【正文】 沈陽理工大學學士學位論文 I 摘 要 空氣質(zhì)量指數(shù)的大小可以用來反應空氣質(zhì)量的好壞，而空氣質(zhì)量指數(shù)主要受， PM10,一氧化碳，二氧化氮及二氧化硫等多種污染物的濃度影響，使得空氣質(zhì)量指數(shù)問題具有很大的不確定性和一定的復雜性。神經(jīng)網(wǎng)絡作為一種描述和刻畫非線性的強有力工具，具有較強的自學習、自組織、自適應能力等特點，特別適合于對具有多因素性、不確定性、隨機性、非線性和隨時間變化特性的對象進行研究。本文基于神經(jīng)網(wǎng)絡 BP 算法和 RBF 算法，利用 MATLAB 神經(jīng)網(wǎng)絡工具箱建立空氣質(zhì)量指數(shù)模型并對空氣質(zhì)量指數(shù)進行預測。 計 算結(jié)果表明 BP 和 RBF 模型應用于大氣污染預報具有較高的預測精度和良好的泛化能力，它為信息社會的城市空氣污染預報工作提供了一種全新的思路和方法。而 通過 BP 和 RBF 算法的比較，更好的展現(xiàn)了神經(jīng)網(wǎng)絡在預測方面應用的可行性。同時表明這兩種方法具有一定的客觀性和積極性。 關(guān)鍵詞 ：空氣質(zhì)量指數(shù)； BP 神經(jīng)網(wǎng)絡； RBF 神經(jīng)網(wǎng)絡； MATLAB；預測 沈陽理工大學學士學位論文 II Abstract The size of the air quality index can be used to response the air quality, and the air quality index mainly due to , 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 plexity. Neural work description and characterization as a powerful tool for nonlinear phenomenon, with strong selflearning, selfanization, the characteristics of adaptive capacity, especially suitable for factor, uncertainty, randomness, nonlinear and timevarying characteristics of the object of research. This design bases on the BP neural work algorithm and RBF neural work algorithm, using MATBLB neural work toolbox to establish air quality model and forecast the air quality index. The putation 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 parison of BP and RBF algorithm, showing the feasibility of neural work 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 work 。RBF neural work。 MATLAB。 Forecast 沈陽理工大學學士學位論文 III 目錄 1 緒論 ....................................................................................................................................... 1 課題的研究背景及意義 .............................................................................................. 1 神經(jīng)網(wǎng)絡的發(fā)展與研究現(xiàn)狀 ...................................................................................... 2 課題設計路線及主要工作 .......................................................................................... 2 2 神經(jīng)網(wǎng)絡簡介 ....................................................................................................................... 4 BP 神經(jīng)網(wǎng)絡簡介 ........................................................................................................ 4 BP 神經(jīng)網(wǎng)絡的定義 .......................................................................................... 4 BP 神經(jīng)網(wǎng)絡的基本原理 .................................................................................. 4 BP 神經(jīng)網(wǎng)絡的應 用 .......................................................................................... 5 BP 神經(jīng)網(wǎng)絡的優(yōu)點及局限性 .......................................................................... 5 RBF 神經(jīng)網(wǎng)絡簡介 ...................................................................................................... 6 RBF 神經(jīng)網(wǎng)絡的定義 ........................................................................................ 6 RBF 神經(jīng)網(wǎng)絡的基本原理 ................................................................................ 7 RBF 神經(jīng)網(wǎng)絡的特點 ........................................................................................ 7 RBF 神經(jīng)網(wǎng)絡的優(yōu)點及局限性 ........................................................................ 8 模糊神經(jīng)網(wǎng)絡簡介 ................................................................................................... 10 模糊神經(jīng)網(wǎng)絡定義 ......................................................................................... 10 模糊神經(jīng)網(wǎng)絡基本形式 ................................................................................. 10 模糊神經(jīng)網(wǎng)絡的發(fā)展動向及用途 ................................................................. 10 3 開發(fā)軟件簡介 ..................................................................................................................... 12 MATLAB 發(fā)展歷程 ................................................................................................... 12 MATLAB 的語言特點 ............................................................................................... 13 MATLAB 的基本功能及應用 ................................................................................... 15 4 基于 BP 神經(jīng)網(wǎng)絡預測仿真研究 ...................................................................................... 16 預測原理 ..................................................................................... 錯誤 !未定義書簽。 預測方法 ................................................................................................................... 17 預測步驟 ................................................................................................................... 20 沈陽理工大學學士學位論文 IV 基于 BP 網(wǎng)絡預測仿真實現(xiàn) ..................................................................................... 25 5 基于 RBF 神經(jīng)網(wǎng)絡預測仿真研究 .................................................................................... 29 預測原理 ................................................................................................................... 29 預測算法 ................................................................................................................... 30 基本步驟 .......................................................................................................... 31 仿真實現(xiàn) .......................................................................................................... 32 BP 與 RBF 神經(jīng)網(wǎng)絡預測仿真比較 ......................................................................... 34 結(jié)論 ......................................................................................................................................... 36 致謝 ............................................................