Limited by hydrometeorological data, flood forecasting in ungauged basins still faces challenges. Parameter regionalization is a common method to solve this problem. The machine learning model has the characteristics of simple modeling and convenient use compared with the traditional flood forecasting model. Taking the West Plain of Nansihu Lake in Shandong Province as the research area, referencing the idea of hydrological regional synthesis, this paper synthesizes the data of 40 floods in 8 watersheds from 2010 to 2021, and builds a regionalized flood forecasting model based on Long Short-Term Memory (LSTM). The results show that the regionalized flood forecasting model can simulate the actual flood process well, the relative error of flood peak in both the training set and the testing set are less than 10%, and the Nash-Sutcliffe efficiency coefficients are all greater than 0.9; In the 15 h forecast period, the regionalized flood forecasting model has higher forecasting accuracy, and when the forecast period is more than 15 h, the forecast accuracy of the model decreases.

Flexible Mattress is the main beach guarding structure in the middle and lower reaches of the Yangtze River. The water flow scouring is likely to cause the deformation of the flexible mattress, affecting the guarding effect. Aiming at its deformation monitoring difficulties, this paper explores the feasibility of using optical fiber sensing to monitor the deformation of the flexible mattress through the indoor experiment. The results of the study show that at the initial stage of tensile deformation (less than 20 mm), the measured strain value deviates less from the actual value. When the tensile length is greater than 20 mm, the error rate of each measurement point is exponentially increasing, and the fixing effect of the optical fiber cable and the flexible mattress at the fixed point determines the monitoring accuracy of tensile deformation. The positioning accuracy of optical fiber sensing to measure bending deformation is 3 times fixed-points interval. For the concentrated stress areas (such as the edge of scour pits), positive strain is mainly generated. Optical fiber sensing has the feasibility of monitoring the tensile deformation of flexible beach protection structures. When applying this technology, it is necessary to consider the coupling of the sensing fiber and the deformation of the flexible structure, the destruction of sinking the mattress and the complexity of the construction process. The results can be reference for the research and development of monitoring and assessment technology of the in-service condition of the flexible mattress.

From the objectives, variables and relations of the time-varying linear confluence model, it can be seen that the model has some constraints such as static parameters and the inability to consider the influence of interval runoff, and there are significant defects in the application of flood prediction. Therefore, the time-varying linear confluence model is improved by combining the chaotic mapping with the rich model to solve the diversity and adding the influence operator to replace the interval runoff. Using the measured runoff data of Maoergai Hydropower Station in the Heishui River basin for many years, and taking the forecast process, flood volume, flood peak and peak time as the evaluation index, the application analysis of the improved time-varying linear confluence model is carried out. The results show that the overall prediction qualification rate of the improved model is increased by 9.13%, and the certainty coefficient is increased by 0.25, which expands the reliability and practicability of the application of the time-varying linear confluence model.

The friction factor of pipeline is a key parameter in the design calculation, operation scheduling optimization and fault diagnosis of water supply system. In order to determine this parameter accurately, an intelligent back-analysis method of pipe section friction factor based on dynamic search fireworks algorithm (dynFWA) coupled with hydraulic calculation model of pipe network was proposed. The partial derivative relationship between node water pressure and friction factor was taken as the node sensitivity. In the improved genetic algorithm, the maximum sum of node maximum sensitivity was taken as the goal to optimize the layout of monitoring points. Based on the optimized water pressure monitoring value at the monitoring point, the dynFWA algorithm was used to inverse the friction factor of each pipe section with the objective of minimizing the average double error between the water pressure monitoring value and the calculated value. In order to verify the inversion performance of dynFWA algorithm, the inversion of friction factor by dynFWA algorithm and particle swarm optimization (PSO) algorithm were compared. The results show that the maximum relative errors of the inverse value of the friction factor are 17.7% and 0.7% before and after the optimization of the monitoring points, which proves the necessity of the monitoring point selection and the superiority of the improved genetic algorithm for the monitoring point selection. Under the condition that the water pressure at the monitoring node is added to noise, the relative errors of the friction factor inversion results based on the dynFWA algorithm and the PSO algorithm are 9.67% and 14.33% respectively, and the maximum relative errors between the actual water pressure value and the simulated water pressure value at the monitoring point are 0.358% and 0.655%, which proves that the dynFWA algorithm has higher accuracy in the parameter inversion problem compared with the PSO algorithm.

