摘要
区域社会经济系统的可持续发展依赖于系统成分的协调程度及与生态系统的增益耦合。传统研究多以系统动力学或多目标规划模型来调控系统。文中以人工神经网络模型来模拟和映射社会经济系统的结构和功能。通过构建、训练、测试、运行河北省环渤海地区生态经济网络模型,提出了以减缓发展速度。
Sustainable development of regional social and economic system depends on the coordinated degree of system components and the gain coupling with ecosystem. Smitulating and regulating socioeconomic system become a challenging work due to its complexity. The weakness in dealing with diffusely coupling, nonlinear and high order equation makes the traditional methods, such as system dynamics and multiobjective planning analysis, become less important in modeling behavior and dynamic character of socioeconomic system. In this paper, a feed forward artificial neural network (ANN) with back propagation (BP) is tested to simulate structure and function of socio economic system. A BP model established by BP builder consists of input layer, hidden layer and output layer. We choose 12 factors, such as total population, cropping land, forest land, water resource and forest cover rate. as variables of neurons at input layer. At output layer, three neurons stand for the ecological quality index, economic development index and social development index. We generate a set of training data by means of real value of socioeconomic system's components from 1991 to 1995 and then train the BP model. Having served to predict and regulate real socioeconomic system around Bohai area of Hebei province, the BPmodel shows that a goal of coordinated development between ecosystem and socioeconomic system will become true by slowing down the economic development speed, increasing land area for ecological purposes, and reducing capital investment and energy input to the ecosystem.
出处
《地理研究》
CSSCI
CSCD
北大核心
1998年第4期351-359,共9页
Geographical Research
关键词
人工神经网络
能量投入
生态经济系统
协调发展
artificial neural networks, energy input, ecological and economic system, Hebei province