Vegetation is an important part of terrestrial ecosystem and is vulnerable to extreme climate events. It is of great significance to explore the spatio-temporal evolution and dynamic response relationship between extreme climate events and vegetation for scientific response to climate anomalies and prevention of natural disasters. Therefore, the RClimDex, trend analysis and detrended fluctuation analysis were used to analyze the spatio-temporal evolution of extreme climate and NDVI in the Huaihe River Basin from 1998 to 2019, and discuss their causes and future variation tendency as well as clarify the response of NDVI to extreme climate. The results demonstrate that the NDVI shows an increasing tendency with an increase of 0.002 5 a^-1 from 1998 to 2019. Among the extreme climate indices, warm temperature indices showed an increasing tendency, while the cold temperature indices and precipitation indices (except RX1day) showed a decreasing tendency. The distribution of NDVI in the Huaihe River Basin was closely related to the distribution of urbanization in the region. The NDVI has a lag response of about one month to extreme climate and will continue to rise. The annual variation of the NDVI is closely related to natural and human factors, and the intra-annual variation of the NDVI is mainly affected by the growth rhythm of crops in the cultivated land.

Taking four small and medium-sized watersheds in the mountainous city (Chongqing) as the research objects, the application prospects of multi-source rainfall data in short-term hydrological forecasting were explored. The GPM-IMERG remote sensing precipitation data was introduced as the source of the satellite precipitation data, and the Mean Bias Correction (MBC) method, Linear Regression Model (LRM) and Geographically Weighted Regression (GWR) method were used to realize the fusion with the ground station measured data. A leave-one-out validation method was used to evaluate the accuracy of the fused precipitation data, and the SWAT model was applied to simulate the runoff of the selected basins using the precipitation data corrected by MBC, LRM, GWR, the original IMERG-early, and the measured data. The results show that the GWR was the best fused effect. The topography had great influence on the accuracy of the IMERG-early data and the fusion method. The accuracy of the fusion method was relatively high in low elevation regions, while in mountainous areas, the accuracy became low comparatively. The results showed a better agreement between the SWAT model and the measured runoff process using all kinds of precipitation data, in which the model using precipitation data corrected by GWR method was more accurate. Results can be referenced for hydrological forecast in similar small basins in the future.

Water footprint evaluation is of guiding significance for the rational development of regional water resources. The input-output method was used to account for the water footprint of Hubei Province from 2007 to 2017, and the driving factors of water footprint changes were analyzed using the structural decomposition model. The results show that the virtual water content in Hubei Province decreased significantly, and water use efficiency improved from 2007 to 2017. The total water footprint showed a trend of rising and then falling, with a net increase of 19.9%, and the main growth sectors were construction and services. Technology level and economic scale were the main factors inhibiting and promoting the increase of water footprint, respectively. However, their effects on water footprint changes gradually weakened, and both showed a reduction effect. From 2007 to 2017, the sectoral linkages changed from positive to negative driving force, and the inhibitory effect on water footprint was revealed, which shows that optimizing the industrial production process helps carry out water conservation. The impact of driving factor on different sectors showed heterogeneity. The adjustment of industrial structure inhibited the increase of water footprint in high water-consuming sectors such as agriculture while promoting the increase in construction and services. In the future, the industry scale and residential demand in sectors producing high-value-added products will grow, such as services, and water-saving technology development should be shifted to these sectors promptly.

The Duncan-Zhang model is not able to reflect the influence of the mid-major stress on the strength and deformation of the rockfill body in the finite element calculations of the panel rockfill dam. The Duncan-Zhang E-B model and its optimization model for the panel rockfill dam has been developed by the secondary development of ANSYS based APDL language. The Duncan-Zhang E-B model and its optimization model were used to calculate the three-dimensional finite element static force of a panel rockfill dam in Yunnan Province. The results show that the settlement of the dam body was reduced, the horizontal displacement was reduced, and the stress change was small. The strain results obtained from the optimization model are closer to the actual data. The optimizaion Duncan-Zhang E-B model considering the medium principal stress can better reflect the stress-deformation characteristics of the panel rockfill dam.

Aiming at the problem that the convective heat transfer coefficient depends on experience in the calculation process of arrester temperature rise, a method of determining convective heat transfer coefficient was proposed. The temperature field distribution inside the arrester was studied. The influences of pollution, wind speed and sunshine on the temperature rise of the arrester at different damp degree and different damp positions were analyzed. The results show that under different influencing factors, the convective heat transfer coefficient of the arrester ranges from 5 to 31, 27 to 35 and 19 to 34 W/(m²·℃), respectively. The pollution location will affect the temperature rise distribution of MOA surface, and the greater the degree of moisture, the more significant the impact on temperature rise. With the increase of wind speed, the temperature on the valve plate decreases slightly. However, the surface temperature decreases greatly. The influence of sunshine on the temperature rise of lightning arrester is obviously increased, but it is smaller than that of pollution and wind speed.

