帶減速器的電機(jī)裝置驅(qū)動(dòng)設(shè)計(jì)
帶減速器的電機(jī)裝置驅(qū)動(dòng)設(shè)計(jì),減速器,電機(jī),機(jī)電,裝置,驅(qū)動(dòng),設(shè)計(jì)
Original text：
X-RAY DETECTS WELDS AND MECHANICAL STRUCTURE DAMAGE MATHODS’SUMMARIZE
The moving small object detection in image is always a difficult problem in field of image processing, which applies in many fields, such as industrial detection and medical detection. The defects, such as blowholes and incomplete penetration, occasionally appear in the welding process. These defects can affect the quality and the security of products. Therefore, defects detection in welding seam is extremely important. Now, the on-line detection of defects in the weld is still done by human interpreter. However, this process is subjective, inconsistent, labor intensive and fatigue of interpreter. It is desirable to find an effective automatic defects detection method to assist human interpreter in evaluating the quality of weld and to make the on-line detection objective, standard and intelligent. Our research is based on this.
We have studied the automatic defects detection in the weld seam and mainly done the following research :(1) There is much redundant background information for the defects detection in the image. Therefore we use an automatically abstracting method of weld area based on the auto- adapted threshold segmentation. This method can reduce the computation and increase the precision. (2) The SUSAN algorithm has good anti-noise ability, which can recognize the image edge very well. So we have studied a defects detection method based on SUSAN algorithm, which associated with the morphology operation. The results indicate that this method is effective.(3) Wavelet analysis method has a very good localization characteristic, which can focus on the arbitrary detail of the analyzed object. Therefore, we studied a method using wavelet decomposition to get the shape and position information of the defects. Then we use the wiener filter and morphology method to complete the detection.
The automatic flaw detection of welded tubes is one of the most important steps to ensure the quality of the tubes. Nondestructive inspection on welding seam of tube is required in the tube production, and real time X-Rayradiography inspection is an effective means. Along with continuous improvement of the productive ratio, the demand for the automatic inspection to the welding seam becomes more and more pressing, so implementation of the automatic inspection possesses important significance on both theory and reality. Wavelet transform is a powerful tool in the signal and image processing, and its fundamental theory has been formed. From the view of engineering applications, however, the wavelet transform is still in the elementary stage, the further
Researches are required for the practical uses. In this thesis, we concentrate mainly on using wavelet analysis for welding seam image processing and recognition, and some related techniques are developed.
For constructing welding seam positioning and detection control system, the multiple computers configuration for weld seam image recognition is proposed. The system adopts the architecture in which multiple CPUs process parallels under the control of the master IPC computer. The system can perform storing the weld seam images, positioning, flaws recognizing and quality prejudging. The Watch-Dog interface card is successfully developed; it can improve the system reliability by redundancies technique of saving breakpoint data and restoring them.
The hardware supporting the system makes use of the high speed digital signal processor TMS320C30 from Tax ax Instruments Company. The frame grabber can capture 25 frames of welding seam image per second continuously and make it possible to fulfill the real time welding seam image processing Andre cognition. The one kind of improved FWT (Fast Wavelet Transform) algorithm for a finite sequence is proposed after studying theory of mustier solution analysis
and analyzing technical characteristics of DSP. The implementation of the
periodic extension of the FWT on DSP is described in detail and the corresponding FWT assembly code is described for the DSP TMS320C3X series.
This dissertation suggests scheme of image demonizing based on two-dimensional discrete wavelet transform. The demonizing algorithm is described with some operators. By threshold the wavelet transform coefficients’ of noisy images, the original image can be reconstructed correctly. Different threshold selections and threshold methods are discussed. The new robust local threshold scheme is proposed. Quantifying the performance of image demonizing schemes by using the mean square error, the performance of the robust local threshold scheme
is demonstrated and is compared with the universal threshold scheme. The experiment shows that image demonizing using the robust local threshold performs better than that using the universal threshold.
