喜歡就充值下載吧。。。資源目錄里展示的全都有，，下載后全都有,,請放心下載，原稿可自行編輯修改=====================喜歡就充值下載吧。。。資源目錄里展示的全都有，，下載后全都有,,請放心下載，原稿可自行編輯修改=====================喜歡就充值下載吧。。。資源目錄里展示的全都有，，下載后全都有,,請放心下載，原稿可自行編輯修改=====================

遼寧科技大學(xué)本科生畢業(yè)設(shè)計 第5頁 焊縫無損檢測現(xiàn)狀 [澳大利亞] RJ Ditchburn SK Burke CM Scala 超聲檢測 20世紀(jì)60年代，超聲檢測就被作為焊縫檢驗(yàn)的一種NDT技術(shù)。從那時起，這種技術(shù)就得到廣泛的發(fā)展，日益被人們所接受。因此，超聲檢測現(xiàn)已成為一項(xiàng)重要技術(shù)，用來判定很多在役檢驗(yàn)的焊接結(jié)構(gòu)，如海上結(jié)構(gòu)、核工業(yè)及壓力容器工業(yè)，以及海洋應(yīng)用的某一范圍。 在在役檢驗(yàn)方面，超聲檢測之所以比X射線照相更受歡迎，是由于射線照相的內(nèi)在局限性和應(yīng)用超聲所獲得的實(shí)際利益。如上所述，射線照相在辨別立體缺陷時非常有效，但在檢查或測量平面缺陷，有可能是最嚴(yán)重的缺陷類型裂紋時，其能力則有限。超聲波經(jīng)平面和立體缺陷散射，可用于探測這兩種類型缺陷，并測量其尺寸，如果采用適當(dāng)?shù)姆椒?，超聲甚至能探測出閉合的裂紋。超聲還能容易地給出與缺陷有關(guān)的深度信息，而X射線，還需專門且昂貴的技術(shù)（如CT）才能獲得這些信息。在節(jié)約資金方面，通過提高生產(chǎn)率，超聲檢測比射線照相可獲得更大的利益。90年代，人們對射線安全性的日益關(guān)注嚴(yán)重阻礙了X 射線照相的繼續(xù)使用。 在過去的幾十年里，超聲檢驗(yàn)已從一種單純的人工操作技術(shù)，經(jīng)過計算機(jī)輔助處理的人工操作技術(shù)，到自動掃描儀的使用，最近發(fā)展為焊縫評估用聯(lián)接多個壓電傳感器的全自動系統(tǒng)。對日益完善的系統(tǒng)在缺陷探測方面應(yīng)用的可靠性研究已形成建立超聲焊縫檢驗(yàn)可信度的一個重要因素。如對鋼構(gòu)件檢驗(yàn)方法（PISC) I ,Ⅱ和Ⅲ進(jìn)行研究的目的是要獲得超聲檢驗(yàn)在核工業(yè)和壓力容器工業(yè)中的最佳應(yīng)用。在PISC研究之外，已進(jìn)行了某些工作來確定檢驗(yàn)特殊焊縫幾何形狀的可靠方法，包括單面V形坡口焊縫、雙面V形坡口焊縫以及對接焊縫。這些研究中的幾項(xiàng)已用射線照相和超聲檢驗(yàn)的可靠性作了比較，結(jié)果表明，用超聲探測缺陷的可靠性隨著該系統(tǒng)完善的程度而增大。根據(jù)Lebowitz 和DeNale，該結(jié)果還表明人工操作超聲技術(shù)，甚至使用最不完善的超聲方法，對存在的不連續(xù)處的拒收率等于或大于射線照相的拒收率。 超聲檢測焊接結(jié)構(gòu)要求不僅能可靠地探測缺陷，而且能精確地對缺陷定位和測量其大小，使接收／拒收準(zhǔn)則能夠正確實(shí)施，人工操作的超聲系統(tǒng)通常使用的技術(shù)是20dB衰減（圖la）、6dB衰減或與來自鉆孔的幅度比較。然而，這些技術(shù)都有誤差，這種誤差不僅是由缺陷形狀、方向和位置的影響引起的，而且還由衰減、耦合、分辨力及設(shè)備特征引起。結(jié)合計算機(jī)輔助處理的超聲系統(tǒng)使得用于缺陷探測及測量的好方法如聲時衍射法( TOFD)（圖lb）便于實(shí)施，在 PISCⅡ中，標(biāo)準(zhǔn)方法加上TOFD，對所要求的缺陷拒收率，可得到近乎完美的結(jié)果。在測量缺陷大小方面，根據(jù)合成的小孔聚焦（SAFT）及其派生方法如 SUPER – SAFT。進(jìn)行超聲成像的自動超聲系統(tǒng)已取得了重要進(jìn)展。 （a）20dB衰減技術(shù) （b）TOFD技術(shù) 圖1 超聲測量缺陷尺寸 對于焊縫超聲檢驗(yàn)可靠方法的開發(fā)要求了解聲波與各種類型焊縫缺陷的相互作用、超聲波在復(fù)雜幾何體中的傳播、檢驗(yàn)?zāi)骋唤Y(jié)構(gòu)中表面閉合缺陷所引起的特殊問題、復(fù)合層和其它顯微組織對超聲波傳播的影響。雖然超聲波在鐵素體和輕質(zhì)合金焊縫中的傳播相對來說不是很復(fù)雜，但對奧氏體焊縫的顯微組織已引起特別注意。這些材料對超聲波的衰減很嚴(yán)重，由于存在大晶粒散射而使背底噪聲高，除非沿著主要晶軸傳播，否則還導(dǎo)致超聲波束的偏移。因此，最近的研究已直接面向開發(fā)專門的超聲技術(shù)去處理這些復(fù)雜情況，特別PISCⅡ,和Ⅲ對超聲波在奧氏體材料中傳播的詳細(xì)模擬試驗(yàn)，已取得了重大進(jìn)展。將來，這些模型應(yīng)能更精確地估算特殊缺陷的位置和大小誤差，為奧氏體鋼和奧氏體焊縫鋼檢驗(yàn)采用改進(jìn)的規(guī)則奠定基礎(chǔ)。 目前，人們對降低焊縫檢驗(yàn)成本的需求日益增長。在現(xiàn)代超聲系統(tǒng)中，自動掃描儀的出現(xiàn)、多個探頭的使用及計算機(jī)輔助處理通過提高檢驗(yàn)速度和檢驗(yàn)的可靠性降低了成本，但使用自動化設(shè)備時，設(shè)備和校準(zhǔn)費(fèi)用增高了。同時、由于近來超聲系統(tǒng)的發(fā)展，分析超聲數(shù)據(jù)的費(fèi)用有可能提高，因?yàn)楦鞣N類型的缺陷（甚至很小的缺陷）都能探測到，而不管這些缺陷是否嚴(yán)重。對這個問題的解決方法就是改進(jìn)缺陷嚴(yán)重性接收／拒收準(zhǔn)則的自動化應(yīng)用。因此，現(xiàn)在要努力開發(fā)神經(jīng)網(wǎng)絡(luò)，使其用于超聲系統(tǒng)中對缺陷類型、尺寸和位置進(jìn)行分級，得出的結(jié)果符合特定的檢驗(yàn)規(guī)則。在模擬的焊縫缺陷（已獲得100％正確的缺陷類型分辨力）和真實(shí)的焊縫缺陷（使用不同方法獲得約為90％的成功率）研究方面，一些試驗(yàn)室已獲得非常有價值的結(jié)果。借助于神經(jīng)網(wǎng)絡(luò)，接收／拒收準(zhǔn)則在自動化應(yīng)用方面的預(yù)備工作也已進(jìn)行。神經(jīng)網(wǎng)絡(luò)只能在有代表性的數(shù)據(jù)上試用后，才能證明它們是成功的。然而，獲得代表性數(shù)據(jù)的費(fèi)用很高。一種替代方法，是在網(wǎng)絡(luò)上采用從超聲波與焊縫缺陷相互作用的加強(qiáng)數(shù)學(xué)模型產(chǎn)生的數(shù)據(jù)。這些模型是PISC Ⅲ下一個PISC繼續(xù)項(xiàng)目的開發(fā)。在這些數(shù)學(xué)模型變得更完善之前，對于焊縫的超聲判定，強(qiáng)調(diào)使用合格的科技人員是有必要的。 