基于BP神经网络城市环境噪声预测及实证研究

摘要

The urban noise pollution influence people's life and work seriously,use the appropriate method to carry on the appraisal and forecast of the noise pollution, then proposed the effective prevention measures is the key question about noise pollution prevention, the article have selected the primary factors which influence the urban environment noise through the gray correlation analysis, used the Back propagation (BP) neural networks to carry on the appraisal and forecast of Beijing's urban environment noise from year 1994 to 2006,the confirmation result error was small, showed the BP neural network based on the gray correlation analysis is the effective way to appraise and forecast the urban environment noise pollution.%城市噪声污染严重影响着人们的生活与工作,采用合适的方法对噪声污染进行评价并预测,进而提出有效的预防及治理措施是噪声污染防治的关键问题,文章利用灰色关联分析选取了影响环境噪声的主要因素,采用BP(Back propagation,BP)神经网络对北京市1994～2006年的环境噪声污染进行评价并预测,验证结果误差较小,说明基于灰色关联的BP神经网络能够有效地对城市环境噪声污染进行评价和预测.