中文摘要：

本文以广州市花都区1990年和2008年TM影像作为数据源,在分析花都区2个时期土地利用格局时空特征的基础上,结合人工神经网络与元胞自动机（ANN-CA）模型对研究区的土地利用空间布局进行模拟.研究表明：根据研究需求构建的ANN-CA模型在实际模拟过程中无需人工定义模型结构和转换规则,只需通过3层神经网络即可模拟复杂的城市形态.因此,通过构建ANN-CA模型对土地利用空间布局进行模拟,可以合理预测城市土地利用发展趋势,具有较强的科学性和可靠性.

英文摘要：

In the paper, based on the analysis of the temporal and spatial characteristics of land use pattern in Huadu District in 2 periods, taking TM image in 1990 and 2008 as data source, ANN-CA model were used to simulate the spatial distribution of land in study area. Research shows according to the ANN-CA model in the actual simulation process, it is not necessary to define the model structure and transformation rules, and it only through 3 layers of neural network. Therefore, the spatial distribution of land use simulated by ANN-CA model can predict the development trend of urban land use, which has strong science and reliability.