| 元胞自动机与神经网络相结合的土地演变模拟 |
| |
| 引用本文： | 赵晶,陈华根,许惠平.元胞自动机与神经网络相结合的土地演变模拟[J].同济大学学报(自然科学版),2007,35(8):1128-1132. |
| |
| 作者姓名： | 赵晶 陈华根 许惠平 |
| |
| 作者单位： | 同济大学,海洋地质国家重点实验室,上海,200092 |
| |
| 基金项目： | 同济大学校科研和教改项目 |
| |
| 摘 要： | 提出了将数据挖掘技术应用于元胞自动机（CA）进行地理模拟的新方法.基于CA原理,利用学习矢量量化神经网络,从不同时相遥感数据中挖掘土地利用演变的内在规律,自动找到土地利用元胞的转换规则,并以该规则反演和预测土地利用格局.在上海市区典型边缘带的应用显示,挖掘出的元胞转换规则,与同期上海城市发展状况相吻合.表明该模型可以满足土地利用演变模拟预测的要求,大大缩短了建立CA转换规则所需时间.
|
| 关 键 词： | 元胞自动机 人工神经网络 土地利用 模拟 转换规则 |
| 文章编号： | 0253-374X（2007）08-1128-05 |
| 修稿时间： | 2005-12-13 |
Simulation of Land Use Evolution Based on Cellular Automata and Artificial Neural Network |
| |
| ZHAO Jing,CHEN Huagen,XU Huiping.Simulation of Land Use Evolution Based on Cellular Automata and Artificial Neural Network[J].Journal of Tongji University(Natural Science),2007,35(8):1128-1132. |
| |
| Authors: | ZHAO Jing CHEN Huagen XU Huiping |
| |
| Institution: | State Key Laboratory of Marine Geology, Tongji University, Shanghai 200092, China |
| |
| Abstract: | In order to find out evolution rules of land use structure,data mining was applied to cellular automata(CA) theory to simulate real land use evolution.By establishing an integrated model made up of CA and learning vector quantization network(LVQ network),evolution rules of land cellular were mined and used to invert the land use pattern and predict land cellular status in future.According to integrated CA-LVQ model,evolution rules of land cellular mined out from past land use pattern were considerably coincident with simultaneous social evolution in research area in Shanghai.The result indicates that such model is able to simulate land use structure evolution and greatly shorten required time of establishing evolution rules. |
| |
| Keywords: | cellular automata artificial neural network land use simulation evolution rule |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