In view of the fact that the reservoirs of medium and small hydropower stations usually operate at a lower level during the flood season because of the smaller regulating reservoir capacity, short decision time for flood regulation, maximum head of the unit and inundation limitation of the upstream reservoir area, etc., the flood resource utilization benefit is not fully utilized during the flood season regardless of the magnitude of the flood. The multi-objective risk analysis model for flood operation and scheduling of medium and small hydropower reservoirs is established with the objective of maximizing the power generation benefit and minimizing the flood risk, and the solution method for the optimal flood level is given when the forecasted incoming flood flow is in different ranges. The results show that when the reservoir faces different levels of floods, the optimal operation level can be obtained by coordinating the benefits and risks to increase the power generation benefits during small floods and reduce the flood losses during large floods, which provides a reference for the flood operation and scheduling of medium and small hydropower reservoirs under changing environment.

The health status assessment of hydraulic turbines is a necessary task for achieving health management of hydraulic turbines, and is a key step in achieving condition based maintenance of hydraulic turbines. Considering the uncertainty and fuzziness of the obtained representation information of the health status of hydraulic turbines, a combination of qualitative and quantitative indicator systems was constructed. The health status of hydraulic turbines was defined as 5 states and transformed into cloud droplets using language scale functions. The evaluation of the health status of hydraulic turbines was achieved through cloud distance. The effectiveness of the health status evaluation model was verified using a certain type of hydraulic turbine as an example, which provides a solid foundation for equipment health management of hydraulic turbine.

The phenomenon of mixed free-surface-pressure flow exists in the process of water level variation in the lower reservoir (roadway group) of mine-type pumped storage power station, which directly affects the operation stability of the system and the hydraulic safety of the roadway group. Based on 3D numerical simulation technology of hydraulic system, considering the three controlling cases including lower reservoir water filling, load rejection and pumping power failure, the characteristics of the transition process of water flow in roadway group and the evolution law of related hydraulic parameters were analyzed, and its influence on the hydraulic safety of roadway group was evaluated. The research shows that there is no obvious unfavorable flow pattern in the initial water filling process of the lower reservoir (roadway group); For load rejection with dead water level of the lower reservoir, the water level of the regulating pool decreases by less than 0.4 m, and for pumping power failure with the normal water level of the lower reservoir, the water level of the regulating pool increases by less than 10.0 m; The roadway group can enter and exhaust normally during the transition process, and the hydraulic transitions in the roadway section is smooth including the possible free-surface-pressure flow; The regulating pool and the ventilation channel meet the requirements of hydraulic optimization to ensure the hydraulic safety of the roadway group.

Timely and accurate forecasting residential water consumption is critical to design and operational management of water supply systems. Long short-term memory (LSTM) is an effective data-driven prediction model for water consumption, but it usually relies on a large number of parameter settings. This paper proposed a multilayer long short-term memory neural network model (MLSTM), which was built on the LSTM model by superimposing a time distribution module. The results indicate that the MLSTM model has lower complexity and higher prediction accuracy than the LSTM model, especially for the prediction of peak water consumption with M_MAPE reduced by about 60%. Meanwhile, the MLSTM model is insignificantly affected by external environmental conditions (e.g., weather).

In response to the problems of low accuracy and poor reliability of water consumption prediction due to the strong randomness and non-stationary state exhibited by the water consumption signal, this paper proposed a hybrid water consumption prediction model based on improved EEMD-WOA-SRU. Firstly, the LSTM prediction method was used to suppress the endpoint effect of the EEMD to obtain the improved intrinsic mode functions (IMF). Then the whale optimization algorithm (WOA) was used to optimize the simple recurrent unit (SRU) and predicted each component. Finally, the final prediction results were obtained by accumulation. The experimental results show that the decomposition error of the EEMD is reduced by 0.94% on average; Compared with the SRU, the average absolute error of EEMD-WOASRU model prediction is reduced by 45.42%, the root mean square error is reduced by 50.43%, and the reliability is improved by 52.38%. It can provide a basis for water resources decision making.

During construction and operation, it is inevitable to have such engineering problems as cutoff wall defects. It is very important to accurately and systematically analyze the influence of defects in anti-seepage wall on seepage and deformation of cofferdam. The finite element model of an anti-seepage wall cofferdam project was established. By using the fluid-solid coupling analysis method, the material parameters of anti-seepage wall element in the finite element model were changed one by one to simulate different defects of anti-seepage wall. The influence of defect location of anti-seepage wall on saturation line, velocity vector field and deformation field of cofferdam was studied. The results show that the bottom defects of the anti-seepage wall have little influence on the saturation line, velocity vector field and deformation field of the cofferdam. With the upward movement of the anti-seepage wall defects, saturation line, pore water pressure and deformation value of the cofferdam behind the anti-seepage wall increase continuously, and the velocity of the cofferdam increases firstly and then tends to be stable. The research results can provide reference for the design and construction of anti-seepage wall and stability analysis of cofferdam.