In order to improve the accuracy and the real time performance of edge detection, a method need to be found to match the detection of low contrast blurred welding seam image. This dissertation analyzed the main sources of noise as well as the different characteristics of noise and signal under wavelet transform, and proposed a Moultrie solution edge detection method based on wavelet transform.
The experimental results show the effect of this algorithm is advantageous over that of traditional edge detection algorithm.
The geometrical relation of elliptic imaging is studied for welding seam image of the butt welds in straight tubes. The region model of welding seam image is proposed, It furnishes a evidence theory to further process to welding seam image. Combining with the region model, a model-based adaptive target segmentation algorithm is proposed. One basis of the algorithm is Otsu's discriminates criterion. The adaptive target segmentation of welding seam image is realized. The effect of target image segmentation is quite well.
The difficult problem of target flaw automatic recognition in welding seam image is analyzed. Using for reference the consciousness organizing process of the human vision system, a knowledge-based target recognition algorithm with mufti-feature fusion, mufti-window architecture and mustier solution is presented. With the help of certain prior knowledge, criteria and means of artificial intelligence, target flaws are extracted and recognized quite well. It is a Prospecting intelligent recognition algorithm.
The fast feature extraction algorithm for target geometrical feature is proposed. This algorithm is different from usual feature extraction methods which first need to change a gray image into binary image. The algorithms proposed get the feature of a image in the gray image directly. Using this algorithm can fast extract features of target flaws in welding seam image.
All kinds of mechanical devices and structure tend to become large-scale and high efficient with the industry developing and progress of science and technology. The mechanical devices and structure become very complex to meet the need of industry. The structure or devices are damaged is couldn't avoided during working under complex load and working for a long time. The loss caused by crash, fatigue, eroding and wear is about 6%一8% of GDP of USA and Japan. In our country, accident number of structural damage is 10 times as many as that in, intense industrialization country in eighties of last century. In 1986, the loss is 12 hundred million $ caused by the space shuttle named '`challenger" of U.S.A. crashed. In 1985，the accident cause of joint of electromotor set of Datong power plane crashed, In 1988, the accident cause of main beam of electromotor set of Qianlong power plane crashed, those accident cause of loss near 1 hundred million RMB. In our country, 6 serious accidents ware been caused by rotor of over 50Mw electromotor set damaged badly during 1984 to 1991.Therefore, the study the theory and technique about large scale and complex mechanical devices and structure online inspect and early fault diagnosis is urgent task. Especially, how to detect the fault of structure as early as possible is engineers most want to do. But it is very difficult that faint signal produced by early fault is recognized.
The research reports of our country and overseas show that at present the study of structure damage inspect by vibration characteristics carry out most on the simple and symmetry structure just as beam and frame etc. and result is given based on finite element numeric calculation. But the structure and join of practical device is quite complex, it is impossible modeling the practical device reliably by FE. So that realizing complex industrial device inspect online and early fault diagnosis by using vibrate testing technique is a problem that wants to be solved urgently.
Fault diagnosis technique is intercross subject. Especially, the base in theory of fault diagnosis of complex system is provided based on modern control theory, signal processing, pattern recognition, optimum method, decision-making and manual intelligent are developed rapidly.
Structure damage detection is a research project that has wide background of industrial application. But realizing large-scale and complex structure damage inspect online is depend on techniques such as development of accurate testing technique and signal processing method, based on getting to the best advantage mix model of structure damage detection, the sensors escape placed on structure reasonably and optimally
The large-scale vibrating device as a researched object, the method structure damage detection is studied. In general NDT technique such ultrasonic test, ray test, magnetism applied in modeling offline mostly. The project test and scheme pervade test etc. are vibrate properties and structure damage characteristics from platitudinous: offline tests and analysis to structure as impotent information of online automatic fault diagnosis database. Then the method of few-testing, points modeling to getting structure damage information has beer researched. Placing sensors reasonably and realizing large-scale an Complex structure damage inspect online are targets of this project.
The large-scale vibrating screen has been applied widely in coax industry and other industrial areas as a kind of important device. A: vibration mechanical device, it works very hardly and works in verb wretched environment so that the beam of screen is damaged easily
Therefore, it is very important how to detect the fault of beam as early a possible to make the repair schedule reasonably and economically and to avoid the body hurt and device damage.
