【正文】 取淹沒在強(qiáng)烈背景噪聲中的沖擊脈沖信號。從共振解調(diào)方法的運用過程[13]如下圖調(diào)制信號低通濾波包絡(luò)分析高通濾波原始信號圖23 共振解調(diào)過程在共振解調(diào)的過程包絡(luò)提取是一項最為主要也是最關(guān)鍵的一項工作。二是在發(fā)現(xiàn)異常情況后對設(shè)備的故障進(jìn)行分析、診斷。但是僅從表面上是很難了解其內(nèi)在規(guī)律的，最好的方法是對其進(jìn)行數(shù)學(xué)分析，因此我們需要建立一個齒輪故障實驗臺。我們知道傳感器將被測系統(tǒng)的機(jī)械振動轉(zhuǎn)換為電信號，但傳感器采集到的數(shù)據(jù)是模擬信號，而電腦所能直接處理的是數(shù)字信號，怎樣解決這樣的矛盾呢？數(shù)據(jù)采集卡能夠?qū)崿F(xiàn)由模擬信號到數(shù)字信號的轉(zhuǎn)換在實際工程實踐中，傳感器采集到的信號時非常的微弱的，并不像教科書上畫的那么明顯，有些信號如果直接用示波器來顯示，根本無法觀察到，所以需要放大器對信號進(jìn)行放大。綜上所述，系統(tǒng)總體方案設(shè)計如圖31所示上一節(jié)中概括性的講了一下實驗臺，傳感器，放大器和數(shù)據(jù)采集卡，計算機(jī)可以把這些東西統(tǒng)稱為硬件。其主要技術(shù)參數(shù)如下：量程： 電壓放大器的輸出信號受連接電纜對地電容的影響,易產(chǎn)生干擾信號,信號不穩(wěn)定,所以選擇電荷放大器。最大輸入電荷量：105PC 不準(zhǔn)確度：% 特點：通用型，高增益，高精度（4）數(shù)據(jù)采集卡數(shù)據(jù)采集卡的主要參數(shù)有通道數(shù)、位數(shù)和最高采樣頻率，它們是選擇數(shù)據(jù)采集卡時的重要指標(biāo)。分辯率：16Bit 模入范圍：177。觸發(fā)方式：定時器觸發(fā),軟件觸發(fā)，可任意設(shè)定采樣頻率 工作模式選擇是指在線監(jiān)測和離線分析兩種工作模式中選擇一種。這兩種方式相輔相成，對設(shè)備進(jìn)行故障監(jiān)測和診斷。信號分析是該軟件的核心功能。綜上所述，軟件總體設(shè)計方案框圖32所示 在NIDAQmx 中，任務(wù)是包括一條或多條通道以及定時、觸發(fā)等屬性的集合。（1） 新建一個空白VI 。將該Express VI放置到程序框圖上。圖32 軟件設(shè)計框圖圖33 數(shù)據(jù)采集控件（5） 選擇Voltage 創(chuàng)建一個新的電壓模擬輸入任務(wù)。 DAQ Assistant 將打開一個新對話框，如圖34所示。（9） 在Samples to Read 文本框中輸入1000任務(wù)創(chuàng)建完成以后，在前面板添加一個波形顯示控件，然后回到框圖用數(shù)據(jù)線連接DAQ assistant和波形顯示器便可以顯示采集到的數(shù)據(jù)的圖形。這些特性在信號的波形圖上表現(xiàn)的尤為明顯。運行之后，顯示器上出現(xiàn)正弦信號的頻譜。圖38 解調(diào)分析程序如圖39所示，模擬一個信號然后提取信號的一部分（在實際工程中采集到的信號的頻譜常有一直流成分，利用此方法可去除直流成分）測量其正峰值，然后將它與一可設(shè)定的常數(shù)（及閾值Xrms0）比較，然后添加一個條件結(jié)構(gòu)用來判斷比較結(jié)果并根據(jù)結(jié)果做出相應(yīng)的相應(yīng)。“讀取數(shù)據(jù)”控件也有一個類似的對話框。第4章 基于labview齒輪故障診斷實驗系統(tǒng) 故障實驗臺分析圖41 實驗臺示意圖如圖41所示，實驗臺將兩臺交流異步電動機(jī)通過一級齒輪傳動連接起來，其中一臺接電源，我們叫它電動機(jī)。（由斷齒故障最能代表齒輪故障信號的特征而且斷齒故障發(fā)生的頻率較高）兩種情況對比分析，驗證基于labview的齒輪故障診斷系統(tǒng)是否能夠準(zhǔn)確的進(jìn)行故障監(jiān)測和診斷。圖42 正常信號時域波形圖43 故障信號時域波形圖44是正常狀態(tài)下信號的頻譜，圖45是故障狀態(tài)時信號的頻譜。圖46 正常信號解調(diào)分析圖47 故障信號解調(diào)分析結(jié)論齒輪故障大體分為兩類，一是局部性故障。具體完成了以下工作：（1）通過研究齒輪故障信號的特征，選擇采用時域分析，頻譜分析，包絡(luò)分析對齒輪的故障信號進(jìn)行分析處理，取得了令人滿意的結(jié)果，證明采用這三種方法對齒輪進(jìn)行故障診斷是可行的；（2）對齒輪故障模擬實驗臺進(jìn)行了結(jié)構(gòu)設(shè)計，搭建了基于振動法的測試裝置，基于labview開發(fā)了數(shù)據(jù)采集模塊；（3）開發(fā)軟件采用了基于圖形化編程的labview軟件，給編程提供了極大的方便，很大程度的縮短了軟件開發(fā)的時間；（4）組建了一套齒輪故障診斷系統(tǒng)，通過齒輪斷齒故障的模擬和信號分析驗證了該系統(tǒng)的可行性。致謝在論文完成過程中，我的導(dǎo)師張洪鑫老師給了我可貴的支持和幫助，我的同學(xué)也提出了寶貴的意見，謹(jǐn)表示誠摯的感謝!參考文獻(xiàn)1 王立中 王 2 熊詩波 ：2772883 4 丁康 李巍華 :12 5 周建民 周佳 6 韓清凱 ：897 :12168 a case based dreasoningmethod for fault diagnosis during 9 陳長征 胡立新 :32332410 丁康 李巍華 朱小勇 :2511 extended FFT algorithm for ARMA spectral 12 續(xù)媛君 基于Labview的齒輪箱故障診斷研究與應(yīng)用13 RolfIsermann .Model based fault detection and diagnosis status and applications .200514 Nc machine tool fault diagnosis 15 周建民 周佳 .附錄NC machine tool fault diagnosisAbstractfault diagnosis technology has a history of more than 30 years of development, but as a prehensive new subject, diagnostics of fault is developed in recent years. From a different perspective, equipment fault diagnosis theory and method are many, the fault diagnosis expert system method is one of the most significant achievements in fault diagnosis field in recent years, its content including diagnosis knowledge expression, diagnosis reasoning method, uncertainty reasoning and diagnosis knowledge acquisition etc. key words：NC machine； tool fault； tree analysis 　A, numerical control machine tool fault diagnosis research is all about .　