Research And Application Of The Ecological Scheduling Of The Lijiang Reservoir Group

In recent years,due to the lack of water in the dry season,a series of serious consequences have been caused.Including inadequate water supply in the coastal cities,damage to the river ecology and the impact of tourism navigation,and so on,it has a serious impact on the local economy and ecology.In order to solve the above problems and ensure the ecological flow of the Lijiang River in dry season,it is necessary to build an ecological water replenishment joint dispatching system for the Lijiang reservoir group.The annual runoff and monthly runoff data of the major sections of the Lijiang River are taken as the research objects.First,In order to excavate the periodic evolution law and multiple time scale characteristics implied in runoff series,In order to excavate the periodic evolution and multi time scale changes in the runoff series,the methods of power spectrum analysis,multiple linear regression and mathematical statistics are used.The periodicity,trend and flood frequency of the runoff are analyzed systematically.The runoff and runoff change characteristics of each section were obtained,and the runoff characteristics of Lijiang River were fully grased,so as to provide information support for decision-makers to judge runoff situation.Secondly,the BP neural network model,multiple linear regression model,Elman model and set model are selected as the medium and long term runoff forecasting model on the basis of the analysis of the requirements of the scheduling target and the actual situation.Through the theoretical analysis of the above models,the calculation process research and code compilation,the development and integration of the forecast module are realized.By setting up different calculation steps,the runoff forecasting scheme based on year,month,ten day and day is compiled and calculated.Through the comparison and analysis of the results of multi model calculation,the analysis and evaluation of the calculation precision of the model are realized.Third,set up a joint operation model of reservoir group to ensure the ecological flow of Guilin,and give the corresponding optimization algorithm.A comprehensive system of constraint is put forward considering reservoir water balance,water restriction and flow restriction.Taking the six reservoirs in the upper reaches of Guilin,such as green lion,Si an river and Xiao dissolve River as the research objects,the ecological scheduling is studied from the angle of protecting the ecological water of the Lijiang River during the dry season with the BP neural network model and the multiple linear regression model used in the previous paper.By setting different calculation steps,we can compile and calculate the water dispatching plan based on year,month,day and real time.Fourth,based on the above research,using the Eclipse2016 development environment and the database engine,a decision support system for the Lijiang River Ecological replenishment scheduling is developed to ensure the 60m3/s ecological flow of the Guilin section of the Lijiang River.The system provides functional services such as information service,runoff analysis,water forecast,ecological scheduling and decision making.It provides a strong support for the decision-makers to formulate a reasonable ecological scheduling scheme.