In this thesis, how can locate a damaged beam of screen is studied serially. The regulation of beam vibration characteristics change depend on damage degree of beam is found. Also the regulation of whole screen vibration characteristics change depend on damage degree of beam is found too. Based on deep research about beam vibration characteristics change regulation and whole screen vibration characteristics change regulation in series, we can get the optimum place to placed sensors for location which beam is damaged.
The target of the thesis is combine the on-line dynamically inspect screen for structure damage with accurately locate fault by acoustic emission technique.
The main content of this thesis consist of (1) Based on modal parameters recognition of whole screen, get location of damaged substructure. (2) Locate fault of substructure accurately by acoustic emission technique. (3)Carry on research about finding a efficient way we can inspect screen for structure damage on-line.
These projects are done step by step. At first, free-free beam vibration characteristics are studied deeply. The first rank and second rank bending vibration modal shape of beam are abstracted as research objects. The study result is shown that the frequencies of FRF peak value drift toward lower frequency and the amplitudes of FRF peak value increase with the damage degree of beam. Then the first rank and second rank bending vibration characteristics of beam fixed on screen are studied. The change regulations of characteristics of FRF with beam damage are agreement to that of free-free beam. Therefore the damage information of beam can be gotten from FRF. The wavelet packet analysis method and spectral analysis calculation method are employed in frequency response and transmissibility processing. The fault characteristics are abstracted. After then, the damaged beam has an effect on whole screen