Font Size: a A A

Research On Land Use Simulation And Prediction Of Dongting Lake Watershed Based On ANN-CA

Posted on:2020-12-15Degree:MasterType:Thesis
Country:ChinaCandidate:S JiangFull Text:PDF
GTID:2439330578951766Subject:Agricultural informatization
Abstract/Summary:PDF Full Text Request
Land use refers to human beings adopting a series of biological and technological means to carry out long-term or periodic management and reform of the land under the natural conditions of the land itself according to certain economic and social purposes.The breadth,depth and reasonable extent of land use are a concentrated reflection of the scale,level and characteristics of land production.With the continuous improvement of human ability to transform and utilize the natural environment,there will be problems of polluting the environment and destroying the ecological balance with a little carelessness.In land use,this phenomenon is often more prominent.The economic development of the Dongting Lake Basin and the expansion of the city scale have increased the pressure on the ecological environment of the basin.The Dongtimg Lake Basin is facing serious comprehensive ecological problems.Therefore,the study of land use change in the Dongting Lake Basin has important ecological significance.In this paper,the Dongting Lake Basin is the main research area.The land use data,digital elevation model data and humanistic economic data of the study area are used as data sources.The spatial transfer matrix model is used to analyze the land use change characteristics of the Dongting Lake Basin from 1995 to 2015.On this basis,the land use change sample data matrix and the land use change driving factor matrix are constructed.According to the selection and setting of the driving factors,the ANN-CA model is constructed.In this model,the Recurrent Neural Network(RNN)model is used.As a transformation rule of Cellular Automata(CA),14 driving factors in the land use change driving factor matrix of the Dongting Lake watershed are used as input neurons of the neural network model,and the output layer is represented by 6 kinds of land.Using six neurons of type conversion probability,the simulation is performed in the cellular automata,and the model is trained and verified,and the land use situation of the Dongting Lake Basin in 2015 is simulated.After comparing with the real land use situation in 2015,The simulation of land use change in the Dongting Lake watershed is obtained.The accuracy of the model is above 80%,and it has good simulation accuracy.After the model is built,the land use situation of the Dongting Lake watershed in 2015 is simulated,and the simulation results are analyzed to provide theoretical support for the future land use mode of the Dongting Lake watershed.
Keywords/Search Tags:Dongting Lake Basin, Artificial Neural Network, Cellular Automata, Land Use Change, Transfer Matrix
PDF Full Text Request
Related items