Font Size: a A A

The Mid-Long Term Forecast And Operation Of Reservoir Based On Extreme-point Symmetric Mode Decomposition

Posted on:2019-01-09Degree:MasterType:Thesis
Country:ChinaCandidate:Z P DuanFull Text:PDF
GTID:2370330548470832Subject:Hydrology and water resources
Abstract/Summary:PDF Full Text Request
Hydrological forecasting and hydropower station reservoir operation are the core contents of water resources operation and management.With the increasingly scarce water resources in recent years,the impact of runoff analysis and forecasting on the joint dispatch of cascade reservoirs has also increased,higher-precision forecasts can give people an idea of the future of hydrological changes in order to take early and co-ordinated arrangements to deal with such issues as flood control,drought control,power generation,water supply and shipping,it is directly related to the optimal allocation of water resources and rational utilization of river basin to maximize the overall benefits.In recent years,the Extreme-point Symmetric Mode Decomposition method,which is good at finding the trend of change from several years of observation sequence,has great advantages in the field of data analysis.Firstly,based on the runoff time series of Tangnaihai station in the Yellow River,the multi-scale characteristics of precipitation changes and the abrupt changes at different time scales were analyzed by using extreme-point symmetric mode decomposition method,and the future trend of precipitation was judged.Secondly,this paper establishes ESMD-BP neural network combination forecasting model by using the advantages of the pole-symmetric modal decomposition method in data analysis and combining with the nonlinear approximation of BP neural network.The example application shows that the combined forecasting model can make full use of data information and improve forecasting accuracy.Finally,taking the Longyangxia and Liujiaxia reservoir group as an example,this paper establishes a joint optimal operation model with the goal of maximizing power generation based on the results of the ESMD-BP combined model prediction.The model is solved using the Evolver Genetic Optimization application,and the forecast scheduling results are compared with the results of the existing operation mode,providing a reference for future hydropower station power generation planning.
Keywords/Search Tags:modal decomposition, time series, runoff analysis, combined forecast, forecasting operations
PDF Full Text Request
Related items