Land Space Optimization Based On Super Learner And Cellular Automata Model

With the institutional reform of the State Council,China has entered the stage of land and space planning with all-round elements such as mountains,waters,forests,fields,lakes and grasses.The spatial classification system formed by the former division management model can not meet the needs of the unified management of natural resources in the new era.At the same time,spatial planning has entered a new era of ecological civilization.How to maximize the benefits of various spatial elements in the whole region under the premise of protecting the environment has become the focus of attention.Starting from the technical method of territorial space optimization,this paper proposes an optimization model construction method which is coupled with algorithm and model.Taking Anhua County of Hunan Province as the research area,this paper sets optimization rules and objectives for regional development,and divides them into three different development scenarios: historical trend development,farmland protection and ecological priority.The optimal solution under each scenario is solved by using the model,which provides scientific reference for spatial planning.Specific research is as follows:(1)By sorting out the models and algorithms used in land space optimization at home and abroad,it is found that the algorithms are gradually evolving from traditional linear mode to machine intelligence mode,and the models are gradually changing from simplification to diversification.With the continuous improvement of the algorithm and model,the optimization ability becomes stronger and the solving speed becomes faster.On the basis of summing up the experience of predecessors,this paper also constructs an effective optimization model of territorial space.(2)As an important national ecological function protection area,Anhua County has a forest coverage rate of 70%,a small area of cultivated land and fragmented distribution,while the urban land is inefficient and extensive,but still maintains the expansion trend.The disorderly spread of urban land not only causes the disorder of spatial development,but also occupies a large number of ecological land.(3)Construct a territorial space optimization model combining super learning and cellular automata model to optimize the territorial space of Anhua.After optimization,the proportion of all kinds of space is 96.44% of the ecological space,9.98% of the agricultural space and 3.56% of the urban space,and the layout of all kinds of space is more reasonable,which accords with the anticipation of Anhua's future development plan.(4)By setting up three scenarios: historical trend,farmland protection and ecological priority,the land spatial pattern of Anhua County in 2025 was optimized.It is found that the spatial agglomeration degree is the highest under the historical trend development scenario;the cultivated land occupied by urban expansion is the least under the agricultural land protection scenario;and the urban expansion scope is the smallest under the ecological priority scenario,which reserves high-quality land resources for the future ecological and agricultural land development.(5)This paper not only realizes the optimal allocation of land space in practical application,but also adopts the method of combining machine intelligence algorithm with spatial optimization model,which makes the solving process more efficient and accurate.The model can also modify the parameters according to the characteristics of the optimized region,which has strong applicability and provides technical support for efficient land and space optimization in China.