Dry-wet cycle effect is an important factor affecting slope stability in red mudstone reservoir area. Triaxial consolidation undrained shear experiments, CT scans and high precision scanning electron microscopy experiments were carried out to investigate the mechanical and fracture evolution characteristics of in situ weathered mudstones during dry and wet cycles. The results show that the stress-strain relationship of red-bed mudstone presents strain softening characteristics at low confining pressures and strain hardening characteristics at high confining pressures. The shear strength parameters of red-bed mudstone decrease gradually during the dry-wet cycle. The volume content of cracks decreases exponentially with the increase of the time of dry-wet cycles, and has a negative linear correlation with cohesion function. The microstructure of mudstone gradually breaks down in the process of dry-wet cycles, leading to the connectivity of cracks and the attenuation of mechanical properties.

To prompt grouting construction quality management of Wuyue Pumped Storage Power Station, combining with exposed geological conditions after excavation, designed grouting parameters, this paper conducts grouting test, and determines the grouting process parameters. The introduction of intelligent grouting system can reduce labor intensity of workers and the labor cost, and improve the traceability of quality work, save the construction period. The introduction of third-party testing units brings advanced technology and management experience, and improves the credibility of testing results. Using the results of water pressure test and comparative analysis of wave velocities of different types of rock masses can evaluate the grouting quality. The evaluation results are mutually verified, which improves the persuasiveness of the grouting effect. The grouting project has achieved satisfactory results for all parties involved in the construction, and this grouting construction process management experience can be a reference for similar projects.

In view of the complex and changeable construction site and high difficulty in construction of water conservancy safety, the risk assessment of all hazard sources is the premise of safety control of a project. The traditional LEC method has the defects of large influence of subjective factors and unclear weight of each hazard source. Based on the introduction of expert credibility, the background differences between risk evaluation subjects were converted to quantitative analysis, and the subjective impairments was modified. This paper combined fuzzy comprehensive evaluation method with traditional LEC method to establish the risk evaluation index system. By calculating every index weight factor with the modified risk values in LEC method, the risk assessment was converted from qualitative to quantitative by membership degree matrix and the defection of imprecise weight factor in LEC method was solved. Finally, a verification of Xinmeng River shows that the objectivity of risk assessment of hazard sources has been improved significantly through the modified method, which provides scientific basis for the risk hierarchical control of water conservancy safety.

The study of classified water price can provide scientific basis for the rational formulation of agricultural water price, which is of great significance for guaranteeing the benign operation of irrigated areas and promoting the high quality development of regional agriculture. Taking grain and oil crops as the research object, two evaluation indexes of irrigation benefit apportionment coefficient and irrigation water output were selected to define the irrigation benefit apportionment coefficient by emergy theory. The agricultural water price share evaluation system of different crops was constructed based on entropy weight method, and the irrigation classified water price evaluation model was established and applied in Baojixia irrigation district of Xianyang City. The results show that the share coefficient of irrigation benefit was 0.42, the share of agricultural water price of grain and oil crops was 0.37, and that of cash crops was 0.63. For agricultural water supply at full cost, the agricultural water price of grain and oil crops was 0.485 yuan/m³, and that of cash crops was 0.827 yuan/m³. As for the water price of agricultural water supply operation cost, the agricultural water price of grain and oil crops was 0.377 yuan/m³, and that of cash crops was 0.641 yuan/m³.

At present, the TV card model of PSD-PBA platform used in power system stability calculation cannot reflect the difference of units under multiple water heads, which is specifically reflected its dynamic response speed and anti-regulation accuracy. Based on the TV card model, an improved scheme is proposed. The water head-opening-power characteristics and power T_w characteristics of the model are obtained by program processing with the early operation data of the hydropower station, and then added to the improved model successively. The results show that the dynamic characteristics of the TV card model do not have the conditions for multi-parameter identification under the power mode primary frequency modulation. Moreover, the accuracy of anti-regulation under different loads is poor. The response characteristics of the improved model accord with the operation law of the real machine and can accommodate the difference of multiple water heads of the unit.

