基于遥感监测的高排污车辆鉴别研究.pdfVIP

th Proceedings of the 25 Chinese Control Conference 7-11 August, 2006, Harbin, Heilongjiang 基于遥感监测的高排污车辆鉴别 1 1,2 1 曾君, 郭华芳 , 胡跃明 1. 华南理工大学自动化科学与工程学院， 广州 510640 E-mail: zengjuns@263.net 2. 广东省科学院自动化工程研制中心，广州 510070 E-mail: ghf833@ 摘 要: 汽车尾气是城市主要污染源之一. 本文在简要介绍汽车尾气遥感监测的基础上, 以广州市2004 年遥感监 测数据作为基础, 主要研究出租车污染情况， 利用 BP 神经网络对汽车尾气排放超标进行判断, 以探讨神经网络 技术在尾气污染预测中的应用. 网络仿真结果表明, 利用神经网络进行预测的准确率达到 93%, 优于传统方法建 模的准确率. 关键词: 汽车尾气，遥感监测，神经网络，超标预测，主成分分析 Modeling of Vehicle Gross Emitter Prediction Based on Remote Sensing Data Jun Zeng1, Huafang Guo1,2, Yueming Hu1 1. College of Automation Science and Engineering，South China University of Technology，Guangzhou，510640 E-mail: zengjuns@263.net 2. Automation Engineering RM Center, Guangdong Academy of sciences, Guangzhou, 510070 E-mail: ghf833@ Abstract: Vehicle emission is a major source of air pollution in urban cities. After the introduction of vehicle emissions remote sensing technology, the neural network model for high emitter prediction is made based on the 2004 remote sensing data of Guangzhou. The results show that satisfactory prediction was obtained by reasonable selection of original data for input layer element and algorithm. And the correct rate and the ability of generalization are superior to the traditional model in prediction. Keywords: neural network, vehicle emission, remote sensing, gross emitter 显示车辆实际运行时的尾气排放,而招致各方严重的 1 引言(Introduction)