摘  要
隨著世界對能源及物質(zhì)需求了不斷增長，煤炭資源作為能源的重要組成部分，其開發(fā)也融入了高科技數(shù)據(jù)監(jiān)測技術(shù)。而煤礦安全在煤炭生產(chǎn)中占有特別重要的地位。煤礦事故發(fā)生的主要形式為瓦斯爆炸、瓦斯突出及中毒窒息。瓦斯治理是煤礦安全研究的主攻方向，是學(xué)術(shù)界和工業(yè)界的研究熱點之一。
礦井中瓦斯涌出濃度是一組時間監(jiān)測序列數(shù)據(jù)，具有數(shù)據(jù)量龐大和波動性兩大特點。如何從當(dāng)前瓦斯涌出濃度數(shù)據(jù)中有效地提取壓縮數(shù)據(jù)并加以分析，是防止瓦斯事故的首要問題。小波分析具有減少數(shù)據(jù)量和降維的特點，特別適合處理海量時間監(jiān)測序列，成為海量數(shù)據(jù)分析的一個強有力的工具。本文主要是以使用小波算法處理一維海量數(shù)據(jù)為例，介紹如何完成對數(shù)據(jù)的小波分解以及小波重構(gòu)，達到數(shù)據(jù)壓縮的目的，為下一步的相似搜索模塊開發(fā)作好準(zhǔn)備。
論文主要介紹了關(guān)于時間序列的簡要分析，簡單概述了數(shù)據(jù)壓縮技術(shù)的中的幾種常用算法，并以小波算法為重點分析了小波算法的性質(zhì)和特點，介紹了用小波算法進行數(shù)據(jù)壓縮的方法。最后將小波處理一維數(shù)據(jù)的原理與計算機技術(shù)結(jié)合起來，利用軟件完成數(shù)據(jù)壓縮并以曲線形式表現(xiàn)出來。經(jīng)測試本人設(shè)計程序結(jié)果正確，可以用于數(shù)據(jù)壓縮。

關(guān)鍵詞：瓦斯?jié)舛?；時間監(jiān)測序列；相似搜索模塊開發(fā)；小波算法；數(shù)據(jù)壓縮
 
ABSTRACT
As the world's energy and material needs of a growing, coal energy resource as an important component of its development is also integrated into the high-tech data monitoring technology. The coal mine safety in coal production is in the possession particular importance. Mine accident occurred as the main form is gas explosion, poisoning and suffocation gas outburst. Gas Control Coal Mine Safety Research is the main direction of attack, the academia and industry on one of the hot spots.
Mine gas emission level is monitoring a group of time-series data, with the huge volume of data and volatility of two major characteristics. How from the current concentration of gas emission data compression effectively extract data and analysis, prevent gas accidents are the most important issue. Wavelet analysis of data and reduce the drop-dimensional characteristics, particularly suitable for mass processing time monitoring sequence, a massive data analysis of a powerful tool. This Dissertation is the use of wavelet algorithm for handling one-dimensional data as an example on completing the wavelet analysis of the data and wavelet reconstruction, to achieve the purpose of data compression, similar to the search for the next step to prepare for the development of modules.
Dissertation introduced a brief analysis of time series, simple overview of the data compression technology in several common algorithms, wavelet algorithm and to focus on the wavelet analysis of the nature and characteristics of algorithm, introduced by wavelet data compression algorithm method. Finally, the wavelet deal with the principle of one-dimensional data combined with computer technology, using software and data compression to the completion of the form shown.

Keywords: gas concentration; time monitoring sequence; similar search module development; wavelet algorithm; data compression
目  錄
1 緒論 1
1.1 選題背景及研究意義 1
1.2 國內(nèi)外研究動態(tài)及發(fā)展趨勢 1
1.2.1 國內(nèi)外研究動態(tài) 1
1.2.2 發(fā)展趨勢 2
1.3 本文的主要工作及章節(jié)安排 3
2 時間序列的分析 4
2.1 時間序列的定義 4
2.2 時間序列的相似性研究 4
2.2.1 相似性度量 4
2.3 時間序列變換 7
2.3.1 變換后的查詢序列特征 7
2.3.2 時間序列變換的原則 7
2.3.3 時間序列相似性研究中特征提取的變換方法 8
3 小波理論簡介 11
3.1 小波簡史 11
3.2 小波分析 12
3.2.1 連續(xù)小波變換 12
3.2.2 離散小波變換 14
3.2.3 小波重構(gòu) 16
3.3 一維哈爾小波變換 17
3.3.1 哈爾基函數(shù) 17
3.3.2 哈爾小波函數(shù) 18
3.3.3 規(guī)范化算法 19
4 基于哈爾小波的特征提取算法 22
4.1 小波分解算法原理 22
4.2 小波逆變換（小波重構(gòu)）原理 23
4.3特征提取算法例題解析 24
5 算法流程圖及程序運行介紹 27
5.1 算法流程圖 27
5.5.1 小波分解的流程圖 27
5.5.2 小波重構(gòu)的實現(xiàn)流程圖 29
5.2 程序運行介紹 31
5.2.1 程序運行環(huán)境 31
5.2.2 程序截圖及說明 31
結(jié)束語 42
致謝 43
參考文獻 44