金屬氧化物氣體感測器由於價格低廉，靈敏度高，穩定性良好而被廣泛用於氣體濃度預估和辨識。但金屬氧化物感測器的氣體選擇性較差，這也導致針對混合物以及純物質氣體時會造成分類錯誤的問題，常見的解決方式是調變溫度改變感測器上的反應以增加其特徵多樣性，本研究以單顆感測器使用溫度調變以取得更多特徵增加氣體選擇性並改善濃度預測演算法分類及量化氣體混合物。
本研究使用單顆金屬氧化物感測器並以方波作為其加熱器波形，使感測器在加熱過程中與氣體反應，藉此獲得感測器在溫度變化的動態反應，動態反應不僅對於不同氣體含有不同特徵，因此能增加氣體選擇性，且針對混合物也能反應出成分的特徵，因此本實驗提出適合動態特徵與氣體濃度的關係式，並整理成多項式互動項混合物模型，藉此預測二元混合物以及純物質氣體的濃度。
在結果中，本研究針對不同濃度甲醇、乙醇及其不同比例的混合物藉由方波所產生的動態特徵以及峰值特徵做為比較，並使用不同的特徵擷取方式以及代入不同的混合物模型得到不同預測結果，最終由本實驗提出之方法可將預測目標氣體的分類效果提升至100%，並針對純物質量化誤差率降到1.4%而針對混合物到13.0%。

Metal-oxide(MOX) gas sensors are widely used for gas concentration estimation and identification due to their low cost, high sensitivity and stability. However, the MOX sensors present low selectivity to different gases, which lead to the problem of classification for mixtures and pure gases. A common solution is to modulate the temperature on sensor films to increase the diversity of features. Therefore, a temperature modulation method was implemented on a sensor to acquire more features to increase gas selectivity and a mixture gas prediction algorithm was improved to classify and quantify gas mixtures properly.
In this study, a square wave was applied as the heater waveform. The sensor did chemical reactions with the gas during heating, thereby a dynamic response of the conductance of the sensor as temperature change was obtained. The information of the dynamic response not only contains different characteristic for different gases, but also retains the each characteristics of the components as reacting with mixture. Therefore, a polynomial interaction term mixture model with dynamic response was proposed to predict concentration of the binary mixtures and pure gases.
In the results, this study acquired sensor response from different concentrations of methanol, ethanol and their different proportions under temperature modulation. Moreover, the dynamic response and steady-state response were compared including two feature extraction methods and two mixture models. Finally, the proposed method in this experiment can improve the classification of the target gas up to 100%. The relative error of quantification was greatly decreased to 1.4% in pure gases and 13.0% in mixture.

摘 要 i
ABSTRACT ii
致 謝 iv
目 錄 v
圖 目 錄 vi
表 目 錄 viii
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 2
1.3 章節簡介 3
第二章 文獻回顧 4
2.1 金屬氧化物氣體感測器簡介 4
2.2 氣體感測器於溫度調變簡介 4
2.3 混合物濃度預測 6
第三章 實驗系統架構 9
3.1 氣體產生與環境控制系統 10
3.2 溫度調變控制系統 12
3.3 訊號讀取系統 13
3.4實驗資料收集流程 13
第四章 溫度調變 15
4.1 溫度調變原理 15
4.1.1 氣體感測器原理 15
4.1.2 反應常數k 16
4.2 動態反應 17
第五章 混合物氣體預測演算法 21
5.1 特徵擷取 21
5.2 混合物模型 23
5.3 預測氣體濃度 24
5.4 留一驗證Leave-One-Out 25
5.5 效果參數 26
第六章 實驗結果 28
6.1 特徵比較 28
6.2 氣體模型 29
6.3 最小平方法視覺化 32
6.4 預測結果 32
第七章 結論與未來展望 37
參考文獻 38
附件一 溫度調變與定溫啟動於分類混合物之比較 42

