In order to study the effects of plants on the migration and accumulation of nitrogen in the biological retention system, and promote the vegetation construction of rainwater biological retention system in loess distribution area, 5 groups of biological retention system simulation devices were constructed by selecting Iris, Hemerocallis, Sedum, Ophiopogon and non-plants. The migration of nitrogen in fillers and the accumulation of nitrogen in each medium under different plant treatments were investigated. The results show that there was a close relationship between plant nitrogen accumulation and plant biomass increase during the experiment, which was as follows: Sedum> Iris>Hemerocallis> Ophiopogon. The nitrogen absorbed by plants was mainly accumulated in the upper structure of plants. The distribution difference of in the five groups of biological retention system fillers is not significant and the content of is low. is mainly removed in the upper part of the packing layer by adsorption, and each system has a good removal effect on . Plants have a significant effect on the distribution of in the packing and the migration of in the system. Compared with the non-plant group, the system with plants can reduce content in the filler, and the lower the content in the filler before influent, the lower the effluent concentration. The average TN removal rates of 5 treatments were as follows: Iris (50.59%)> Sedum (43.99%)> Hemerocallis (37.85%) Ophiopogon > (30.51%)> non-plant (26.09%).

It is of great significance to discuss the relationship between ecosystem services and matching status of water and land resources for the sustainable utilization of ecological resources. Therefore, the equivalence factor method, matching coefficient of water and land resources, and coupling coordination degree model were used to reveal the spatiotemporal evolution characteristics of the ecosystem service value, matching status of water and land resources, as well as the coupled coordination relationship between them in China. The results show that the total value of ecosystem services in China increased 2144.45 billion yuan. The overall spatial pattern of the ecosystem service value per unit area was high in the south China and low in the north China, and the ecosystem service value per unit area in all provinces except Shanghai was improved during the study period. The matching coefficient of water and land resources in China continued to decrease, showing a clockwise spatial variation characteristic of first decreasing and then increasing roughly with Qinghai province as the endpoint in space. The coupling coordination degree between ecosystem services and matching of water and land resources in China from 2010 to 2018 first increased and then decreased, and the overall belonged to the basic coordination type, which was in transition. The coupling coordination degree shows the spatial distribution characteristics of high in the south Chin and low in the north China. The results can provide decision-making basis for the policy formulation of coordinating ecological protection and water and land resources utilization.

Taking the upper reaches of the Yangtze River at Guangyang Dam as an example, a physical model was constructed to analyze the changes of water depth and flow field in the corresponding river section before and after the implementation of the continuous pool-riffle type natural habitat restoration technology scheme. And then the effects of the habitat restoration scheme on fish spawning grounds, habitats, overwintering activities and waterways were discussed. The study show that after the implementation of the habitat restoration scheme, the area of the slow flow area in the test reach increased significantly, and a local slow flow was formed between the group of spur-dike, which was conducive to fish spawning and habitat; The implementation of the scheme did not adversely affect the vessel traffic in the channel of the river.

In order to scientifically evaluate vulnerability of heavy rainfall and waterlogging in subway Stations, determine the sensitivity index as well as improve the management level of the heavy rainfall and waterlogging in subway stations, an evaluation method based on IOWA-VAC was proposed. Firstly, an evaluation index system was constructed based on Pressure-State-Response (PSR) theory. Then, the index data was reordered in ascending order by using the IOWA operator, and the final weight of the new data was obtained by introducing coefficients θ to dynamically adjust the interval boundary weights. The Vector Angle Cosine (VAC) was introduced to test the closeness of the consistency between the target score vector to be evaluated and the ideal target vector, so as to realize the transformation from algebraic thinking to spatial geometry thinking, and the target grade to be evaluated was obtained. Single factor sensitivity analysis was used to determine the sensitivity index. The practicability of the evaluation model was verified by taking the Yansi station of Zhengzhou Metro Line 5 construction projects as examples. The results show that the vulnerability level of the Yansi station of Zhengzhou Metro Line 5 is Ⅳ, with high risk. Thus, it provides feasible suggestion for vulnerability management of heavy rainfall and waterlogging in subway stations.