At present, most variable-speed pumped-storage units use converters (AC excitation) to control power. There is a problem of reverse speed overshoot caused by rapid power changes, and the traditional governors are difficult to ensure good control effects, which affects the operating efficiency of variable-speed units. In order to study the control of the speed regulation system under the active power regulation of the variable speed unit, the speed regulation system of the variable speed pumped storage unit with adaptive backstepping sliding mode control (ABSMC) was proposed. Firstly, a mathematical model of the variable-speed pumped-storage unit speed control system including the optimal speed module was established. Secondly, taking the electromagnetic power as the system disturbance, an adaptive backstepping sliding mode controller for the variable-speed pumped-storage unit speed control system was designed and its stability was analyzed. Finally, a simulation test platform for the speed regulation system of the variable speed pumped storage unit was built, and the dynamic response simulation analysis of the speed of the variable speed pumped storage unit was carried out under the given power of the power generation condition. Compared with the sliding mode controller (SMC) and PID controller, the results show that the reverse overshoot of the controlled unit speed by the ABSMC is small, the adjustment time is short, and the effect of tracking the optimal speed is ideal. The effectiveness of the designed controller was verified.

The primary frequency modulation function of the hydropower unit plays an increasingly important role in maintaining the safe and stable operation of the power grid, and it is necessary to ensure that the units integrated into the power grid have good primary frequency modulation performance. The influence of various dead zone nonlinear factors in the hydraulic turbine adjustment system on the dynamic performance of primary frequency modulation cannot be ignored. The nonlinearity of the dead zone in all links of the adjustment system is comprehensively analyzed. Based on the established high-precision primary frequency modulation model of hydropower units, the influence of frequency dead zone, openness dead zone, power dead zone and dead zone of the follow-up system on the performance of the primary frequency modulation response performance of large and small frequency difference is explored, and it is emphasized that the next frequency modulation inaction of small disturbances may be related to the dead zone of the follow-up system, which provides important theoretical guidance for the improvement and optimization of the primary frequency modulation function of the hydropower unit.

The load rejection process is one of the most dangerous transients in pumped storage hydropower plants. The current design criteria of plant guarantee that the lowest pressure in draft-tube inlet is higher than the saturation pressure during transients, but there is still local cavitation inside the pump-turbines. The one-dimensional pipeline and three-dimensional pump-turbine coupled computational fluid dynamics simulation method was used to simulate the simultaneous load rejection process in a pumped-storage hydropower plant. The results show that a spiral cavitation cavity is in the center of the draft-tube inlet, and five wedge-shaped cavitation cavities are in the outlets of runner channels. The collapse of the spiral cavitation cavity in the draft-tube leads to instantaneous pulse impacts on the pressure and runner forces. There is no obvious impact when the wedge cavities collapse.

In the hydraulic model test of the CB hydroelectric station in Jinsha River, occasional penetrating vortexes were observed in front of the sluice gate. In order to eliminate the vortex, this paper improved a vertical eddy-eliminating barries and proposed a new horizontal eddy-eliminating barries. The flow field in front of the sluice gate, vortex scale, and vortex generation frequency of the new horizontal eddy-eliminating barries were measured and analyzed for its vortex elimination effect. The results show that the proposed horizontal eddy-eliminating barries has better vortex elimination effect than the previous vertical eddy-eliminating barries, and it is a low-cost and efficient vortex elimination facility for sluice gate.

In order to address the limitations of the existing vibration trend prediction model for hydroelectric units, a vibration trend prediction method for hydroelectric units based on optimal variational mode decomposition (OVMD), time-varying filter empirical mode decomposition (TVFEMD), hunter-prey optimization algorithm (HPO), and extreme learning machine (ELM) is proposed. This method first applies OVMD to adaptively decompose the original vibration signal of the hydroelectric unit, and then further employs TVFEMD to perform a secondary decomposition of the residuals obtained from the first decomposition. Subsequently, vibration trend prediction models HPO-ELM are established for each subsequence. The final predicted vibration signal is obtained by aggregating and reconstructing the prediction results of all the sub-sequences. The research results demonstrate that this method outperforms traditional methods in terms of prediction accuracy for the vibration trend of hydroelectric units, and it has good engineering application value.