[1] CHEN, Li-Ying, et al. Development of a dual MOS Electronic nose/camera system for improving fruit ripeness classification. Sensors, 2018, 18.10: 3256.
[2] RYAN, Margaret Amy, et al. Monitoring space shuttle air quality using the jet propulsion laboratory electronic nose. IEEE sensors journal, 2004, 4.3: 337-347.
[3] SCHNABEL, R. M., et al. Electronic nose analysis of exhaled breath to diagnose ventilator-associated pneumonia. Respiratory medicine, 2015, 109.11: 1454-1459.
[4] MACHADO, Roberto F., et al. Detection of lung cancer by sensor array analyses of exhaled breath. American journal of respiratory and critical care medicine, 2005, 171.11: 1286-1291.
[5] WILSON, Alphus Dan. Recent applications of electronic-nose technologies for the noninvasive early diagnosis of gastrointestinal diseases. In: Multidisciplinary Digital Publishing Institute Proceedings. 2017. p. 147.
[6] MADROLLE, Stéphanie; GRANGEAT, Pierre; JUTTEN, Christian. A Linear-Quadratic Model for the Quantification of a Mixture of Two Diluted Gases with a Single Metal Oxide Sensor. Sensors, 2018, 18.6: 1785.
[7] ZHANG, Dongzhi, et al. Quantitative detection of formaldehyde and ammonia gas via metal oxide-modified graphene-based sensor array combining with neural network model. Sensors and Actuators B: Chemical, 2017, 240: 55-65.
[8] HOSSEIN-BABAEI, Faramarz; AMINI, Amir. Recognition of complex odors with a single generic tin oxide gas sensor. Sensors and Actuators B: Chemical, 2014, 194: 156-163.
[9] RASTKHADIV, Ali; JENABI, Amin; SOURI, Asma. Discriminating among different tea leaves using an operating temperature-modulated tin oxide gas sensor. In: IOP Conference Series: Materials Science and Engineering. IOP Publishing, 2016. p. 012019.
[10] FORT, A., et al. Selectivity enhancement of SnO2 sensors by means of operating temperature modulation. Thin Solid Films, 2002, 418.1: 2-8.
[11] AMINI, Amir; BAGHERI, Mohammad Ali; MONTAZER, Gholam Ali. Improving gas identification accuracy of a temperature-modulated gas sensor using an ensemble of classifiers. Sensors and Actuators B: Chemical, 2013, 187: 241-246.
[12] NGO, Kieu An; LAUQUE, Pascal; AGUIR, Khalifa. High performance of a gas identification system using sensor array and temperature modulation. Sensors and Actuators B: Chemical, 2007, 124.1: 209-216.
[13] HUANG, Xingjiu, et al. Gas sensing behavior of a single tin dioxide sensor under dynamic temperature modulation. Sensors and Actuators B: Chemical, 2004, 99.2-3: 444-450.
[14] BARSAN, N.; KOZIEJ, D.; WEIMAR, U. Metal oxide-based gas sensor research: How to?. Sensors and Actuators B: Chemical, 2007, 121.1: 18-35.
[15] WANG, Chengxiang, et al. Metal oxide gas sensors: sensitivity and influencing factors. Sensors, 2010, 10.3: 2088-2106.
[16] FIGARO USA., INC. http://www.figarosensor.com/
[17] SEARS, W. M.; COLBOW, K.; CONSADORI, Franco. General characteristics of thermally cycled tin oxide gas sensors. Semiconductor Science and Technology, 1989, 4.5: 351.
[18] WLODEK, Stanislaw; COLBOW, Konrad; CONSADORI, Franco. Signal-shape analysis of a thermally cycled tin-oxide gas sensor. Sensors and Actuators B: Chemical, 1991, 3.1: 63-68
[19] HUANG, Xingjiu, et al. Quantitative analysis of pesticide residue based on the dynamic response of a single SnO2 gas sensor. Sensors and Actuators B: Chemical, 2004, 99.2-3: 330-335.
[20] PERERA, A., et al. Evaluation of fish spoilage by means of a single metal oxide sensor under temperature modulation. Sensors and Actuators B: Chemical, 2010, 146.2: 477-482.
[21] LEE, Andrew P.; REEDY, Brian J. Temperature modulation in semiconductor gas sensing. Sensors and Actuators B: Chemical, 1999, 60.1: 35-42.
[22] RYAN, Margaret Amy, et al. Monitoring space shuttle air quality using the jet propulsion laboratory electronic nose. IEEE sensors journal, 2004, 4.3: 337-347.
[23] TANG, Kea-Tiong; LIN, Yi-Shan; SHYU, Jyuo-Min. A local weighted nearest neighbor algorithm and a weighted and constrained least-squared method for mixed odor analysis by electronic nose systems. Sensors, 2010, 10.11: 10467-10483.
[24] WOLD, Svante; ESBENSEN, Kim; GELADI, Paul. Principal component analysis. Chemometrics and intelligent laboratory systems, 1987, 2.1-3: 37-52.
[25] MORÉ, Jorge J. The Levenberg-Marquardt algorithm: implementation and theory. In: Numerical analysis. Springer, Berlin, Heidelberg, 1978. 105-116.