BP人工神经网络在生态环境脆弱性评价方面的应用——以乌拉特前旗为例

Study on the evaluation of eco-environmental vulnerability by BP artificial neural network:A case study of Wulate County

以乌拉特前旗为研究区域,生态环境为评价目标,依据自然和人为因素要求,选取11个评价指标,建立网络输入和网络期望输出的Ⅰ,Ⅱ,Ⅲ三级标准,构建生态环境脆弱性评价模型,运用BP人工神经网络方法进行评价.结果表明,网络仿真的输出值为0.205 9,1.007 6,-0.175 6,判定乌拉特前前旗的生态环境脆弱性为Ⅱ级,属于中度脆弱.最后对脆弱性形成的原因进行了分析.

Take Wulate County as study area, and the ecological environment as evaluation target, basing on the natural and man-made factors, 11 evaluation indicators are selected to establish I , II and III grade standards of network input and network expectation ouput, thus the vulnerability assessment model of ecological environment in Wulate County is constructed. The outputs of network simulation are 0. 2059, 1.0076 and --0. 1756. The vulnerability is determined to the second level, which belongs to the moderately fragile. At the end of this study, the reasons of vulnerability formation are discussed.