应用B—P神经网络的城市空气污染预测研究

Research on A Kind of Urban Air Quality Forecast Model Based on BP Neural Network

在具有城市重点污染源的排放数据的条件下,利用人工神经网络的自学习、自组织、自适应和较强容错性等特点,探讨B-P神经网络应用于城市空气质量预测的适用性。试验结果表明:此种方法基本适用,且在污染源监测数据充足的基础上,预测结果会更加准确。

Under the terms of having continuous monitoring data of urban key pollution sources, this article discusses the feasibility of B-P nerve network technology for urban air quality forecast by B-P nerve network technology＇s advantages of self-study, self-organization, self-adoption and unusual fault tolerance. Results shows： This method is generally applicable, and the prediction results will be more accurate if monitoring data is enough.