發(fā)布時(shí)間：2018-07-09 16:27

  本文選題：齒輪故障 + 數(shù)字圖像處理　； 參考：《武漢工程大學(xué)》2012年碩士論文

【摘要】：齒輪是機(jī)械設(shè)備中最常用的一種傳動(dòng)零件，在動(dòng)力的傳遞過(guò)程中起著重要的作用，其正常的工作狀態(tài)和可靠性是確保傳動(dòng)系統(tǒng)效率的有力保障。目前，對(duì)齒輪進(jìn)行故障診斷的常用技術(shù)包括傳統(tǒng)診斷技術(shù)和智能診斷技術(shù)，如振動(dòng)監(jiān)測(cè)技術(shù)、噪聲監(jiān)測(cè)技術(shù)、油液分析、紅外熱成像、無(wú)損探傷、聲發(fā)射技術(shù)和神經(jīng)網(wǎng)絡(luò)、模糊邏輯、專家系統(tǒng)、遺傳算法等，它們?cè)邶X輪運(yùn)行狀態(tài)監(jiān)測(cè)和故障診斷中均有著重要的應(yīng)用。隨著圖像處理技術(shù)的不斷發(fā)展，在機(jī)械工業(yè)中的應(yīng)用也越來(lái)越廣泛。為此，本文提出了一種基于數(shù)字圖像處理技術(shù)的故障診斷研究思路，將其應(yīng)用到齒輪磨損故障診斷中進(jìn)行了研究，為齒輪磨損檢測(cè)提供了一種全新的手段，實(shí)現(xiàn)了通過(guò)磨損圖像識(shí)別的方法進(jìn)行齒輪磨損故障的判別。 本文的研究工作以建立可供實(shí)際應(yīng)用的齒輪磨損故障判別系統(tǒng)為目標(biāo)，分為五大部分：首先對(duì)齒輪圖像進(jìn)行了預(yù)處理研究，通過(guò)對(duì)圖像中存在的各種失真和噪聲進(jìn)行了處理，獲得了一幅干凈且目標(biāo)特征清晰的圖像；其次是對(duì)齒輪圖像的邊緣檢測(cè)和提取算法進(jìn)行了研究，通過(guò)比較分析選擇最優(yōu)算子，提取出了能夠真實(shí)反映實(shí)際輪齒邊緣的齒廓曲線；再對(duì)圖像進(jìn)行離散化處理，將通過(guò)解析式獲得的標(biāo)準(zhǔn)齒輪齒廓與通過(guò)圖像識(shí)別獲得的實(shí)際齒廓進(jìn)行了對(duì)比研究，得出了齒輪磨損量；然后為了獲得齒輪磨損量的閾值，進(jìn)行了齒輪磨損實(shí)驗(yàn)；最后對(duì)故障診斷系統(tǒng)進(jìn)行了開發(fā)，利用相關(guān)軟件建立了一套可視化的齒輪磨損故障診斷系統(tǒng)。 本文研究?jī)?nèi)容涉及到多種數(shù)字圖像處理技術(shù)和系統(tǒng)開發(fā)技術(shù)，研究思路與方法比較新穎，目前國(guó)內(nèi)進(jìn)行的相關(guān)研究較少，因此具有一定的理論意義和創(chuàng)新價(jià)值。該方法豐富了齒輪故障診斷技術(shù)，，為齒輪零件的磨損故障診斷提供了一個(gè)新的方向，具有極大的現(xiàn)實(shí)意義。
[Abstract]:Gear is one of the most commonly used transmission parts in mechanical equipment, which plays an important role in the transmission of power. Its normal working state and reliability are the effective guarantee to ensure the efficiency of transmission system. At present, the commonly used technologies for gear fault diagnosis include traditional diagnosis technology and intelligent diagnosis technology, such as vibration monitoring technology, noise monitoring technology, oil analysis, infrared thermal imaging, non-destructive flaw detection, acoustic emission technology and neural network. Fuzzy logic, expert system, genetic algorithm and so on, they have important applications in gear running state monitoring and fault diagnosis. With the development of image processing technology, it is widely used in machinery industry. Therefore, this paper puts forward a research idea of fault diagnosis based on digital image processing technology, and applies it to gear wear fault diagnosis, which provides a new method for gear wear detection. The fault identification of gear wear is realized by the method of wear image recognition. The aim of this paper is to set up a gear wear fault discrimination system for practical application, which is divided into five parts: firstly, the preprocessing of gear image is carried out, and the distortion and noise in the image are processed. A clean image with clear target features is obtained. Secondly, the edge detection and extraction algorithm of gear image is studied. By comparing and selecting the optimal operator, the tooth profile curve which can truly reflect the actual tooth edge is extracted. Then the image is discretized, the standard gear tooth profile obtained by analytic formula is compared with the actual gear profile obtained by image recognition, the gear wear quantity is obtained, and then in order to obtain the gear wear threshold, Finally, the fault diagnosis system is developed, and a set of visual gear wear fault diagnosis system is established by using related software. The research content of this paper involves a variety of digital image processing technology and system development technology, the research ideas and methods are relatively new, at present, there are few related studies in China, so it has certain theoretical significance and innovative value. This method enriches gear fault diagnosis technology and provides a new direction for gear wear fault diagnosis, which has great practical significance.
【學(xué)位授予單位】：武漢工程大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2012
【分類號(hào)】：TH132.41;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 何毅斌;陳定方;張娟;;VB與Matlab混合編程的圖像處理研究[J];電腦開發(fā)與應(yīng)用;2008年11期

2 徐獻(xiàn)靈;林奕水;;圖像邊緣檢測(cè)算法比較與分析[J];自動(dòng)化與信息工程;2007年03期

3 李捷;唐星科;蔣延軍;;幾種邊緣檢測(cè)算法的比較研究[J];信息技術(shù);2007年09期

4 厲丹;錢建生;王超;;圖像復(fù)原技術(shù)的研究[J];計(jì)算機(jī)工程與應(yīng)用;2010年25期

5 萬(wàn)力,易昂,傅明;一種基于Canny算法的邊緣提取改善方法[J];計(jì)算技術(shù)與自動(dòng)化;2003年01期

6 張新華,魯志康,李新榮;齒輪故障自動(dòng)監(jiān)測(cè)的頻譜分析和研究[J];機(jī)械傳動(dòng);2000年02期

7 李佳;禮賓;王夢(mèng)卿;;基于神經(jīng)網(wǎng)絡(luò)的齒輪故障診斷專家系統(tǒng)[J];機(jī)械傳動(dòng);2007年05期

8 芮執(zhí)元;徐龍?jiān)?馮瑞成;任麗娜;;基于小波神經(jīng)網(wǎng)絡(luò)(WNN)的齒輪故障診斷[J];機(jī)械傳動(dòng);2008年01期

9 韓捷,張琳娜;齒輪故障的振動(dòng)頻譜機(jī)理研究[J];機(jī)械傳動(dòng);1997年02期

10 張新華;齒輪故障識(shí)別[J];機(jī)械強(qiáng)度;1999年04期

本文編號(hào)：2109944