今后，進(jìn)一步優(yōu)化焊縫超聲檢驗(yàn)仍有很多困難，先進(jìn)方法如TOFD的應(yīng)用有可能取得實(shí)質(zhì)性的改進(jìn)，同時，神經(jīng)網(wǎng)絡(luò)的實(shí)施還只是個開始。在焊接應(yīng)用中，改進(jìn)的超聲產(chǎn)生和探測有多種選擇方法，如使用相控陣、激光技術(shù)（如下所述）和其它特種探頭。在超聲可靠性范圍內(nèi)需考慮其它一些因素，如殘余應(yīng)力、較高頻率的影響及對真實(shí)缺陷而不是模擬缺陷更全面的考慮。 對檢驗(yàn)的總經(jīng)濟(jì)效益也應(yīng)給予更多的考慮。要獲得最大經(jīng)濟(jì)效益就需對所給的某項(xiàng)檢驗(yàn)選擇最適合的技術(shù)，包括使用一種以上的技術(shù)來判定焊接結(jié)構(gòu)的不同部位。例如，磁粉檢驗(yàn)與超聲結(jié)合快速而經(jīng)濟(jì)有效地探測焊縫表面裂紋。新的電磁方法在這種情況下也起著作用（如下面所討論的）。還存在的問題是如何將超聲檢驗(yàn)技術(shù)的進(jìn)展補(bǔ)充到焊縫判定規(guī)則中。 在線焊縫檢驗(yàn) 在自動焊接過程中，焊接工藝的在線監(jiān)測和控制能提高焊縫質(zhì)量并增加產(chǎn)量，通過實(shí)時無損評價技術(shù)與焊接工藝的結(jié)合，可達(dá)到焊縫的監(jiān)測和控制。另外，如果產(chǎn)生了焊接缺陷，則在這些缺陷被后續(xù)焊道覆蓋前，對這些缺陷進(jìn)行修補(bǔ)以減少焊后的檢驗(yàn)和修補(bǔ)。 高質(zhì)量焊縫取決于與焊接準(zhǔn)備工作有關(guān)的正確的焊池尺寸、幾何形狀和位置。在線自動焊縫監(jiān)測系統(tǒng)通常具有可提供焊池狀況信息的傳感器。利用這些信息，確定焊池狀況與至少一個臨界焊接參數(shù)（如電流、電壓、焊距位置和傳送速度）間的關(guān)系，通過傳感器的反饋線路調(diào)整焊接工藝（圖3）。該系統(tǒng)可連續(xù)調(diào)節(jié)工藝參數(shù)來維持期望的穩(wěn)定工藝狀態(tài)，且不需要操作人員介入。 圖2 描述焊接工藝引入檢測原理示意圖 焊接的動態(tài)特征意味著數(shù)據(jù)采集和處理必須能迅速獲得焊接過程中的有用信息，兩級實(shí)時射線照相分析包括快速尋找缺陷區(qū)域，并對缺陷進(jìn)行仔細(xì)辨別和定位的系統(tǒng)已達(dá)到這個要求。實(shí)時射線照相圖像已用于對接接頭焊縫電弧焊條件的控制；焊池凹陷的實(shí)時射線照相圖像和池后焊縫迅速固化圖像的綜合方法可用于焊縫熔透和質(zhì)量控制。 在線超聲傳感已用于確定金屬極氣體電弧焊(GMA)和鎢極氣體保護(hù)電弧焊（GTA）焊接工藝的質(zhì)量，這種技術(shù)可實(shí)時探測焊池幾何形狀和焊縫缺陷。該技術(shù)評價了焊條處熔化的焊池質(zhì)量和焊條后固化的焊縫金屬質(zhì)量，可探測側(cè)壁不完全熔透和密集氣孔。使用具有92％成功率的專家系統(tǒng)技術(shù)可將這些不連續(xù)與完好焊縫區(qū)分開來，遺憾的是這個專家系統(tǒng)的算法不能成功地識別不連續(xù)的類型。正確辨別密集氣孔的比率為70%，正確辨別側(cè)壁不完全熔化則為63％。Bull 等人已表明GTA和電阻點(diǎn)焊可用超聲法監(jiān)測，并在電阻點(diǎn)焊在線監(jiān)測系統(tǒng)開發(fā)方面有所進(jìn)展。 使用壓電傳感器和耦合劑因存在耦合介質(zhì)使焊縫不純凈，對生產(chǎn)是不利的，為了解決這個問題，已研制出一種非接觸式超聲系統(tǒng)。該系統(tǒng)用一種受脈沖作用的釹YAG激光來產(chǎn)生超聲波，用電磁聲傳感器（EMAT）接收。 另外，兩種非接觸檢測技術(shù)正在研究過程中，第一種技術(shù)使用雙波長光導(dǎo)纖維傳感器可同時觀察來自焊接過程的紅外線（IR）和紫外線（UV），射線由熱熔池和由光束/氣體相互作用發(fā)出的等離子體產(chǎn)生。這種技術(shù)已成功地用于顯示激光焊中所遇到的干擾。第二種技術(shù)使用錄像圖像在線處理，它可保證GMA焊接中焊縫接頭區(qū)和焊道中心線的冷卻速率，模糊邏輯控制裝置和人工神經(jīng)網(wǎng)絡(luò)可用這個信息來修改工藝參數(shù)。 結(jié)論及今后的工作 近幾年在焊縫NDT技術(shù)方面已取得了重大進(jìn)展，特別是用超聲和電磁方法測量裂紋尺寸和殘余應(yīng)力，以及焊接工藝的在線監(jiān)測方面。成本效益與諸多因素有關(guān)，如NDT技術(shù)的可靠性、靈敏度、速度和覆蓋面，這些方面的要求導(dǎo)致焊縫自動超聲檢驗(yàn)系統(tǒng)的應(yīng)用得以擴(kuò)大，特別是在核工業(yè)和壓力容器領(lǐng)域。非接觸式NDT技術(shù)已取得迅速發(fā)展，今后在自動數(shù)據(jù)處理的神經(jīng)網(wǎng)絡(luò)不斷發(fā)展的同時，還應(yīng)提高其檢驗(yàn)速度。 總之，焊縫NDT特別是在使檢驗(yàn)成本降至最低而不損壞結(jié)構(gòu)完整性方面，仍有很多難題，通過焊接工程師與NDT專家之間的密切合作，這些難題正在得到解決，既具備NDT知識又具備焊縫生產(chǎn)，斷裂力學(xué)和結(jié)構(gòu)力學(xué)知識，是獲得最佳效果的基礎(chǔ)。 第8頁 遼寧科技大學(xué)本科生畢業(yè)設(shè)計 NDT of Welds: state of the art R.J. Ditchburn , S.K. Burke and C.M. Scala Ultrasonics Ultrasonics was introduced as an NOT technique for weld inspection in the 1960’s .Since then, the technique has undergone extensive development and gained increasing acceptance. Consequently, ultrasonics is now the major technique used for validation of welded structures in many in-service inspection applications, eg in off-shore structures. In nuclear and pressure vessel industries and in a range of naval applications. The emergence of ultrasonics as a preferred technique over X-radiography in these in-service inspections is due both to inherent limitations in radiography and to actual benefits in applying ultrasonics. As described above, radiography