vibration characteristics are researched. From above work, the damaged beam of screen could be located from whole screen. Then the acoustic emission technique is used to locate fault of the damaged beam accurately. Because the too many sensors couldn't place on the working screen so that we must find limited positions to place sensors getting enough structure damage information. At last, the method of the finding optimum places to placed sensors for location which beam is damaged is studied. The efficient way of optimum place to placed sensors is found.
In this thesis, the different spectral analysis calculation methods are employed in vibration signal processing to abstract fault characteristics. The processing result indicates that methods of vibration signal procession are efficiently.
This thesis provides some realizable ways to realize the on-line dynamically inspect screen for structure damage.
Translation:
X射線探測(cè)焊縫及機(jī)械損傷方法概述
圖像中運(yùn)動(dòng)小目標(biāo)的檢測(cè)一直是圖像處理與分析領(lǐng)域中的難題，它涉及到很多領(lǐng)域，具有很廣泛的研究?jī)r(jià)值和應(yīng)用價(jià)值。在工業(yè)探傷領(lǐng)域，由于焊接過程出現(xiàn)的各種問題，會(huì)導(dǎo)致焊縫中含有氣孔和未焊透等缺陷，影響產(chǎn)品的質(zhì)量和安全，所以焊接圖像中缺陷目標(biāo)的檢測(cè)十分重要。目前X射線無損探傷系統(tǒng)主要采用人工方式進(jìn)行在線檢測(cè)與分析，而人工檢測(cè)存在主觀標(biāo)準(zhǔn)不一致、勞動(dòng)強(qiáng)度大等缺點(diǎn)。因此，急需要研究一種有效的缺陷自動(dòng)檢測(cè)方法來代替人工檢測(cè)，從而使在線檢測(cè)工作客觀化、規(guī)范化和智能化。本文的研究工作就是基于此而展開的。
本文探討了焊縫圖像中缺陷目標(biāo)的自動(dòng)檢測(cè)方法，主要做了以下幾個(gè)方面的研究:(1)針對(duì)X射線焊縫檢測(cè)圖像中存在大量與缺陷檢測(cè)無關(guān)的背景冗余信息，采用了一種基于自適應(yīng)閉值分割的焊縫區(qū)域的自動(dòng)提取方法，以減少計(jì)算量，提高檢測(cè)精度，取得了較好的效果。(2)由于SUSAN算法具有良好的抗噪能力，對(duì)圖像的邊緣、角點(diǎn)能夠很好的識(shí)別，所以本文研究了一種以SUSAN算法為基礎(chǔ)的，焊縫缺陷自動(dòng)檢測(cè)算法，同時(shí)輔助以形態(tài)學(xué)去噪和填充等運(yùn)算，取得了較好的效果。(3)因?yàn)樾〔ǚ治龇椒ň哂泻芎玫木植炕匦?，它能?duì)高頻采取逐漸精細(xì)的時(shí)域或空域步長(zhǎng)，從而可以聚焦到分析對(duì)象的任意細(xì)節(jié)。所以研究了一種利用小波分解來得到缺陷目標(biāo)的形狀和位置信息，并結(jié)合維納濾波和形態(tài)學(xué)運(yùn)算的焊縫缺陷檢測(cè)方法，結(jié)果比較理想。為了驗(yàn)證本文提出的兩種算法的有效性，本文對(duì)在工廠實(shí)際得到的含有缺陷目標(biāo)的焊接圖像進(jìn)行了檢測(cè)，取得了較好的效果，證明了本文方法的可行性。
