the accuracy of the combined plan, corrects the flow pattern in the forebay, and reduces the sediment content.

In the circular forebay of sediment-laden rivers, adverse flow patterns such as vortices emerge, leading to substantial sediment accumulation along the sidewalls of the diversion channel and in the forebay. This sediment accumulation significantly impacts the operational efficiency of water pumps. To address these challenges, a combination of numerical simulations and engineering experiments was employed to reveal the factors contributing to the development of these patterns. Based on these findings, seven distinct flow rates and operating conditions were introduced to obtain the optimal inlet flow rate and increase the sustainability of engineering use. The findings revealed that the different inlet flow rates exert varying influences on the flow patterns in the circular forebay, leading to changes in sediment levels in both the forebay and diversion channels. Following comprehensive comparison, Scheme 6 was identified as the optimal flow rate scheme. In this scheme, the flow uniformity, deviation angle, ratio of the vortex reduction area parameter, and rate of vortex area reduction in sections z1−3 were markedly enhanced. The effectiveness of this scheme was validated through rigorous numerical simulations and engineering experiments, thereby offering indispensable theoretical insights into the operation of pumping stations in circular forebays.

Water quality directly determines our living environment. In order to establish a more scientific and reasonable water quality evaluation model, it needs a lot of data support, but it will lead to a large increase in the calculation time of the evaluation model. This paper proposes an improved particle swarm optimization SVM model (CPOS-SVM) to solve this problem. In this paper, the Pareto optimal solution concept is used to sparsely process the training set, which can ensure that the number of training sets is reduced without loss of data characteristics, thus reducing the training time. In order to solve the problem of the kernel parameter g and penalty factor c on the SVM algorithm, which affects the accuracy of the SVM model but it is difficult to select why, a particle swarm optimization algorithm is used in this paper to optimize the kernel parameter and penalty factor and improve the accuracy of the model. In this paper, 480 sets of data from Ming Cui Lake from 2014 to 2022 are taken as the research object, and examples are analyzed in MATLAB 2020a. The results show that the training time of the CPOS-SVM model can be completed within 2 s and does not increase with the increase of data volume. Meanwhile, by comparing the SVM model, POS-SVM model, and POS-BP model, training time increases dramatically with the amount of data. The accuracy of the POS-SVM model is the highest, and the accuracy of the CPOS-SVM model is basically consistent with that of the POS-SVM, reaching 94%, while the accuracy of the SVM model and the POS-BP model are slightly worse. This indicates that the CPOS-SVM model has good application value in water quality evaluation.

In light of the forebay of a lateral pump station in Shaanxi Province, numerical simulation and engineering tests were employed to analyze the correlation between the formation of multiple vortices and sediment deposition in the lateral forebay. Building upon this foundation, a total of 24 distinct rectification measures were proposed to scrutinize the alteration in flow regime within the lateral forebay, followed by an engineering test to validate the optimal scheme in the absence of rectification measures. The findings from the engineering test of the lateral forebay were consistent with the outcomes of the numerical simulation. This marked the inaugural proposition of a trapezoidal array clapboard within a lateral forebay. By adjusting the position, rotating the clapboard angle, and selecting the optimal scheme 24, the flow velocity uniformity at section C1-2 reached 93.16%, the deflection angle measured 94.32°, and the vortex area reduction rate was 87.73%, significantly mitigating the vortex impact on this section. Within section C1-3, the flow velocity uniformity attained 96.58%, the deflection angle measured 86.43°, and the vortex area reduction rate in this section amounted to 81.26%. Through the analysis of these two sections, the scheme's accuracy was validated. Additionally, the sediment deposition thickness decreased by 2.39 m under scheme 24, thus ensuring favorable operational conditions for the inlet forebay of the lateral pump station.

Dissolved oxygen (DO) is a key indicator for assessing the ecological health of nearshore harbors and managing the aquatic environment. Its concentration dynamics are governed by the nonlinear coupling of multiple factors, and obtaining accurate, interpretable estimates remains a central challenge in water quality modeling. Current research on DO estimation generally suffers from a disconnect between feature selection and model building, neglecting the systemic compatibility between the two. This results in suboptimal model configurations with poor site adaptability, redundant features that compromise model accuracy, and weak explanations of physical processes, making it difficult to meet the demands for refined water quality early warning systems. To address this issue, this study developed a Model-Feature System Matching (MFSM) framework based on 35 years (1988–2023) of monthly water quality monitoring data from Victoria Harbor in Hong Kong. By comparing all possible combinations of eight feature importance algorithms and five machine learning models, the framework enables adaptive diagnosis and adaptation of optimal modeling schemes for DO estimation at specific monitoring sites within the harbor. The results indicate that the optimal configuration for Station VM1 is a Pearson feature selection-coupled MLP model, which achieves the highest estimation accuracy (RMSE = 0.1292 mg·L⁻¹, R² = 0.9904). The Random Forest model demonstrated the most significant improvement in accuracy after feature optimization, with RMSE decreasing from 0.4898 to 0.2009. Chlorophyll a, water temperature, and ammonia nitrogen are the key factors driving DO variations. SHAP LOWESS interpretable analysis identified the transition intervals for the empirical effects of temperature and salinity (water temperature 95% CI: [21.69, 22.26] °C; salinity 95% CI: [32.79, 33.16] psu), indicating a unique nonlinear response pattern for DO in the semi-enclosed Victoria Harbor. Independent results from multiple sites demonstrate that the MFSM framework improves the estimation performance of feature-redundancy-sensitive models (Random Forest) across multiple subregions of Victoria Harbor (RMSE reduction ranging from 31.1% to 59%). The adaptive modeling framework developed in this study addresses the issue of blind configuration in traditional DO estimation and provides reliable technical support for accurate water quality estimation, ecological association analysis, and intelligent estimation and control in coastal semi-enclosed bays.