In order to reveal the process and mechanism of outbreak of flash floods and causing disasters in ungauged areas, a numerical model of the transport of loose bodies under heavy rainfall floods was established in the Xiangxi River subbasin of Zhangshu City, Jiangxi Province. The process of heavy rainfall and the law of sediment transport were analyzed. The results show that for the river, the flow velocity is higher at the high terrain, slope drop change and the confluence area, and for the area with gentle slope and large catchment area, the increase of flow velocity at the confluence point is especially significant compared with other areas; The local water accumulation at the beginning of the basin, surface runoff occurs in a short period of time, and the gathered water scours out several gullies, and sediment is mostly deposited at the downstream outlet; The water depth and flow velocity of the basin are related to the rainfall. The greater the rainfall, the greater the water depth and flow velocity is, the more significant the scouring situation is. This study has positive significance for flash flood warning in the region.

Aiming at the shortcomings of insufficient population diversity, easy to fall into local optimum, and slow initial solution speed of intelligent water drops algorithm (IWD), an improved IWD algorithm (IIWD) was proposed, which introduces the Logistic chaotic initialization method and the differential variation operator of the difference evolution algorithm (DE), and improves the neuronal update mode of the self-organizing mapping algorithm (SOM) and introduces it into the sub-update process of the optimal individual of the IWD algorithm. The IIWD was applied to the peak shaving scheduling of cascade reservoir groups. Compared with IWD and DE, the efficiency of the IIWD was verified.

Currently, the electricity market is undergoing a new round of reform, and deviation assessment mechanisms have been introduced in various places to promote fair competition. In the face of the new market rules, the traditional dispatching model with maximum generation cannot be applied, and hydropower companies need to take market factors into account to control deviations. In order to save hydropower companies from paying costs, a study was conducted on the deviation control rules, and a bi-level optimization model with the objective of maximizing generation capacity and minimizing deviation was established to facilitate trading decisions. Taking the Three Gorges hydropower plant and Hubei province electricity market as examples, the Harmonic Search algorithm was used for analysis. The results show that the deviation rate in each interval is within the deviation control range, and the hydropower plant is exempt from deviation control. Thus, the bi-level optimization model mentioned above is feasible.

In order to realize the remote mobile monitoring of the dam, the automatic identification of important defects and the measurement of key monitoring equipment, the research on the intelligent mobile inspection technology of dam safety based on image recognition technology and track robot was carried out. Firstly, an intelligent mobile inspection system for dam safety was constructed based on the orbital robot. On this basis, the automatic identification of typical dam defects and the measurement of key monitoring equipment were realized based on image recognition technology. It had been applied in a dam with good results. The remote and large-scale inspection of dam safety was realized, which can significantly reduce the workload of manual inspection, replace manual inspection under extreme natural conditions, and timely grasp the operation of hydraulic structures such as dams. The research results can provide important reference and technical support for the safety monitoring of dams.

Robotic total station plays an important role in long-term, continuous and automatic displacement monitoring of reservoir dams. But the reliability of the monitoring results is easily affected by atmospheric refraction. Based on the theoretical analysis of the meteorological correction model and reference point differential correction model, the applicability of atmospheric refraction correction model was experimentally analyzed by using the automatic monitoring system based on robotic total station established in Wanmipo power station. The results show that for small monitoring areas, only using the meteorological data at the station for correction can achieve good results, and the accuracy of reference point differential correction model is equivalent to that of meteorological correction model. It has reference value for the construction of dam automatic monitoring system based on robotic total station.

In order to explore the influence of geographical location (latitude) differences on the water temperature of the reservoir area, the temperate Gongguoqiao Reservoir and the tropical Jinghong reservoir, which are similar in scale and operation mode, were taken the research objects. Based on the prototype observation data, the CE-QUAL-W2 model was established to compare and analyze the water temperature distribution characteristics of the Gongguoqiao and Jinghong reservoir. The results show that the CE-QUAL-W2 model can be applied to the simulation of temperate and tropical reservoirs, and the difference of water temperature changes in reservoirs in different climatic zones can be clarified. The temperature increase and decrease of the surface water temperature in the Jinghong reservoir area obviously lags behind the surface water temperature in the Gongguoqiao reservoir area. Both Gongguoqiao and Jinghong reservoir are weakly stratified reservoirs. Gongguoqiao Reservoir is weakly stratified in spring and summer, while the stratification in autumn and winter is not obvious. Jinghong Reservoir shows weak stratification in summer and autumn, but not in spring and winter. The results can provide reference for water temperature research of similar-scale power stations on the Lancang River.