is excellent for identifying volumetric defects but is limited in its ability to detect or size planar defect, such as cracks, which are likely to be the more serious defects type. Ultrasonic waves are scattered by planar and volumetric defects, making the ultrasonic technique useful for detecting and sizing both types of defects. Even closed cracks are detectable by ultrasonic. Provided that appropriate procedures are used. Ultrasonics also readily gives depth information concerning a defect, whereas for X-rays, specialized and expensive techniques such as computer tomography are needed to obtain such information. Ultrasonics also offers benefits over radiography in terms of cost savings through increased productivity. Finally, in the 1990’s the increasing concerns about radiation safely are a severe disincentive to the continued use of X-radiography. In the last few decades, ultrasonics has developed from a purely manual technique to a manual technique with computer-assisted processing to the use of automatic scanners and more recently to the development of fully automated systems incorporating multiple piezoelectric studies on the use of this range of increasingly sophisticated systems for defect detection have formed a major factor in establishing the credibility of ultrasonics for weld inspection. Studies such as the Programme for Inspection of Steel Components (PISC) ultrasonic inspection in the nuclear and pressure vessel industries. Outside the PISC studies, useful work has been carried out to determine reliable procedures for inspecting specific weld geometries including single-V and double-V welds and also compared the reliability of radiography vs. ultrasonic inspection Overall, the results of the reliability studies indicate that the probability of detecting a defect with ultrasonics increases with the degree of sophistication of the system. According to Lebowitz and DeNale the results also indicate that manual ultrasonic procedures, can be expected to reject an equal or greater percentage of the discontinuities present than will radiography. Ultrasonic validation of welded structures requires not only reliable defect detection but also sufficiently accurate defect location and sizing to allow acceptance/rejection criteria to be correctly implemented Manual ultrasonic systems usually rely on the use of amplitude dependent techniques for defect sizing. Techniques commonly used are the 20dB drop (shown in Figure la), the 6dB drop, or comparison with the amplitude form a drilled hole. However these techniques are known to be inaccurate. The inaccuracies are caused not only by the effects of defect shape, orientation and location, but also by attenuation, coupling, resolution and equipment characteristics. The incorporation of computer-assisted processing into ultrasonic systems has allowed the easy implementation of potentially better methods for defect detection and sizing such as time-of-flight-diffraction (TIFD) (see Figure 1b),eg in PISCⅡ the addition of TOFD to standard procedures gave nearly perfect results in terms of required rejection rate for-defects. Important advances in defect sizing have