基于神经网络的天津市城市化水平分析及预测

Forecast and Analysis of Urbanization Level of Tianjin Based on Neuro-Network

针对一般城市化水平分析只侧重人口指标测定的不足,在比较了灰色预测与神经网络预测优劣的基础上,以天津市为例,采用误差逆传播人工神经网络(简称BP神经网络)方法对人口、产业以及人居环境三项城市化水平指标做出短期预测,在与北京、上海等城市的同一指标的比较后得出天津市当前及未来的城市化综合发展指标,并据此提出了相关建议。

To be a modem metropolis, the scale of its city people ratio must reach some extent; at the same time, corresponding economic and environmental level is necessary. The forecast for the ratio of urban population, the ratio of third industry production value and the public greenbelt per person in Tianjin for several years short are made by error back propagation neuro-network （BP neuro-network） model with MATLAB. Based on the comparisons of the three factors with Beijing and Shanghai, the development of Tianjin in these years and integrated progress of Tianjin in future are given, and then some advice is also put forward.