Fault diagnosis begins with mechanical equipment fault diagnosis, mainly refers to the manufacturing equipment and manufacturing process of condition monitoring and fault diagnosis. Manufacturing equipment is mainly refers to the processing machine tools, fixtures, measuring and cutting tools。 Japan absorption characteristics between them, and put forward the total productive maintenance (TPM) point of view. 　　Since the Apollo programme started in 1961 in the United States, appear a series of accidents caused by equipment failure, lead to the NASA (NASA) in 1967, under the sponsorship by the . naval research (ONR) host formed the mechanical failure prevention group (MFPG), and actively engaged in the development of diagnosis technology. The diagnosis technology in aviation, aerospace, military, nuclear and other cuttingedge department is still in the world leading position. 　　UK in the last century 60 s 70 s, in order to machine care and state monitoring association (MHMGamp。 (3) in view of the expert system knowledge acquisition bottleneck problem, more emphasis on the adaptive ability and self learning ability research, online diagnostic technique, multisensor technology research. 　　Three, numerical control machine tool fault diagnosis 　　Nc machine tool fault diagnosis has been a problem in operation and maintenance personnel. Due to the work safety and reliability of nc machine tool for the benefit of the production units directly to produce very big effect, and application in the field of fault diagnosis expert system, based on human experts experience knowledge of equipment and system fault diagnosis technology. 　　CNC machine tools as a plex nonlinear system, fully considering the natural conditions change and manmade wrong operation, how to bine the advantage of fuzzy technology and artificial intelligence, summed up the more intelligent fault diagnosis methods, will be need to efforts in the direction of the later. 　With further improve equipment automation, its fault diagnosis is also being more plicated, especially for engineering machinery, to solve all fault is very difficult in the process of operation. In view of this situation, the technical strength of scientific research institutes to establish a system of remote fault diagnosis through the Internet and engineering machinery field connections, establish a realtime fault detection system, found the fault during the process of operation in a timely manner and quickly diagnose. In fault diagnosis of local system cannot solve, use the Internet access to remote fault diagnosis center, and through technical strength of scientific research institutes to deal with these failures, in time to resume production, and effectively implement the technical resources sharing, therefore the remote fault diagnosis system based on Internet will be a importan