焊管缺陷的自動(dòng)檢測(cè)是保證鋼管產(chǎn)品質(zhì)量的重要環(huán)節(jié)。在鋼管生產(chǎn)過程中需要對(duì)焊管焊縫進(jìn)行無損檢測(cè)，X射線實(shí)時(shí)成象檢測(cè)是一種比較有效的檢測(cè)手段。隨著生產(chǎn)率的不斷提高，對(duì)焊管焊縫的自動(dòng)化X射線檢測(cè)要求越來越迫切，實(shí)現(xiàn)焊管焊縫的自動(dòng)化檢測(cè)具有重要的理論意義和實(shí)際意義。小波變換作為信號(hào)和圖象處理的一種強(qiáng)有力的工具，其理論框架己基本形成，但從工程應(yīng)用的角度，小波變換技術(shù)還處于初級(jí)階段，還需進(jìn)一步完善。.本文主要研究小波分析技術(shù)如何用于焊縫圖象處理與識(shí)別以及一些相關(guān)技術(shù)。
為建立焊管焊縫自動(dòng)定位檢測(cè)控制系統(tǒng)，提出了焊縫圖象識(shí)別的多機(jī)系統(tǒng)結(jié)構(gòu)方案，該系統(tǒng)采用多處理器并行處理的結(jié)構(gòu)，并由上位機(jī)工PC協(xié)調(diào)控制管理。系統(tǒng)能完成對(duì)焊縫圖象的存貯、焊縫定位、缺陷識(shí)別和質(zhì)量評(píng)定。并成功地研制了基于工SA總線的Watch-Dog接口板，使用冗余法進(jìn)行斷點(diǎn)數(shù)據(jù)存儲(chǔ)和恢復(fù)，實(shí)現(xiàn)了系統(tǒng)的可靠運(yùn)行。
硬件系統(tǒng)使用了Taxax儀器公司的高速信號(hào)處理器'I'MS320C30。圖象采集卡能每秒連續(xù)采集25幀焊縫圖象，使得實(shí)時(shí)完成焊縫圖象處理與識(shí)別成為可能。在充分研究多分辨分析理論和分析信號(hào)處理器技術(shù)特點(diǎn)的基礎(chǔ)上，針對(duì)DSP TMS320C3X的特點(diǎn)，提出了一種有限序列的FWT(快速小波變換)的改進(jìn)算法，詳細(xì)闡述了信號(hào)處理器上FWT的周期性擴(kuò)展的實(shí)現(xiàn)問題，用DSP TMS320C3X匯編語言實(shí)現(xiàn)了改進(jìn)的FWT算法。
通過對(duì)小波變換系數(shù)進(jìn)行閡值處理，給出了一種基于二維離散小波變換的圖像去噪方法并用算子的形式加以描述。討論了幾種閡值選取方法和閡值策略，并提出了一種魯棒局部閉值去噪法。用均方差衡量去噪性能，實(shí)驗(yàn)表明用魯棒局部閉值去噪法好于全局閉值去噪法。
為提高邊緣檢測(cè)的準(zhǔn)確性和實(shí)時(shí)性，需要尋找一種適合于低對(duì)比度模糊焊縫圖象邊緣檢測(cè)的快速方法。本文分析了焊縫圖象的主要噪聲來源及噪聲與信號(hào)在小波變換下呈現(xiàn)的不同特點(diǎn)，提出了一種基于小波變換的多分辨率邊緣檢測(cè)方法。實(shí)驗(yàn)表明該算法的邊緣檢測(cè)效果明顯優(yōu)于經(jīng)典的邊緣檢測(cè)方法。 針對(duì)具體的鋼管直管對(duì)接焊縫圖象，研究了其橢圓成象的幾何關(guān)系，提出了焊縫圖象區(qū)域模型，為進(jìn)一步處理焊縫圖象提供了理論依據(jù)。提出了一
種模型基多分辨率圖象自適應(yīng)分割算法。該算法以O(shè)tsu判別準(zhǔn)則為基礎(chǔ)，結(jié)合焊縫區(qū)域模型進(jìn)行焊縫圖象的自適應(yīng)目標(biāo)分割，具有較好的分割效果。
研究了在焊縫圖象中目標(biāo)缺陷的自動(dòng)識(shí)別這一難題，在借鑒人類視覺系統(tǒng)知覺組織過程的基礎(chǔ)上，提出了一種基于知識(shí)的多特征融合多窗口結(jié)構(gòu)多分辨率目標(biāo)識(shí)別算法。該算法依據(jù)一定的先驗(yàn)知識(shí)和準(zhǔn)則，輔以人工智能的手段，能夠得到較為精確的目標(biāo)識(shí)別結(jié)果，是一種極有前途的智能識(shí)別算法。
