基于BP人工神经网络的小城镇生态环境质量评价模型

针对中国小城镇生态环境质量综合评价存在的问题，以生态环境质量指标体系作为神经网络的输入、以生态环境等级评分作为输出，基于BP人工神经网络，建立了具有20个隐含层节点、3层网络的小城镇生态环境质量评价模型；以生态环境指标的各级评价标准作为模型的训练样本，以训练样本数量的10%以及各指标各等级的临界值、中间值作为检验样本，以研究区生态环境质量的实际监测值作为预测样本，利用MATLAB软件对BP人工神经网络进行训练，并对鄂州市杜山镇生态环境质量等级进行了模式识别.结果表明：利用BP人工神经网络方法对小城镇生态环境质量进行预测是可行的、可靠的，它不仅能很好地评价区域生态环境质量，而且能够与区域生态环境的实际特征相结合.

Assessment model of townlet eco-environmental quality based on BP-artificial neural network.

Aiming at the problems in the townlet eco-environmental quality assessment in China, a comprehensive assessment model of townlet ecological environmental quality based on BP-artificial neural network was set up, which contained 20 cryptic layer nodes and 3 layers. The rank classification criterion of eco-environmental quality’s assessment indicator system were chosen as the training sample of the model, the 10% of training sample as well as the middle and critical values were regarded as examining sample, and the monitoring values of assessment region were treated as test sample. Choosing the Dushan Town in Ezhou City as an example, the training and prediction were made by using MATLAB software. The results showed that BP-artificial neural network was not only feasible and dependable, but also could provide acceptable results in accord with the regional realistic eco-environmental feature.