The water level of Poyang Lake is affected by both the Yangtze River mainstream and the Fiver Rivers (namely, Ganjiang River, Fuhe River, Xinjiang River, Raohe River and Xiushui River) in the lake basin. The runoff of the Five Rivers is the main inflow of Poyang Lake. In order to explore the temporal and spatial impact of the Five Rivers on the water level of Poyang Lake, the response of Poyang Lake level variation to the runoff change of the Five Rivers was quantitatively studied by data-driven simulation and sensitivity analysis. Firstly, the lake level prediction model was established by combining PSO-GA algorithm and support vector regression (SVR) technology. The temporal sensitivity of water level of Poyang Lake to runoff change of the Five Rivers was simulated. Then, based on this model, the spatial distribution characteristics of water level response of Poyang Lake to runoff change of the Five Rivers were studied using continuous model simulation. The results show that the most sensitive period that Poyang Lake level response to the runoff change of the Five Rivers is from May to July, especially in May. In terms of spatial distribution, water level in the middle part of Poyang Lake is most sensitive to the runoff change of the Five Rivers. For instance, based on the data in May 2010, when the runoff of the Five Rivers was increased by 40%, water level at Duchang station was averagely increased by 1.16 m within one week; Average water level increased at Xingzi station and Hukou station in the north lake area was 1.15 m and 1.03 m, respectively, and that at both Tangyin station and Kangshan station in the south lake area was 0.97 m. This study clarified the temporal and spatial response characteristics of water level of Poyang Lake to runoff of the Five Rivers, and provides scientific basis for water resources management and water projects optimization in the Poyang Lake basin.

To avoid flow surface morphology change caused by wear and cavitation erosion of turbine movable guide vane and the unit efficiency decrease, ultrasonic vibration cavitation machine, high-speed jet of gaza's grinding apparatus and circle grinding apparatus were used to carry out cavitation erosion test for epoxy mortar, mortar composite resin and polyurethane and movable guide vane base material of ^#45 steel. Based on the test results, the cavitation erosion resistance and impact wear resistance of different materials were analyzed, and the service life of materials under strong cavitation erosion was predicted. The influence of flow rate, attack angle and sand content on material wear was studied, and the prediction model of material wear was established. The results show that the corrosion resistance of the materials is as follows: polyurethane > composite resin mortar > epoxy mortar; The abrasion loss of material has a linear or exponential relationship with the increase of velocity. The relationship between wear and sand content is linear. When the attack angle increases gradually, the impact wear first reaches the peak value and then decreases gradually. The critical attack angle of different materials is different.

In view of a flood discharge hole of a hydropower station, two types of flip buckets were proposed. The experimental study was carried out with the arc bevel and the new upper and lower composite flip bucket for hydraulic model in flood discharge hole. The parameters of two types of flip buckets such as shape of the water jets, jet trajectory length and the depth of downstream scour hole were compared and analyzed. The test results show that the new type of flip bucket divides the water flow into two parts, forming two upper and lower parallel floors water jets, which fall into the river channel in a parabolic shape, and the width of the two layers of water jets is controlled within the range of the river channel. The trajectory water jets have a stable flow state and are fully diffused in space and plane. Compared with the arc bevel flip bucket, the scour depth of downstream riverbed, the surge height, and the bank flow velocity have all been reduced to a certain extent. Finally, the Ansys Fluent software was used to carry out numerical simulation verification of the designed new flip bucket, and the shape of the tongue, jet trajectory length, turbulent kinetic energy and turbulent kinetic energy dissipation rate under each working condition were obtained. Comparing with the results of numerical calculation and model test, the flow patterns of the two are consistent, and the results of the trajectory length and width of the water jets are basically consistent, which verifies the reliability of the numerical simulation, and provides convenience and technical support for research of ski-jump energy dissipation.

Aiming at the problem that the reliability calculation results of embankment slope stability are not accurate due to the fact that the correlation between variables cannot be considered in two-dimensional independent distribution, based on 171 sets of test data of the shear strength of the Yangtze River embankment shoreline soil, Copula function was used to study the correlation between the shear strength parameters of the Yangtze River embankment shoreline soil. The AIC and BIC criteria were used to identify the optimal edge distribution and Copula function, and a two-dimensional joint distribution model was constructed. The influence of data volume on the identification of correlation structure between parameters was analyzed. The results show that the two-dimensional joint distribution model based on Copula function can accurately characterize the correlation between soil parameters of the Yangtze River embankment shoreline; When the data volume is more than 24 groups, AIC and BIC criteria can accurately identify the optimal Copula function, which provides a reference for the reliability calculation of embankment slope and the construction of Copula function model.

In order to explore the relationship between macro and micro characteristics of hybrid fiber reinforced concrete under freeze-thaw cycles, rapid freeze-thaw cycle tests were carried out with freeze-thaw cycles and fiber volume content as variables. The test results show that under different freeze-thaw cycles, the compressive strength has a positive correlation with air content, bubble spacing coefficient and average chord length of bubbles, and a negative correlation with specific surface area of pores, while the freeze-thaw damage and pore parameters have an opposite law, and they all change approximately linearly; The larger the freeze-thaw cycles, the lower the compressive strength, the greater the freeze-thaw damage, the greater the air content, the bubble spacing coefficient and the average chord length of bubbles, and the smaller the specific surface area of pores; The incorporation of hybrid fibers optimizes the pore structure and improves the frost resistance.