提出了一種幾何特征快速提取算法，該算法改變了通常先圖象二值化后提取目標(biāo)參數(shù)特征的做法，而是直接對(duì)灰度圖象進(jìn)行目標(biāo)參數(shù)特征提取。使用本文提出的幾何特征快速提取算法可以有效地實(shí)現(xiàn)缺陷目標(biāo)的快速識(shí)別處理。
隨著生產(chǎn)的發(fā)展與科學(xué)技術(shù)的進(jìn)步，各類機(jī)械設(shè)備和結(jié)構(gòu)向著大型、高效化發(fā)展，從而也使得這些機(jī)械設(shè)備和結(jié)構(gòu)趨于復(fù)雜化。復(fù)雜的承載條件與長(zhǎng)時(shí)間的連續(xù)工作，導(dǎo)致設(shè)備結(jié)構(gòu)的損傷不可避免。因斷裂、疲勞、腐蝕和磨損而造成的破壞，其損失達(dá)美、日等國家每
年國民經(jīng)濟(jì)總值的6%^' 8%。而在我們國家20世紀(jì)80年代的結(jié)構(gòu)損傷事故率比工業(yè)化國家高10倍，人員累計(jì)傷亡居國內(nèi)勞動(dòng)安全事故第二位。1986年，美國的“挑戰(zhàn)者，，號(hào)航天飛機(jī)失事?lián)p失高達(dá)12億美元;蘇聯(lián)的切爾諾貝利核電站的核泄漏事故，對(duì)整個(gè)地區(qū)的人員、
生態(tài)環(huán)境都是無法估量的損害。1985年我國大同電廠一機(jī)組聯(lián)軸器斷裂事故、1988年秦嶺電廠機(jī)組主軸斷裂，造成的經(jīng)濟(jì)損失均近億元，并嚴(yán)重影響華北和西北地區(qū)供電。從1984年到1991年，我國50MW以上的汽輪發(fā)電機(jī)組轉(zhuǎn)子嚴(yán)重?fù)p壞等重大事故就達(dá)6起。因此，研究防止這類事故發(fā)生的根本途徑一一大型復(fù)雜機(jī)械結(jié)構(gòu)的健康狀況監(jiān)測(cè)與故障診斷(尤其是早期故障診斷)的理論和技術(shù)，實(shí)現(xiàn)結(jié)構(gòu)損傷的早期識(shí)別，及時(shí)采取措施，防止損傷的發(fā)展，以保證這些系統(tǒng)安全、可靠、長(zhǎng)壽命、高效率地運(yùn)行成為緊迫的任務(wù)。然而，對(duì)于大型的復(fù)雜機(jī)械結(jié)構(gòu)，實(shí)現(xiàn)合理、優(yōu)化地布置傳感器，監(jiān)測(cè)其運(yùn)行結(jié)構(gòu)的局部損傷，從中識(shí)別出損傷的結(jié)構(gòu)件及損傷狀態(tài)，進(jìn)行早期故障預(yù)報(bào)是具有相當(dāng)?shù)碾y度的。 國內(nèi)外的研究報(bào)道顯示，由于結(jié)構(gòu)振動(dòng)模態(tài)參數(shù)對(duì)不同損傷各有其敏感性，加之大型復(fù)雜結(jié)構(gòu)振動(dòng)模態(tài)識(shí)別技術(shù)發(fā)展的限制，以振動(dòng)特性為參數(shù)對(duì)結(jié)構(gòu)進(jìn)行損傷檢測(cè)的研究大多集中在諸如直梁、析架等簡(jiǎn)單對(duì)稱結(jié)構(gòu)，且多是基于有限元分析數(shù)值計(jì)算得出相應(yīng)的結(jié)論。而實(shí)際設(shè)備的構(gòu)造及聯(lián)接都是相當(dāng)復(fù)雜的，無法實(shí)現(xiàn)可靠的有限元建模。如何實(shí)現(xiàn)用振動(dòng)測(cè)試技術(shù)對(duì)實(shí)際復(fù)雜上業(yè)結(jié)構(gòu)的健康狀況進(jìn)行監(jiān)測(cè)與故障的在線診斷還是一個(gè)急待解決的問題。
故障診斷技術(shù)具有很強(qiáng)的學(xué)科交叉性，尤其是現(xiàn)代控制理論，信號(hào)處理，模式識(shí)別、最優(yōu)化方法、決策論、人工智能等的迅速發(fā)展，為解決復(fù)雜系統(tǒng)的故障診斷問題提供了理論基礎(chǔ)，形成了許多具體的方法。故障信號(hào)的特征提取為故障準(zhǔn)確診斷的前提條件。近三十年來，
