基于细胞状演化神经网络的黄河三角洲区域预测

Prediction of the Yellow River Estuary Area Based on Cellular Evolutionary Neural Networks

将细胞状遗传算法（cGA）与反向传播（BP）神经网络相结合，构建了一个细胞状演化神经网络时间序列模型，以1984～2000年间的多时相遥感影像为主要数据源，应用该模型对黄河口地区陆地面积进行预测，并分析了其将来的演变趋势．结果表明，在2001～2010年间研究区域陆地面积将呈现增长与蚀退交替演变、增长高峰逐渐下降的趋势．

An cellular evolutionary neural networks time series model was constructed to predict the change in the coastline of the Yellow River Estuary area. Firsdy, the location and evolution of our research region was described. Secondly, an evolutionary neural networks time series model was constructed. Finally, the model was applied to predict the evolution of our research region, using Remote Sensing （RS） images between 1984 and 2000 as data source. In the end of this paper, the results were analyzed. It indicates that the increase and decrease in the area of our research region appears alternately from 2001 to 2010, and the change tends to slow down.