To address the issue of degraded predictive model performance caused by high-dimensional breach features and strong correlations among features, a peak discharge prediction model for earth-rock dam breaches was developed by integrating the Lasso algorithm with the XGBoost model. This model uses the Spearman correlation coefficient method to analyze the correlations between features, utilizes the Lasso algorithm for further feature selection, and obtains an optimal feature subset by eliminating redundant features. The feature subset is input into the XGBoost model to predict the peak breach discharge. Comparative results with support vector regression and ridge regression machine learning models show that the proposed model exhibits strong nonlinear information mining capability, effectively reduces the dimensionality of high-dimensional features, and improves the prediction accuracy while reducing model complexity.

In order to study the effect of wind fields during the water diversion period on the water transmission fields of Gaoyou Lake and Shaobo Lake in the inlet waterway of the Huaihe River, a two-dimensional wind flow numerical model of the study area was established by MIKE21 Flow Model, and the hydrodynamic changes and water quality renewal laws of the two lakes were analyzed under the action of prevailing wind fields. The results show that under the action of wind load, wind circulations of different structure are formed in Gaoyou Lake, and the flow direction and shape of the circulation are affected by the wind direction of the wind field; the upper and lower lakes of the Xinmintan control line are more significant areas of wind surge, and the higher the wind speed, the greater the water level variation, which may lead to a low design head of the pumping station project and insufficient water transfer power; the shear and drag effect of the wind field also has an impact on the renewal and replacement law of the water bodies in the lake area and the Yangtze River, and the renewal speed and degree of the water quality in the lake area are significantly improved under downwind conditions. Under headwind conditions, the degree of renewal of water quality in the lake area is decreased significantly.

In order to control the effluent discharge of the Shaying River more effectively and to ensure the normal operation on the water diversion project on the Xifei River line, a one-dimensional hydrodynamic-water quality coupling model was established from MIKE11, and the degradation and dilution characteristics of pollutants in the process of sewage effluent discharge were analyzed in terms of the mechanism of the influence on the water quality of the water-intake of Xifei River, using ammonia nitrogen as the water quality index. The results show that before flood, with the increase in the ratio of feeder flow, the pollution mass transitions from being mainly subject to degradation to dilution and the effluent discharged at the beginning of the flood mainly relies on the dilution effect of the main stream to reduce the concentration. When the sewage flow of the Shaying River before and at the beginning of the flood is greater than that of the main stream by 10% and 7%, respectively, the water quality at the water intake is at risk of exceeding Class Ⅲ of surface water standard. There is a power function relationship between the concentration of ammonia nitrogen at the intake and the branch to trunk flow ratio, and the fitting effect of the two is related to the degree of dilution of the polluted mass by the main stream.