各類機(jī)械設(shè)備基于實(shí)時(shí)監(jiān)測(cè)(包括振動(dòng)監(jiān)測(cè))的故障診斷技術(shù)的研究和應(yīng)用，促進(jìn)了故障信號(hào)處理和特征提取技術(shù)的發(fā)展。這些技術(shù)包括時(shí)域信號(hào)波形分析和統(tǒng)計(jì)特征值提取;基于FFT分析的高斯平穩(wěn)隨機(jī)信號(hào)現(xiàn)代譜分析技術(shù):自功率譜和互功率譜、高階譜、倒譜、復(fù)倒譜以及譜嫡與極大嫡譜估計(jì)(也包括與之相對(duì)應(yīng)的自相關(guān)與互相關(guān)函數(shù)、高階自相關(guān)函數(shù)等時(shí)域分析);非平穩(wěn)信號(hào)的時(shí)頻分析和多元統(tǒng)計(jì)分析等。故障診斷技術(shù)還涉及到材料的選擇、制造工藝、結(jié)構(gòu)設(shè)計(jì)、斷裂力學(xué)等多種學(xué)科和專業(yè)技術(shù)領(lǐng)域。隨著力學(xué)、材料科學(xué)、物理學(xué)、化學(xué)領(lǐng)域的學(xué)科交叉與發(fā)展，可從缺陷的背景和損傷、斷裂機(jī)制來研究從材料變形、損傷到失效的全過程。而計(jì)算機(jī)數(shù)據(jù)處理、模式識(shí)別的技術(shù)發(fā)展為與早期故障相關(guān)的微弱信號(hào)的捕捉和提取，提供了有利的手段。
結(jié)構(gòu)損傷檢測(cè)是一個(gè)具有廣闊工程應(yīng)用背景的研究課題，而大型復(fù)雜結(jié)構(gòu)損傷檢測(cè)的研究離實(shí)際應(yīng)用還有距離，還有一許多問題需要在今后的研究中加以解決:如精確的測(cè)量信息處理技術(shù)的發(fā)展，以期獲得更加精確的測(cè)量模態(tài);基于吸收各種方法優(yōu)點(diǎn)的混合模型的結(jié)構(gòu)損傷檢測(cè)方法的研究;大型設(shè)備的工況監(jiān)測(cè)，傳感器的合理布置與優(yōu)化配置問題的具體應(yīng)用;其它理論方法的引入，如模糊數(shù)學(xué)的應(yīng)用以及子結(jié)構(gòu)振動(dòng)分析方法的應(yīng)用等等。
本課題以大型振動(dòng)機(jī)械為研究對(duì)象，進(jìn)行結(jié)構(gòu)損傷檢測(cè)方法研究。常規(guī)的無損檢測(cè)方法，如超聲波、射線以及磁力探傷與滲透法探傷等大多是用于生產(chǎn)過程中間環(huán)節(jié)的零件離線檢測(cè)和設(shè)備檢修，通常為靜態(tài)檢測(cè)。本項(xiàng)目研究的基本思路是對(duì)大型振動(dòng)機(jī)械進(jìn)行多測(cè)點(diǎn)建
模，利用振動(dòng)測(cè)試技術(shù)進(jìn)行充分的離線試驗(yàn)和分析來獲取被診斷結(jié)構(gòu)的振動(dòng)特性細(xì)節(jié)、故障機(jī)理及其特征，作為結(jié)構(gòu)動(dòng)力學(xué)本質(zhì)特征庫的先驗(yàn)知識(shí)與在線自動(dòng)故障診斷信息庫的重要內(nèi)容;研究以結(jié)構(gòu)的少測(cè)點(diǎn)獲取結(jié)構(gòu)損傷信息的建模方法，合理配置傳感器，實(shí)現(xiàn)對(duì)大型振動(dòng)
結(jié)構(gòu)進(jìn)行健康狀況的在線監(jiān)測(cè)。通過振動(dòng)特性的微小變化，發(fā)現(xiàn)與定位結(jié)構(gòu)早期損傷，離線實(shí)現(xiàn)結(jié)構(gòu)損傷細(xì)節(jié)的分析與估計(jì)。上述思想是本項(xiàng)目研究的創(chuàng)新性思維。
課題具體內(nèi)容包括:通過對(duì)整體結(jié)構(gòu)的模態(tài)參數(shù)識(shí)別，判斷損傷的結(jié)構(gòu);用聲發(fā)射技術(shù)對(duì)損傷的結(jié)構(gòu)進(jìn)行損傷細(xì)節(jié)分析;研究對(duì)大型振動(dòng)機(jī)械進(jìn)行健康狀況監(jiān)測(cè)的方法。