The escape of oil mist from bearing oil groove of power station will not only pollute the environment of power station, but also bring safety hazard to electrical equipment. Combined with the actual situation of hydropower station, the installation of oil mist absorption device can help to discharge excess oil mist from oil tank and effectively avoid the pollution problem of oil mist in power station. In order to investigate the variation of oil mist distribution, velocity, pressure at measuring points and discharge amount of oil mist in the oil tank, numerical simulation calculation of thrust bearing oil tank with negative pressure oil mist pipe was carried out. It is found that after complete oil mist discharged, most of the oil mist can be effectively discharged; The overall change of the amount of oil mist at the outlet follows the law of rising first and then falling. Under the suction pressure of -500 Pa, the maximum discharge amount of oil mist could reach 0.008 23 kg/s. After the oil mist is discharged, the pressure drop at the oil tank upper region reaches 96.07%, and the reduction of the pressure difference can effectively prevent the oil mist from escaping to the cover plate, the wind tunnel, the rotor and so on. The research can provide some theoretical help for the oil mist discharge process of oil tank so as to better formulate the oil mist control strategy.

Aiming at the shortcomings that traditional water weirs cannot measure the sand content of runoff and the traditional measurement methods are not accurate, it is difficult to realize online measurement in the field of dam water weirs. By analyzing the negative correlation between the sand content of the water flow of the water weir and the gray value of the image, sand volume measurement technology was proposed to automatically measure the sand content of the water flow of the weir. The test results show that the deviation between the detection result of sand content and the actual value is low, and it is a convenient, efficient and real-time determination method. The research results have certain reference value for dam seepage monitoring.

In order to understand whether the power frequency electric field level generated by the UHV AC power transmission and transformation project meets the relevant requirements, this paper uses the EFA-300 power frequency field strength meter to conduct the cross-section of the substation and transmission line of the Ximeng-Shengli 1 000 kV AC power transmission and transformation project, and evaluate the uncertainty of the measurement results. The results show that the maximum measured value of the substation boundary is (3.65±0.18) kV/m, k=2, which is lower than the electric field limit of the substation boundary, 4 kV/m; the maximum measured value of the line cross-section is (5.41±0.28) kV/m, k=2, which is lower than the electric field limit below the overhead transmission line, 10 kV/m, which meets the requirements of relevant national standards. It is found that the error of the field strength meter has a great influence. The uncertainty of the power frequency electric field measurement can be reduced by improving the calibration accuracy of the power frequency field strength meter. This work is supported by the Science and Technology Project of the Headquarters of State Grid Corporation of China-Research on Power Frequency Electric Field Calibration Technology (GYW1720200201).

Prediction of the vibration trend of hydropower units is an important measure to ensure the normal operation of the unit. However, due to the complexity and non-stationarity of the vibration signal of the unit, accurate prediction becomes a difficult problem. To this end, this paper proposes a combined trend prediction model based on adaptive variational modal decomposition and temporal convolutional network (TCN). Firstly, the Whale Swarm Algorithm (WOA) was used to optimize the parameters of Variational Mode Decomposition (VMD) to avoid the drawbacks of blindly selecting parameters, and to achieve adaptive decomposition of vibration signals. And then each decomposed component signal was normalized to establish TCN for trend prediction. Finally the original vibration signal prediction was obtained by superimposing the results. The proposed model was demonstrated and tested with the actual operation data of a domestic power station. The results show that the proposed model has high prediction accuracy and can be used in engineering practice.

The emergency shutdown device of the hydro-generator governor should not only meet the requirements of "power loss action" in the 25 anti-accident measures, but also avoid the hidden danger of unplanned outage of the unit caused by the electromagnetic valve coil of the device being charged for a long time and easy to burn. This paper introduces three generations of technical improvement of the emergency shutdown device of hydro-generator governor of a large hydropower station, from double electromagnet two-position solenoid valve, single electromagnet spring return two-position solenoid valve to double solenoid valve and linkage valve, expounds their working principle and compares their respective advantages and disadvantages. It is proved that the emergency stop device of hydro-generator governor based on double solenoid valve and linkage valve is a practical and reliable emergency stop control optimization scheme of speed regulating system, which has great popularization value in hydropower plants.