Data-intelligent methods designed for forecasting the streamflow of the Fenhe River are crucial for enhancing water resource management. Herein, the gated recurrent unit (GRU) is coupled with the optimization algorithm improved grey wolf optimizer (IGWO) to design a hybrid model (IGWO-GRU) to carry out streamflow forecasting. Two types of predictive structure-based models (sequential IGWO-GRU and monthly IGWO-GRU) are compared with other models, such as the single least-squares support vector machine (LSSVM) and single extreme learning machine (ELM) models. These models incorporate the historical streamflow series as inputs of the model to forecast the future streamflow with data from January 1956 to December 2016 at the Shangjingyou station and from January 1958 to December 2016 at the Fenhe reservoir station. The IGWO-GRU model exhibited a strong ability for mapping in streamflow series when the parameters were carefully tuned. The monthly predictive structure can effectively extract the instinctive hydrological information that is more easily learned by the predictive model than the traditional sequential predictive structure. The monthly IGWO-GRU model was found to be a better forecasting tool, with an average qualification rate of 91.66% in two stations. It also showed good performance in absolute error and peak flow forecasting.

To quantify the flood risks in cascade dam systems, it is critical to analyze the risk factors and potential breaking failure paths. In this study, Bayesian networks (BNs) were applied to create a flood risk analysis model for a cascade dam system. Expert experiment, historical data, and computational formulas were employed to estimate the prior probability and original conditional probability tables (CPTs) in the BN model; sensitivity analysis was used to ensure the original continuous breaking failure path in the system. To avoid the possible misperceptions of the probability of a certain event, Dam Breach Analysis Model (DB-IWHR) 2014 software and the flood regulation method were used to simulate the dam breaking progress. The posteriori continuous breaking failure paths were obtained, and then the original CPTs were refined based on the new evidence. The proposed method was applied to the Bala-Busigou-Shuangjiangkou (BL-BSG-SJK), which is located upstream of the Dadu River basin in China. The results show that three continuous breaking failure paths could be identified in the researched cascade dam system. A new BN model was created to determine the failure probability of the cascade dam system under the three continuous breaking failure paths. This analytical method may also be useful for other similar cases.

Reservoir-operation optimisation is a crucial aspect of water-resource development and sustainable water process management. This study addresses bi-objective optimisation problems by proposing a novel crossover evolution operator, known as the hybrid simulated binary and improved arithmetic crossover (SBAX) operator, based on the simulated binary cross (SBX) and arithmetic crossover operators, and applies it to the Non-dominated Sorting Genetic Algorithms-II (NSGA-II) algorithm to improve the algorithm. In particular, the arithmetic crossover operator can obtain an optimal solution more precisely within the solution space, whereas the SBX operator can explore a broader range of potential high-quality solutions. Considering the advantages of both operators, this study introduces an improved arithmetic operator to reduce the risk of local convergence inherent in conventional arithmetic operators. Subsequently, two strategies for the SBAX operator are discussed: SBX operator + new arithmetic operator and new arithmetic operator + SBX operator. The convergence of the bi-objective Pareto solution set is evaluated based on the generation and inverted generational distances. This method is used for the collaborative optimisation of the water supply and ecological operation of the Fenhe Reservoir, where its effectiveness is demonstrated. A comparative analysis of the bi-objective optimisation schemes obtained using different crossover operators indicates the following: (1) the NSGA-II algorithm based on the SBAX operator achieves a convergence efficiency that is 14.25–41.95% higher than that of the conventional NSGA-II algorithm; (2) the reservoir operation indices of the scheduling scheme derived from the NSGA-II algorithm based on the SBAX operator significantly outperform those obtained using the conventional NSGA-II algorithm. The optimal strategy reduces the annual average water abandonment by 11.2–14.52 million m³. This study provides a novel approach for bi-objective optimisation and sustainable reservoir management.

The impacts of temperature and precipitation changes on regional evaporation and runoff characteristics have been investigated for the Biliu River basin, which is located in Liaoning Province, northeast China. Multiple climate change scenarios from phase 3 and phase 5 of the Coupled Model Intercomparison Project (CMIP3 and CMIP5) (21 scenarios in total) were utilized. A calibrated hydrologic model—SWAT model—was used to simulate future discharges based on downscaled climate data through a validated morphing method. Results show that both annual temperature and precipitation increase under most of the CMIP3 and CMIP5 scenarios, and increase more in the far future (2041–2065) than in the near future (2016–2040). These changes in precipitation and temperature lead to an increase in evaporation under 19 scenarios and a decrease in runoff under two-thirds of the selected scenarios. Compared to CMIP3, CMIP5 scenarios show higher temperature and wider ranges of changes in precipitation and runoff. The results provide important information on the impacts of global climate change on water resources availability in the Biliu River basin, which is beneficial for the planning and management of water resources in this region.