Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/2107
標題: 類神經網路應用於引擎振動診斷之辨識
Application of Neural Network on the Recognition of Vibration Diagnosis for Engine
作者: 廖基屹
Liaw, Ji-Yih
關鍵字: Engine;引擎;Vibration;Fault Diagnosis;Neural Network;振動;故障診斷;類神經網路
出版社: 機械工程學系所
引用: 【1】吳啟明,“汽油噴射引擎”,全國工商出版社圖書,民國80 年。 【2】林慶銘譯,“最新汽車控制技術”,全華科技圖書股份有限公司,民國87 年。 【3】劉興華、張志瑋、張嘉仁,類神經網路在引擎故障種類診斷上的應用,第二十屆機械工程研討會,台北市,國立台灣大學,民國92年12月5~6日。 【4】陳聖洲,汽車引擎之類神經模糊故障診斷,南台科技大學機械工程學系碩士論文,民國92年。 【5】Mitchell, S. J, “Machinery Analysis and Monitoring,” PennWell Publishing Company, Oklahoma, 1981. 【6】Eshleman, R. L, “Machinery Vibration Analysis,” The Vibration Institute, VIPress,1995. 【7】Eshleman, R. L. and Lewis, F. M., “Torsional Vibration in Reciprocating and Rotating Machines,” in Shock and Vibration Handbook, edutor Harris, C.M., McGraw-Hill,1988. 【8】Nurthadi, I., Bagiasna, K. and Wediyanto. “Signature Analysis of 4-Stroke 1-Cylinder Engine”,SAE Transaction,paper No.932011, 1993. 【9】DeBotton, G., Ben-Ari, J., Itzhaki, R. and Sher, E., “Vibration Signature Analysis as a Fault Detecion Method for SI Engine,” J. of Comml Veh. SAE paper No.980115 (Transaction), 1998. 【10】Macian, V. and Lerma, MJ, Barila., “Condition Monitoring of Thermal Reciprocating Engines Through Analysis of Rolling Block Oscillatuons,” SAE Transaction, paper No.980116, 1998. 【11】Jesion, G., Gierczak, C. A., Puskorius G. V., Feldkamp L. A., and Butler J. W., “The Application of Dynamic Neutral Networks to the Estimation of Feedgas Vehicle Emissions,” IEEE paper 0-7803-4859-1/98, pp. 69-73, 1998. 【12】D. Yuanwang, Z.Meilin,X.Dong,C.Xiaobei,An analysis for effert of cetane number on exhaust emissions from engine with the neural network,Fuel 81,2003,pp.1963-1970. 【13】王泗華、彭錦樵,“排氣背壓與廢氣排放診斷引擎故障之研究”, 行政院國家科學委員會專題研究計劃成果報告,計劃編號:NSC90-2313-B-005-133,2001. 【14】張嘉仁,“振動與噪音信號在引擎診斷上的應用”, 碩士論文,國立台北科技大學/車輛工程系,2002。 【15】葉懷仁,“類神經網路應用於引擎聲訊之辨識”, 碩士論文,國立中山大學/物理研究所,2003。 【16】張志瑋,“類神經網路於引擎故障診斷之研究”, 碩士論文,國立台北科技大學/車輛工程系,2003。 【17】Adnan, P.,Yasar, I.,Halit, Y.and Aysun,E., “Application of artificial neural network to predit specific fuel consumption and exhaust temperature for a Diesel engine” Applied Thermal Engineering 26, 2006, pp. 824-828. 【18】陳永耀,“以類神經網路模式來進行引擎故障診斷”, 碩士論文,國立中興大學/機械工程系,2006。 【19】王進德,“類神經網路與模糊控制理論入門”,全華圖書公司,民國90 年。 【20】K.A.et al Marko, K. A.‚ James, J., Dosdall, J. and Mirphy, J., “Automotive control system diagnosis using Neural Nets for rapid pattern classification of large data set”,IJCNN-89-San.,II,pp.13-16.1989. 【21】葉怡成,“類神經網路模式應用與實作”,儒林出版社,民國82年。 【22】VOLVO 引擎技術手冊,瑞典國寶富豪汽車有限公司。
摘要: 
摘 要
車輛引擎異常時常導致振動量的增加,本研究利用類神經網路將引擎振動信號經頻譜分析後擷取其異常值進行故障診斷。引擎為一複雜系統並包含非線性因素,若以理論數學解析進行故障診斷非常困難,而類神經網路具有解決高度非線性問題與學習記憶的能力,能夠達到對複雜系統的解析,因此類神經網路非常適合使用於引擎故障診斷。
