区域生态安全评价的BP神经网络方法

生态安全评价是建立生态安全预警系统及进行环境管理的基础。为确立有效的区域生态安全评价方法，基于压力－状态－响应概念模型，构建了区域生态安全评价指标体系和评价标准；采用BP神经网络的途径，对安徽省17个地级市的生态安全评价问题进行了研究。结果表明：宣城、池州、黄山为Ⅱ级，亳州、宿州、阜阳、滁州、六安、巢湖、安庆为Ⅲ级，合肥、淮北、蚌埠为IV级，淮南、马鞍山、芜湖、铜陵为Ⅴ级。区域生态安全的BP神经网络评价方法涵义明确，计算过程简单，能区分生态安全的等级，可以应用于具有评价标准的其它生态系统综合评价。

ASSESSMENT OF REGIONAL ECOLOGICAL SECURITY USING BACK PROPAGATION NEURAL NETWORK METHOD

The assessment of ecological security is a new topic in the field of untraditional research on security and sustainable development,and is also the basis to set up warning system for ecological security and environment management.In order to establish an effective evaluation method for regional ecological security,the authors built an evaluation index system and then decided the evaluation criteria based on pressure-state-response conceptual model.The level of ecological security of 17 regional cities in Anhui Province was evaluated by backpropagation (BP) neural network channels.It is shown that Xuancheng,Chizhou and Huangshan are in grade Ⅱ;Bozhou,Suzhou,Fuyang,Chuzhou,Liuan,Chaohu and Anqing are in grade Ⅲ;Hefei,Huaibei and Bengbu are in grade Ⅳ;Huainan,Ma'anshan,Wuhu and Tongling are in grade Ⅴ.It is indicated that the method can truly reflect the situation of regional ecological security.Since regional ecological security evaluation of BP neural network method has explicit meanings and simple computational process,and is able to sort and rank subjects according to their ecological security level,it can be applied in the comprehensive assessment of other kinds of ecological systems which have evaluation criteria.