Risk Analysis And Multi-attribute Risk Decision Of Combined Flood Control Operation Of Reservoirs

As we all know,the temporal and spatial distribution of water resources in China is extremely uneven.Flood disaster is a serious threat to the affected area of people's life and property safety and restricting the sustainable development and the economic and social modernization process.With the aggravation of global climate change,the further manifested spatiotemporal variability of water resources,and the frequent extreme climate and hydrological events,hydrological uncertainty has become one of the key factors affecting the cascade reservoirs operation and flood risk management and control system.Therefore,studying of the main hydrological uncertainty factors which affect the joint operation flood control risk of cascade reservoirs,analysing of reservoir water level risk and flood diversion risk in joint operation of reservoirs in the upper reaches of the Yangtze River quantificationally,establishing a risk decision model for the joint scheduling of cascade reservoirs,providing decision-making basis for the decision-makers about the optimization of the scheduling scheme scientifically reasonably,are of great significance.Combining the research tasks undertaken by our laboratory,in this paper,we select the flood forecasting error,wind wave effect,water level reservoir capacity and water level discharge capacity as the risk factors for flood control operation of the reservoir.And on the basis of existing work,we launch the risk analysis of flood control operation under the influence of multiple risk uncertainties about joint operation of Three Gorges Reservoir with Xiluodu and Xiangjiaba reservoir in the upper reaches of the Yangtze River.In addition,for the risk decision-making problem of flood control scheduling scheme,a decision method based on grey correlation degree and Euclidean distance ideal solution is introduced to construct a multi-attribute decision-making model for flood-control scheduling of cascaded reservoirs,and the optimization of flood-control scheduling plan set is realized.The research results can provide a theoretical basis for the safe utilization of regional flood resources,and provide technical support for risk analysis and risk decision-making of cascade reservoirs flood control operation.The main contents and conclusions are as follows:(1)The main contents of the risk analysis of reservoir flood control are determined.Based on Monte Carlo simulation technology,the risk estimation process of reservoir flood control scheduling is drawn up,and the LHS method is used to improve the sampling performance.(2)This paper takes Xiluodu,Xiangjiaba and Three Gorges Reservoir as well as Li Zhuang,Zhu Tuo and Cun Tan hydrological stations corresponding to downstream flood protection points as the research objects in the upper reaches of the Yangtze River,and has analyzed the probability distribution of 5 uncertain factors which affect the risk of flood control dispatching,including the flood forecast error,wind wave effect,water level reservoir capacity and water level discharge capacity.It focuses on the analysis of the probability distribution of flood forecasting errors through using 10 distribution functions,such as normal distribution,exponential distribution,P-III distribution,generalized extreme value(GEV)distribution,Weibull distribution,Gumbel distribution and Wakeby distribution.And the distribution form of the probability distribution of flood forecasting error which is most suitable for each station is selected through comparison of goodness of fit test.The results show that the fitting effect of the weickby distribution is the best for the flood forecast error of the Li Zhuang station,the Cun Tan Station and the Yi Chang Station,and the GLO distribution of the flood forecast error of the Zhu Tuo station is the best.(3)In view of the risk assessment of cascade reservoirs in the upper reaches of the Yangtze River,based on the Monte-Carlo stochastic simulation technology and the probability distribution of five uncertain risk factors,Getting uncertainties sampling random sequence by LHS,the risk assessment model of the cascade reservoir group combined flood control scheduling is established.Based on the flood control simulation of design flood after different typical years,the risk of combined flood control of cascade reservoirs with different risk uncertainty factors is obtained.(4)A multiple attribute decision-making method based on grey correlation degree and Euclidean distance ideal solution is introduced to build a multi-attribute risk decision-making model for flood control scheduling.And the decision making optimization is carried out with the example of the joint operation of Three Gorges Reservoir with Xiluodu and Xiangjiaba reservoir in the upper reaches of the Yangtze River.The results are compared with those of the traditional multiple attribute decision making method and it is proved that the method of multiple attribute decision making based on grey correlation degree and Euclidean distance ideal solution is scientific and reasonable.