In order to explore the cause of the flow pattern problem in the forebay of the pumping station at the diversion operating condition in Taihuzhuangyuan in Changzhou City, and find the corresponding solution, the three-dimensional turbulent numerical simulation method was used to calculate the inlet flow pattern after the prototype of the lock station and the addition of the diversion pier. By analyzing the flow uniformity and weighted average angle of the inlet section in each case, it can be concluded that the flow pattern of the forepbay is mainly caused by the large diffusion angle. The longer a single diversion pier is added to the forebay, the better the rectification effect will be. Compared with the single pier, the combined pier can weaken the recirculation region more and improve the inlet flow pattern. The research results can provide reference for engineering practice.

Aiming at the problem of sluice flow, based on the smooth particle hydrodynamics method (SPH), the boundary treatment method was improved, and the main parameters of repulsion force were re-calibrated, which can avoid non-physical oscillation with n₁=4 and n₂=2. At the same time, the comparison and selection of water replenishment modes were carried out, and it was clear that the bottom hole water replenishment mode with 3 m elevation difference in the optimal water replenishment mode can reduce the net outflow of particles from the upstream water body. On this basis, the numerical model of the flow through the gate impacting the stilling pool downstream was established, and the process of the flow from the sluice to the stilling pool was simulated. The results show that at the initial time, the maximum pressure at the bottom of the upstream water is 127.4 kPa; At the 7th second, the average velocity at the right end of the stilling pool is 7.07 m/s, at the 19th second, the average velocity at the right end of the stilling pool decreases to 1.4 m/s. Thus, the SPH method can accurately simulate the changes of flow velocity, pressure and flow pattern in the discharge process.

The weir-gate structure has larger discharge capacity. To accurately and efficiently check the discharge of weir-gate, three intelligent algorithms including BP neural network, SVM and GRNN were used to predict the discharge coefficient of cylindrical weir-gate. The correlation analysis and variation law between dimensionless parameters and discharge coefficient were discussed. The results show that the GRNN and the BP can accurately predict the discharge coefficient of the cylindrical weir-gate. The determination coefficient of the BP in the test stage is 0.997, the root mean square error is 0.009, the average absolute percentage is 0.801 %, and the Nash efficiency coefficient is 0.997, which is superior to the GRNN, and it can be used as an efficient and high-precision prediction model for the discharge coefficient of the weir-gate. There is a stronger correlation between the ratio of gate opening to cylinder diameter (a/D), the ratio of weir head to cylinder diameter (H_w/D) and C_d. The C_d increased with the increase of upstream Froude number (Fr) and H_w/D, and the greater the a/D is, the greater the increase of C_d is. The search results provide theoretical reference and technical support for the popularization and application of cylindrical weir gate in practical engineering.

In order to response the request of hydraulic digital twin, this paper studied how it implement in long linear engine hydraulic project so that the manage system can show any position and any changes of the project. The data linkage inquiry method in both two and three dimension and the shortest distance algorithm innovatively were proposed based on BIM, GIS and geological profile graph. It is implemented through data fusion, data structure design, software design etc., which can make digital scene and geological profile graph's view move and inquire together. The user can immediately acquire mileage, geological conditions, design section, construction and operating state through moving and pre-clicking the cursor. It's proved in a certain water diversion project that the project manager can inquire and acquire real-time project information immediately by using this method without inspect the whole engineering. This method lays the foundation of real-time data inquiry, engineering safety intelligent analysis and production operation management, which implement engineering digital twin to some extent.

To solve the problem of low recognition accuracy caused by insufficient power data feature mining, this paper proposed a novel power data identification method based on multi-domain feature analysis and feature selection. Firstly, aiming at the shortcomings of existing power data feature extraction methods, a feature extraction method based on empirical mode decomposition (EMD) and Hilbert transform (EMD-Hilbert) was proposed, and the power features and V-I trajectory features of power data were quantified. Secondly, based on random forest and generalized sequence backward selection search strategy, the optimal feature subset was obtained. The random forest was employed to build a recognition model for the power data. Finally, the experimental results verified the effectiveness and identification accuracy of the proposed method. The results show that the proposed method can utilize the complementarity of different features to overcome the problem of low accuracy by single feature, and further improve the model recognition performance through feature selection.

Considering the wide frequency range, strong nonlinearity, and multimodal characteristics of "double high" power systems, an adaptive identification method for broadband oscillation modes was proposed. Firstly, an adaptive threshold filtering technique was applied to the spectrum of the Blackman-Harris windowed signal to extract the dominant components. Then the three-spectrum line interpolation FFT and the ZoomFFT were used to adaptively analyze the signals after using main-lobe interference criterion of modes with close frequencies. Finally, the proposed method was tested by using 20-mode example signals and a measurement signal of a power system model. The results demonstrate that the method can accurately identify the parameters of complex broadband oscillation modes, adapt to various oscillation scenarios, and meet the signal characteristics of "double high" power systems, thus confirming its applicability and feasibility.