大型直線振動(dòng)篩是洗煤廠的主要設(shè)備之一，作為振動(dòng)機(jī)械，不僅工作強(qiáng)度大，且工作環(huán)境十分惡劣，連續(xù)工作使其結(jié)構(gòu)極易產(chǎn)生疲勞斷裂。本課題的研究目標(biāo)是采用對(duì)振動(dòng)篩的結(jié)構(gòu)進(jìn)行系統(tǒng)動(dòng)力學(xué)健康狀況監(jiān)測(cè)與聲發(fā)射技術(shù)離線檢測(cè)相結(jié)合的方式確定結(jié)構(gòu)損傷的部位
與損傷的程度。
下橫梁是振動(dòng)篩的主要承載結(jié)構(gòu)件，也是易損傷結(jié)構(gòu)。本文從振動(dòng)篩的損傷下橫梁識(shí)別入手，研究大型機(jī)械結(jié)構(gòu)的早期故障診斷方法。
具體研究方法是:
研究近自由狀態(tài)的下橫梁的基本振動(dòng)特性，提取前兩階彎曲模態(tài)振型為研究基礎(chǔ)。系統(tǒng)研究了不同程度損傷所引起這兩階模態(tài)頻率的相應(yīng)變化以及頻響函數(shù)幅值的變化規(guī)律，獲得了梁的損傷程度與模態(tài)頻率以及頻響函數(shù)幅值間變化規(guī)律的經(jīng)驗(yàn)公式;
研究在連接約束狀態(tài)下的下橫梁的基本振動(dòng)特性，發(fā)現(xiàn)在振動(dòng)篩整體結(jié)構(gòu)中，下橫梁的這兩階彎曲模態(tài)振型仍然存在，只是模態(tài)頻率產(chǎn)生向低頻方向的移動(dòng)，且振型略有變異，但彎曲模態(tài)的主要特征依然存在。因此，以此兩階彎曲模態(tài)振型為研究基礎(chǔ)，研究損傷所引起
這兩階模態(tài)頻率的相關(guān)變化以及頻響函數(shù)幅值的變化規(guī)律，提出了相應(yīng)的經(jīng)驗(yàn)公式;
基于對(duì)振動(dòng)篩的子結(jié)構(gòu)一一下橫梁及其損傷特征的充分研究，進(jìn)一步研究下橫梁損傷對(duì)振動(dòng)篩的整體振動(dòng)特性的影響。發(fā)現(xiàn)在特定的頻率區(qū)間(這里是20Hz-30Hz，含損傷下橫的振動(dòng)篩的頻響函數(shù)非常明顯地向低頻方向移動(dòng)。對(duì)此，作者做了三項(xiàng)工作:(1)對(duì)頻響函數(shù)在這段頻率范圍的變化規(guī)律做了詳細(xì)分析;(2)用小波分析與功率譜分析對(duì)時(shí)域振動(dòng)加速度信號(hào)進(jìn)行了處理，提取損傷特征;③選取特定的參考點(diǎn)，研究振動(dòng)信號(hào)傳輸率的變化，提取出損傷特征;
基于對(duì)振動(dòng)篩的振動(dòng)特性及其子結(jié)構(gòu)損傷的系統(tǒng)研究，提出了振動(dòng)篩子結(jié)構(gòu)損傷定位方法。通過含損傷子結(jié)構(gòu)與完好子結(jié)構(gòu)振型的相關(guān)分析，可準(zhǔn)確確定損傷子結(jié)構(gòu)的位置。這個(gè)方法的提出是本文的創(chuàng)新性土作之一;
研究運(yùn)行狀態(tài)下以少測(cè)點(diǎn)獲取結(jié)構(gòu)損傷信息的建模方法，提出傳感器合理布置方法，實(shí)現(xiàn)大型結(jié)構(gòu)構(gòu)件局部損傷的在線檢測(cè)與定位 (確定構(gòu)件位置)。這項(xiàng)研究是本文工作的創(chuàng)新點(diǎn)性;
最后，本文用聲發(fā)射技術(shù)對(duì)損傷下橫梁進(jìn)行了損傷部位的精確定位，并對(duì)裂紋的聲發(fā)射信號(hào)進(jìn)行了小波分析及功率譜分析，可提取出裂紋的聲發(fā)射信號(hào)特征，為進(jìn)一步的后續(xù)研究探尋路徑。
在本文的研究思路上，將設(shè)備結(jié)構(gòu)健康狀況的在線動(dòng)力學(xué)監(jiān)測(cè)與離線的聲發(fā)射損傷細(xì)節(jié)分析技術(shù)相結(jié)合，提出了大型結(jié)構(gòu)損傷檢測(cè)的一個(gè)新思路。
本文對(duì)振動(dòng)篩結(jié)構(gòu)的健康狀況監(jiān)測(cè)、局部結(jié)構(gòu)損傷檢測(cè)與定位做了基礎(chǔ)性研究工作，用多種物理方法提取結(jié)構(gòu)損傷特征，為實(shí)際工業(yè)應(yīng)用做了理論準(zhǔn)備。
收藏