本研究是以瑞典富豪汽車公司所生產的VOLVO 94O SE四行程之電子點火電腦控制噴射引擎,進行惰速定速運轉實驗,作測試故障的項目包括引擎各缸噴油嘴失效或異常、各缸高壓線鬆脫或火星塞不良、引擎腳鬆動、空氣芯阻塞、進氣管洩漏、以及噴油嘴與空氣芯同時異常、高壓線或火星塞與空氣芯同時異常等,利用訊號擷取儀器紀錄引擎的振動信號。引擎振動與其轉速有關,以振動頻譜中轉速頻率的倍頻振幅來表現故障的特徵。根據實驗取得引擎故障分類資料,建立引擎故障類型,作為倒傳遞診斷與類神經網路學習的依據。
由於類神經網路具有廣義的預測能力,能夠準確辨識未學習的相似引擎故障。因此在診斷時同樣使用所有故障的新範例來驗證學習後的網路。結果顯示已學習的網路不僅對引擎故障的種類能作辨別,且對故障程度也能作良好的診斷,這也證實了引擎中的某些故障種類在網路學習前是無需作複雜的多種程度量測。這種故障診斷的方式可以提供引擎維修人員作為參考依據,以及協助其診斷引擎故障,予以故障的排除,這也對於提昇新進技術維修員的訓練和修護能力而言,將大有助益。

Abstract
Engine always produces a large amount of vibration when engine in idel conditions. This research uses neural network to develop a fault diagnosis method by using features collected from spectral analysis of vibration signal of engine. Engine is a complicated system and has some nonlinear factors in general. It is very difficult to diagnose such system by mathematical analysis. The neural network has the capabilities of solving nonlinear problems, learning and memory. Therefore, the neural network is quite suitable for the purpose of engine fault diagnosis.
This research performed idle and steady speed experiments with Volvo 940SE equipped with electronic ignition and engine-control computer. The malfunction testing items include fuel nozzle malfunction, ignition malfunction, the mount of the engine becomes flexible, the air filter is blocked, the intake pipe is leaked, and the fuel nozzle and air filter are unusual at the same time, ignition and air filter are unusual at the same time, In the meantime, a signal collector is utilized to record vibration signal of an engine; engine vibrations are relevant to the engine speed. The amplitude of vibration both in a vibration spectrum displays features of malfunction. According to the experiments above, the acquire different data were used to establish types of engine malfunction, both of which are served as basis for anti-communicative diagnosis and neural network emulation.
Due to generalized predictability, of possessing data the neuron network can identify similar engine fault patterns. After the networks were learned and verified with new measured data and the results show that it not only can identify different types of fault but also can distinguish between degrees of fault levels. It also indicated that for certain types of engine fault it is even not necessary to measure many different fault levels for the network learning. The method of malfunction diagnosis may provide references for engine users and maintenance staff, assisting them in diagnosing and obviating engine malfunction, in terms of the training and maintenance ability of the newly recruited technical repairmen, their capability for exact and reasonable recognition of the fault types can be substantially promoted.
URI: http://hdl.handle.net/11455/2107
其他識別: U0005-2308200812492100
Appears in Collections:機械工程學系所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.