Aiming at the problems of difficult modeling, low prediction accuracy and poor adaptability of the operation trend prediction of cascade pumping station units under the process of multi-factor participation, this study took Yanhuanding Yellow River Project in Ningxia as the research object, introduced the time series analysis method, and put forward the operation trend prediction method of pumping station units based on ARIMA and SVM combination model. The energy consumption and average load in the operation technical parameters of the unit were selected as the test samples. The ARIMA model was used to linearly fit the processed data, and the SVM model was used to predict the residual error to compensate for the nonlinear change in the operation of the unit. The prediction results of the combination model were obtained by combining the two prediction results. The results show that the optimal models are ARIMA (1, 1, 3) and ARIMA (2, 1, 1), and the optimal parameters of SVM model are c=38, g=0.06 and c=68, g=0.18, respectively. The goodness of fit of the combined model for the test samples were 0.999 2 and 0.998 4, R_RMSE were 1.67×10^-5 and 3.9×10^-8, the M_MAPE were 0.036 1 % and 0.074 7 %, indicating that the combined model has high accuracy and good effect in predicting the operation trend of pumping stations. ARIMA-SVM combination model can provide a theoretical basis for the optimization and upgrading of pumping station unit operation condition monitoring system.

In order to study the applicability of recycled aggregate in hydraulic asphalt concrete, the mix proportion test of recycled aggregate hydraulic asphalt concrete was carried out, and its mix proportion parameters were determined through uniaxial compression, splitting test and permeability (porosity). The asphalt aggregate ratio is 7.0%, the grading index is 0.37, the filler content is 13%, and the replacement rate of recycled aggregate is 50%. The recycled aggregate was modified with modifiers of different concentrations. The water absorption, apparent density, crushing index and adhesion of the recycled aggregate were analyzed to determine the modifier concentration that meets the test requirements. The modifier concentration in this test is 10% of water glass and 12% of silane coupling agent. Based on the above research, the applicability of recycled aggregate in hydraulic asphalt concrete was studied through water stability coefficient, thermal stability coefficient and permeability coefficient. The results show that the application of recycled aggregate in hydraulic asphalt concrete can meet the requirements of the specification, which can provide a reference for the popularization and application of recycled aggregate in hydraulic asphalt concrete.

In view of the problem that the compaction coefficient of cement improved silty fine sand road in Gobi area is difficult to meet the standard after the completion of rolling inspection as an engineering material, after comprehensive analysis, it is found that the timeliness of cement improved silty fine sand is not considered in on-site construction and inspection. Through the design of compaction test and unconfined compression test of cement improved silty fine sand under different delay time, the influence of delay time on the compaction characteristics and strength characteristics of cement modified silty fine sand was analyzed. Considering the strength change of Subgrade in seasonally frozen area after freeze-thaw cycle, the freeze-thaw cycle test of cement modified silty fine sand under the influence of delay time was designed. The results show that with the increase of delay time, the maximum dry density of cement modified silty fine sand first increases and then decreases, and the optimal water content first increases, then tends to be flat and then increases; The saturated 7 d unconfined compressive strength and the strength after freeze-thaw cycle decreased gradually with the increase of delay time; The structure formed by cement hydration inside admixture has been destroyed after being rolled and formed, reducing the cement effectiveness.

Geostress field plays an important role in the stability of tunnel surrounding rock. Based on the in-situ stress results and geological conditions in the cracked area of the secondary lining concrete of the Qinling water conveyance tunnel from Hanjiang-to-Weihe, three-dimensional hydraulic fracturing method was used to analyze the in-situ stress of the surrounding rock in Qinling tunnel. The results show that the maximum principal stress on the cross section of the horizontal hole is 16.3-37.9 MPa and the minimum principal stress on the cross section is 2.4-9.1 MPa in the horizontal drilling depth range of 9.2-26.6 m. In the vertical drilling depth range of 3.8-20.7 m, the maximum horizontal principal stress is mainly 21.5-30.5 MPa, the azimuth average is N57°E, and the minimum horizontal principal stress is mainly 12.7-16.5 MPa. The maximum and minimum and the spatial maximum principal stress are 24.3 MPa, 14.4 MPa and 24.6 MPa respectively, which are nearly horizontal distribution. The stress field of surrounding rock at the test site of tunnel rock mass is mainly horizontal stress and the rock mass stress is classified as "high in-situ stress".

Due to the randomness of concrete carbonation process, the results obtained by the existing carbonation depth prediction models are difficult to characterize the actual carbonation status of in-service concrete structures. Therefore, combining the existing stochastic carbonization model with the actual measurement information of the carbonization depth of in-service concrete structures, a calculation method of a priori distribution hyper-parameters was given when the mean and variance of carbonized population samples were unknown. A Bayesian random model for the carbonation depth of in-service concrete structures was established. Taking the measured concrete carbonation depth of spillway pier and other structures of Yanshan Reservoir as an engineering example, the results show that the Bayesian random model can better represent the actual carbonation condition of in-service concrete structure.