Examining Land-Use Change Trends in Yucheng District, Ya’an City, China, Using ANN-CA Modeling
Journal of Urban Planning and Development(2023)
摘要
Land systems are those that most closely interact with human beings in terms of natural resources. The relationship between the simulation of their dynamic evolution and their microscopic and macroscopic elements is complex and nonlinear. In this study, we aimed to design a scientific model that could unify the macroscopic statistical distribution with microscopic individual movements, in order to simplify the model structure and definition of transformation rules. Taking Yucheng District, City, China, as a study site, we constructed an artificial neural network cellular automata (ANN-CA) model that uses neural network algorithms to predict the land area required to restrict the number of cellular automat (CA) space simulations. First, we took the raster data of the current land-use map of Yucheng District for the years 2001 and 2009, and the effect coefficients of slope, the river system, and other elements were used as initial data for the ANN-CA model. We then analyzed the superimposed land-use evolution law from 2001 to 2009. Next, within the error limits of the results for the 2017 neural network area prediction and the space center of gravity for each land type, we predicted the land-use changes for 2017 and determined the final land type for all grid units based on the set evolution rules and land conversion threshold. Finally, the land-use prediction simulation map for 2017 and actual land-use conditions were verified for quantitative and spatial accuracy. The results reveal that the model simulated the land-use changes for Yucheng District relatively accurately. It was also found that selecting the appropriate thresholds and random variable parameters improved the model's simulation accuracy and that the land use in the study area exhibits a clear urbanization trend.
更多查看译文
关键词
Land-use change,ArcGIS,Neural network,Cellular automata,Simulation and prediction
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要