also been made possible by the incorporation in automated ultrasonic systems of ultrasonics imaging based on Synthetic aperture focusing (SAFT) and variants such as SUPERSAFT. The development of reliable procedures for the application of ultrasonics to weld inspection had required an understanding of the interaction waves with the various types of weld defects, of wave propagation in complicated geometries, of particular problems caused by inspecting for defects close to the surface of a structure of the effects of cladding and other micro structural influences influences on wave propagation. While wave propagation in ferritic and light-alloy welds Is relatively uncomplicated, the microstructures of austenitic welds have caused special concerns. These materials strongly attenuate ultrasonic waves, cause high background noise due to scattering from the large grains present, and result in skewing of the ultrasonic beam unless the propagation is along principal crystallographic (a) 20 dB Drop Technique (b) TOFD Technique Figure 1 Ultrasonic defect sizing (a) The 20dB intensity drop technique. Where the transducer is positioned at points is used along with probe calibration characteristics to estimate the defect length.(b) Time-of-flight-diffraction (TOFD) technique. A two probe technique used to determine crack size and location utilizing the diffracted waves from the tips of the defect. axes. Thus, much recent research has been directed toward the development of specialized ultrasonic techniques to deal with these complications. Considerable progress has already been made, especially under PISCⅡ and Ⅲ where detailed has been undertaken. In the future, these models should allow more accurate estimation of location and sizing errors for specific defects, and provide the basis for improved codes for inspection of austenitic steels and weld steels. In today’s world, there is an increasing need to minimize the cost of weld inspection. The advent of automated scanners, the use of multiple probes and computer-assisted processing in modern ultrasonic systems have reduced costs by increasing both the speed and reliability of inspection. On the negative side, equipment, and calibration costs are higher with automated equipment. Also, the costs in actually interpreting ultrasonic data could rise due to the recent advances in ultrasonic systems, since all types of defects and even very small defects can be detected, whether or not the defects are critical. The solution to this problem would be improvements in the automated application of acceptance/rejection criteria on defect criticality. Hence, considerable effort is now being directed towards the development of neural networks to be used in ultrasonic systems to classify defect type, size and location, and resulting conformance with a particular inspection code. Very promising results have already been obtained in several laboratories in studies both on simulated weld defects, where a 100% correct classification rate was achieved in defect type, and on real weld defects where success rates of the order of 90% were achieved using a variety od methods. Preliminary work has