臺灣博碩士論文加值系統

　　本文以類神經網路為基礎，提出汽油引擎故障診斷系統架構，包括引擎運轉狀態診斷系統、故障來源診斷系統、空氣流量感知器故障程度診斷系統、洩漏來源診斷系統、曲軸通風系統洩漏程度診斷系統及燃油壓力調整器洩漏程度診斷系統。
　　先以引擎運轉狀態診斷系統來判別引擎之運轉狀態；當引擎被判定為不正常運轉時，由引擎故障源診斷系統來判別引擎之故障源。故障源可分為空氣流量感知器故障與引擎洩漏。若空氣流量感知器故障，接著用空氣流量感知器故障程度診斷系統來判別故障程度。若引擎產生洩漏，先以引擎洩漏來源診斷系統來判別洩漏來源。洩漏來源可分為曲軸通風系統洩漏與燃油壓力調整器洩漏兩種。當引擎為曲軸通風系統洩漏時，則以曲軸通風系統洩漏程度診斷系統來判別其洩漏程度；否則以燃油壓力調整器洩漏程度診斷系統來判別其洩漏程度。
　　由實驗結果知，診斷結果的均方根誤差皆小於0.1%，因此本文所提出之汽油引擎故障診斷系統架構是可行的

　　In this paper, neural networks are employed to construct a gasoline engine fault diagnostic system including engine-state diagnostic system, fault source diagnostic system, air-flow sensor fault diagnostic system, leaking source diagnostic system, crankcase ventilation system leaking-degree diagnostic system and fuel pressure regulator leaking-degree diagnostic system
　　The running engine is detected using engine-state diagnostic system to identify the state first, if the engine state is abnormal, distinguish the fault source by engine fault source diagnosis system. The fault sources include fault air-flow sensor and engine leakage. If it is fault air-flow sensor, air-flow-sensor fault-degree diagnostic system is employed to indentify the fault degree, otherwise use leaking source diagnosis system to identify the leaking source. Leaking sources are divided into leakage of crankcase ventilation system and leakage of fuel pressure regulator. If crankcase ventilation system leaks, detect the leaking degree using crankcase ventilation system leaking degree diagnosis system, otherwise use fuel pressure regulator leaking degree diagnosis system to identify the leaking degree.
　　The experiment results show that the root-mean-square errors of diagnostic results are all less than 0.1%, therefore, the proposed gasoline engine fault diagnostic structure is feasible.

目次
中文摘要 ........................................................................................................................... i
英文摘要 .......................................................................................................................... ii
誌謝 ................................................................................................................................. iii
目次 ................................................................................................................................. iv
表目錄 ............................................................................................................................. vi
圖目錄 ............................................................................................................................ vii
第一章 緒論 .................................................................................................................. 1
1.1 研究動機 ........................................................................................................... 1
1.2 文獻回顧 ........................................................................................................... 2
1.3 論文大綱 ........................................................................................................... 3
第二章 類神經網路 ...................................................................................................... 5
2.1 生物神經網路 ................................................................................................... 5
2.2 類神經網路之神經元介紹 ............................................................................... 7
2.3 活化函數類型 ................................................................................................... 9
2.4 類神經網路系統架構 ..................................................................................... 12
2.5 類神經網路的學習 ......................................................................................... 14
2.6 倒傳遞類神經網路 ......................................................................................... 16
第三章 汽油引擎多點燃油噴射系統 ........................................................................ 20
3.1 汽油引擎運轉參數 ......................................................................................... 20
3.2 空燃比與汽油引擎之關係 ............................................................................. 21
3.3 空氣流量感知器故障對汽油引擎之影響 ..................................................... 23
3.4 洩漏來源對汽油引擎之影響 ......................................................................... 24
第四章 引擎故障診斷系統 ........................................................................................ 27
4.1 資料的正規化 ................................................................................................. 28
4.2 引擎運轉狀態診斷系統 ................................................................................. 28
4.3 引擎故障來源診斷系統 ................................................................................. 30
4.4 空氣流量感知器故障程度診斷系統 ............................................................. 32
4.5 引擎洩漏來源診斷系統 ................................................................................. 34
4.6 曲軸通風系統洩漏程度診斷系統 ................................................................. 36
4.7 燃油壓力調整器洩漏程度診斷系統 ............................................................. 38
第五章 實驗設備介紹 ................................................................................................ 41
5.1 汽油引擎系統 ................................................................................................. 41
5.2 汽油引擎動態監控系統 ................................................................................. 44
第六章 實驗結果 ........................................................................................................ 55
6.1 引擎訊號擷取與儲存 ................................................................................... 55
6.2 引擎運轉狀態診斷結果 ............................................................................... 57
6.3 引擎故障來源診斷結果 ............................................................................... 59
6.4 空氣流量感知器故障程度診斷結果 ........................................................... 61
6.5 引擎洩漏來源診斷結果 ............................................................................... 63
6.5.1 曲軸通風系統洩漏程度診斷結果 .......................................................... 66
6.5.2 燃油壓力調整器洩漏程度診斷結果 ...................................................... 68
第七章 結論及建議 .................................................................................................... 70
7.1 結論 ................................................................................................................. 70
7.2 後續研究建議 ................................................................................................. 71
參考文獻 ........................................................................................................................ 72

1. Huang, Y. C., Yang, H. T. and Huang, C. L., An evolutionary computation based fuzzy fault diagnosis system for a power transformer, Soft Compution in Intelligent System & Information Processing Proceedings Asian, pp.218-223, 1996.
2. Dai, F. and Sugisaka, M., Fuzzy fault query approach in the fault diagnosis of a power plant system, Proceedings IEEE/RSJ International Conference on Intelligent Robots and System, Vol.2, pp.752-754, 1999.
3. Buehler, P. J., Freanchek, M. A. and Makki, Mass air flow sensor diagnostics for adaptive fuel control of internal combustion engines﹐Proceedings of the American Control Conference Anchorage, Vol.3, pp.2064-2069, 2002.
4. Chen, P. C, Neuro-fuzzy-based fault detection of the airflow sensor of an idling gasoline engine,Proc.IMechE. Part D:J.Automobile Engineering,2005,219,1-14.
5. 楊啟良，發動機電路多重故障診斷法，國立成功大學航空太空工程研究所碩士論文，民國91 年。
6. 陳聖洲，汽油引擎之類神經模糊故障診斷，南台科技大學機械研究所碩士論文，民國92 年
7. 廖哲鴻，汽油引擎的空氣流量及噴油嘴故障檢測，南台科技大學機械研究所碩士論文，民國94 年。
8. Chen, P. C, and Liu, J. Z., Fault correction of Air-flow Sensor by Using Gentic Algorithm and Neural Fuzzy Technique，中華民國第十五屆模糊理論及其應用會議，pp.421-426,2007。
9. 劉建志，應用類神經網路於汽油引擎的漏氣偵測及動態資料估測，南台科技大學機械研究所碩士論文，民國96年。
10. 張斐章、張麗秓，類神經網路，東華書局，民國 94 年。
11. 中華汽車，LACER VIRAGE 工作手冊，引擎與全車電路。
12. 惠汝生編著，LabView7.1 Express 圖控程式應用，全華科技圖書股份有限公司，民國93 年。
13. 汪國禎編著，汽車學（I）汽油引擎篇，復文書局，民國89年。
14. William H.Crouse & Donald L.Anglin編著，劉崇富譯，汽車學（一）汽油引擎，