毕业论文---bp神经网络预测空气质量指数设计.doc

摘 要 空气质量指数的大小可以用来反应空气质量的好坏，而空气质量指数主要受PM2.5，PM10,一氧化碳，二氧化氮及二氧化硫等多种污染物的浓度影响，使得空气质量指数问题具有很大的不确定性和一定的复杂性。神经网络作为一种描述和刻画非线性的强有力工具，具有较强的自学习、自组织、自适应能力等特点，特别适合于对具有多因素性、不确定性、随机性、非线性和随时间变化特性的对象进行研究。本文基于神经网络BP算法和RBF算法，利用MATLAB神经网络工具箱建立空气质量指数模型并对空气质量指数进行预测。计算结果表明BP和RBF模型应用于大气污染预报具有较高的预测精度和良好的泛化能力，它为信息社会的城市空气污染预报工作提供了一种全新的思路和方法。而通过BP和RBF算法的比较，更好的展现了神经网络在预测方面应用的可行性。同时表明这两种方法具有一定的客观性和积极性。 关键词：空气质量指数；BP神经网络；RBF神经网络；MATLAB；预测 Abstract The size of the air quality index can be used to response the air quality, and the air quality index mainly due to PM2.5, PM10, carbon monoxide, nitrogen dioxide, sulfur dioxide and other factors, makes the issue of air quality is a great uncertainty and a certain degree of complexity. Neural network description and characterization as a powerful tool for non-linear phenomenon, with strong self-learning, self-organization, the characteristics of adaptive capacity, especially suitable for factor, uncertainty, randomness, non-linear and time-varying characteristics of the object of research. This design bases on the BP neural network algorithm and RBF neural network algorithm, using MATBLB neural network toolbox to establish air quality model and forecast the air quality index. The computation results showed that the BP model had good quality on forecasting precision and generalization ability. Besides,it provided a new method for urban air pollution forecasting. Through the comparison of BP and RBF algorithm, showing the feasibility of neural network in the prediction aspect better, and achieved good results which indicate the objectivity and enthusiasm of the design. Key words: Air quality index; BP neural network ;RBF neural network; MATLAB; Forecast 目录 1 绪论 1 1.1课题的研究背景及意义 1 1.2神经网络的发展与研究现状 2 1.3课题设计路线及主要工作 3 2 神经网络简介 4 2.1 BP神经网络简介 4 2.1.1 BP神经网络的定义 4 2.1.2 BP神经网络的基本原理 4 2.1.3 BP神经网络的应用 5 2.1.4 BP神经网络的优点及局限性 5 2.2 RBF神经网络简介 6 2.2.1 RBF神经网络的定义 6 2.2.2 RBF神经网络的基本原理 7 2.2.3 RBF神经网络的特点 7 2.2.4 RBF神经网络的优点及局限性 8