The precipitation data with high quality and high spatial and temporal resolution is of great significance to the research of hydrology, meteorology and other fields. At present, remote sensing and reanalysis of precipitation data are widely used, but there are problems such as low resolution, high uncertainty of accuracy, etc. In this paper, a random forest precipitation fusion algorithm considering covariates is proposed to fuse seven sets of precipitation products, namely CMA, CN05, ERA5, GLDAS, TRMM, IMERG and PERSIANN. Three typical sub-basins of the Yangtze River basin (Jinsha River, Sanxiaqujian and Poyang Lake) are selected to test the effect of random forest fusion data (RFF). The results show that for the accuracy of precipitation products, the accuracy of random forest fusion data is improved compared with the original precipitation products. For the accuracy assessment of different precipitation events, with the increase of rainfall intensity, the T_TS score of each precipitation product shows a decreasing trend, and the T_TS score of RFF is better than the original precipitation product. Data fusion of precipitation products through random forest model considering covariates can improve the accuracy of precipitation data and the reliability of different precipitation events, which provides support for hydrological simulation.

The Lhasa River Basin is a typical arid and semi-arid basin in the Qinghai-Tibet Plateau, where the ecosystem is extremely fragile. It is of great significance to study the spatio-temporal variation of vegetation index (NDVI) in response to the changes of meteorological factors, and to explore the adaptability of vegetation on the Qinghai-Tibet Plateau to the meteorological factors under the background of climate change. Based on the monthly NDVI, precipitation (P), and average temperature (T) time series dataset in the Lhasa River Basin from 1982 to 2017, using Pettitt, Mann-Kendall trend test and Pearson correlation analysis, this paper analyzed the spatio-temporal variation characteristics of growing period NDVI (G-NDVI) and meteorological factors, and identified responses patterns of G-NDVI to climate factors. The results show that the G-NDVI changed abruptly in 1997, and there was a trend shift from increasing to decreasing. The climate of the watershed changed from “wetting-colding” before the abrupt point to “drying-warming” after the abrupt point, and its effect on vegetation growth changed from promoting to inhibiting. There are two zones in the watershed where the responses patterns of G-NDVI to meteorological factors changed before and after the abrupt point. In the western permafrost areas, G-NDVI shows significant correlation with P and T in the second phase after 1997, i.e., emerging the “responding” function on meteorological factors variation. In the southern seasonal frozen zone, after the abrupt point, time lags of G-NDVI to P were elongated, and meanwhile the “responding” of G-NDVI to T has been triggered. The latency of NDVI response to P in the western permafrost region is longer than that in the southern seasonal frozen zone. In the second phase after 1997, the response area of NDVI to meteorological factors increased compared with that before the abrupt point, and the effect of meteorological factors on vegetation growth in the Lhasa River basin was enhanced.

Based on the monthly data of precipitation and average temperature from 36 reference meteorological stations in the six basins in Yunnan Province, 11 climate models published by CMIP6 were used to simulate and predict the change process of meteorological elements in six basins. Three model integration technologies were used to improve the simulation ability and evaluate the suitability of each model and integration method. The optimal method was chosen to predict the possible trend of climate change in six basins under the scenario of SSP2-4.5 in the future. The results showed that precipitation and temperature would increase gradually in the near, middle and long term in the six basins of Yunnan Province. Precipitation might increase significantly after 2060, and temperature might increase significantly after 2050.

The drought in the Fenhe River Basin has the characteristics of serious disaster and wide range of influence, which has a great blocking effect on the development of local society economy, so it is of great significance to accurately evaluate the drought situation in the basin. Based on Standardized Precipitation Index (SPI), Standardized Precipitation Evapotranspiration Index (SPEI) and Standardized Runoff Index (SRI), the data were grouped with Tyson polygons. A new comprehensive drought index Comprehensive Index (CI) was constructed by Caussian Copula function. Mann-Kendall trend test was used to test the trend of CI annual series, and the run-length theory was used to extract drought features. It is found that CI can respond well to both hydrological and meteorological droughts. The number of M-K trend series of CI sequence is -3.78. There is a trend towards drought in the upper reaches of the Fenhe River. The intensity of drought in 1970 is the smallest, which is 0.06. Four years from 2013 to 2016 appeared continuous drought, and the maximum intensity is 2.66.