Aiming at the shortcomings of slow convergence and easily falling into the local minimum in the learning process for the traditional BP neural network, particle swarm algorithm with fast convergence speed and strong global optimization ability was introduced so as to establish the PSO-BP model. Taking the seepage monitoring data of an earth and rockfill dam as an example, the seepage was predicted. Compared with the prediction model, the BP model and the traditional statistical regression model, the results show that the PSO-BP model has a higher goodness-of-fit and convergence.

To improve the added value utilization rate of reservoir sediment in the field of construction materials, the reservoir sediment was used as auxiliary cementitious material to prepare pervious concrete. The results show that with the increase of reservoir sediment content, the compressive strength of the specimens will decrease, and the permeability will increase. With the extension of mechanical activation time, the compressive strength of the specimen increases and the permeability decreases. When the water cement ratio is 0.3, the mechanical activation time is 45 min, the 28 d compressive strength of the prepared pervious concrete can reach 21.44 MPa, and the permeation coefficient is 0.56 mm/s. The results of this study can provide reference for the resource utilization of reservoir sediment in the field of building materials.

With the transformation of the southwest power grid with a high proportion of hydropower into an UHV AC-DC network structure, ultra-low frequency oscillations have occurred in the power grid. According to the characteristics of the hydropower unit governor, a small disturbance model of the hydropower unit was established. The impact of various parameters of the hydropower unit governor on the system stability in primary frequency modulation was discussed by using the eigenvalue analysis method, and the power mode and the opening mode were compared. Finally, the impact of the opening mode and the power mode with AGC was studied. The results show that the parameter adjustable range of the opening mode is larger than that of the power mode, but it is more easily affected by AGC. This study lays a foundation for further research on suppression methods of ultra-low frequency oscillations.

In order to study the mechanical characteristics such as fracture mode, crack propagation characteristics, and strength characteristics of intersecting multi-fractured rock masses in uniaxial compression tests, the particle flow program was used to simulate the mechanical characteristics of intersecting multi-fractured rock masses in uniaxial compression tests. The impact of the angle between intersecting cracks on the strength characteristics, failure mode, microcrack development and evolution law of multi-fractured samples during uniaxial compression was analyzed. The experimental results show that the failure strength of the cross multi-crack specimen exhibits a slow bottom "U" shaped characteristic with the increase of crack angle; The failure mode of intersecting multi-crack specimens is mainly tensile shear failure, with some specimens exhibiting shear failure; The failure process of intersecting multi-crack specimens first involves the generation and intersection of cracks between the tips inside the cracks, followed by the generation of cracks at the outer tips of the cracks and their transmission parallel to the loading direction to both ends of the rock mass, ultimately leading to macroscopic failure of the rock mass. The results of this study contribute to the understanding of the mechanical mechanism of cross fissured rock masses in compression failure, and are a supplement to the research on the mechanical properties of cross fissured rock masses.

To effectively judge the vibrational stationary state of hydropower units, the whirling characteristics of the rotor vibration and the evaluation method of stability of shaft vibration were studied. The theoretical expression of instantaneous whirling velocity was derived, and the characteristic parameters of whirling velocity were analyzed. A holographic whirling velocity analysis method for shaft vibration was proposed. The whirling characteristics indicators and vibration stability evaluation index of shaft vibration were established. Finally, the theory of this paper was applied to the shaft vibration data analysis of Zhejiang Xianju pumped storage power station. The key characteristic indicators of vibration were extracted and the shafting stationary was evaluated. The results show that the established characteristic indicators have good performance in evaluating the stationary of shaft vibration, and have broad application prospects for the monitoring and performance evaluation of actual hydropower units.

Aiming at the problem that the background harmonic voltage fluctuation and the abnormal harmonic measurement samples significantly affect the accuracy of harmonic impedance estimation, a method for calculating utility harmonic impedance based on generalized jackknife least square estimation and data screening is proposed. According to the principle that the assumption of the fluctuation method and least squares regression are the same, the measurement period when the background harmonic voltage fluctuation is stable is selected for the regression calculation, which can restrain the influence of background harmonic voltage fluctuation on impedance estimation. According to the jackknife principle, the harmonic measurement data is resampled, and the generalized least square method is combined to effectively suppress the adverse effects of abnormal measurement data on the regression. The accuracy and practicability of the proposed method are verified by simulation and actual measured data test.

In order to explore the impact of moisture change on the pore structure of reshaped loess, different times of humidification-dehumidification cycle tests were carried out for remodeled loess with different dry densities, and the pore structure inside the soil under different cycles was tested by nuclear magnetic resonance technology. The results show that the NMR T₂ pattern of reshaped loess presents a bimodal structure, that is, the sample contains two pore structures; The dry and wet cycles have an impact on the change of the pore structure of the loess, and the pore structure shows different laws under different times of dry and wet cycles. Finally, based on the quantitative parameters extracted from the T₂ pattern, the permeability of reshaped loess under different cycles were calculated by SDR permeability model, and the permeability mechanism of loess was explained from the microscopic aspect.

In order to improve the prediction accuracy of vibration signals of hydropower units, a prediction method based on variational mode decomposition (VMD), sample entropy (SE) reconstruction and particle swarm optimization (PSO) optimization of bi-directional gated circulation unit (BIGRU) was proposed. In the VMD-SE-PSO-BIGRU model, firstly, the vibration signal was decomposed into several subsequences by VMD, and the subsequences were reconstructed by SE. The trend, oscillation and noise components of the vibration signal were obtained. Then, parameter optimized BIGRU prediction models were established for the reconstructed components. Finally, the prediction results of each component were superimposed to achieve vibration prediction. Compared with other models, example analysis shows that the prediction error of the model is smaller and the prediction accuracy is higher, which can effectively predict the vibration signals of hydropower units.

In order to explore the causes of urban waterlogging, taking Guangzhou as an example, a drainage model was established based on SWMM. A waterlogging cause analysis method based on real-time monitoring data was proposed. Combining with the real-time monitoring data, the water-collecting capacity of water-logging points, the flow capacity of pipeline and the runoff generated by rainstorm were calculated. At the same time, the corresponding waterlogging measures were given according to different waterlogging causes. The results show that 75% of waterlogging cases involved insufficient surface drainage capacity, 57.7% of waterlogging cases involved watercourse top support, and 44.2% of waterlogging cases involved excessive rainfall intensity and insufficient pipe drainage capacity.

The design of culverts in hydraulic engineering involves many common and repetitive tasks. In order to achieve the parametric application of BIM design in culvert engineering, this study has carried out the research on the parametric modelling method of culvert engineering based on Revit platform, and developed a parametric modelling system for culvert through secondary development of Revit according to the relevant technical specifications. By inputting specified parameters, the system can complete the model building, calculation, and drawing, which greatly improves the efficiency of culvert engineering design. The research results achieve a high standard of engineering parametric design, which can provide technical reference for BIM parametric design of similar projects.