also been made in the automated application of acceptance/rejection codes via neural networks. Clearly neural networks will only prove successful if they can be trialed on representative data. However, representative data can prove expensive to acquire. For example, the PISC programme is currently costed at ＄200M, and it seems unlikely that this type of effort will be duplicated in other industries in the near future. An alternative approach would be to trial the networks using data generated from robust mathematical models of the interaction of ultrasonic waves with weld defects. The development of such models is a continuing PISC objective under PISCⅢ。 Until these mathematical models are more complete, an emphasis staff seems necessary and technically qualified staff seems necessary for ultrasonic weld validation. For the future, many challenges remain in optimizing ultrasonic inspection of welds. Substantial improvements are possible in the application of neural networks has only just started. Various options exist for the improved generation and detection of ultrasonic in welding applications, eg by the use of phased arrays, laser techniques (as described below) and other specialist probes. A number of additional factors need to be considered in the ultrasonic reliability area, eg residual stress, the effect of higher frequencies, more extensive consideration of real rather than simulate defects. Greater consideration also needs to be given to the overall cost effectiveness of inspection. One of the elements in maximizing cost-effectiveness is the selection of the most appropriate techniques for a given inspection, including the possible use of more than one technique to validate different parts of a welded structure. For example, magnetic particle testing is already used in conjunction with ultrasonics for rapid and cost-effective detection of surface cracks in welds. New electromagnetic methods could also have a role to play here (as discussed in the following section). Finally, there are clearly challenges in implementing advances in ultrasonic inspection technology into the codes for weld validation. In-production weld inspection On-line monitoring and control of the welding process has the potential to improve weld quality and increase productivity in automated welding. Weld monitoring and control can be achieve by the integration of real-time nondestructive evaluation techniques with the welding process. In-production weld inspection can improve weld quality and may provide a significant cost reduction. The welding parameters can usually be adjusted to prevent defects do occur the flaws can be found and repaired before they are covered by subsequent welding passes, leading to a decrease in the level of post-weld inspection and repair. Good quality welds rely in the correct weld pool size geometry and position relative to the weld preparation. In-production automated weld monitoring systems usually have sensors providing information on the stare of the weld pool. Using this information and determining a relationship between the stare of the weld pool and at least one of the critical welding parameters (eg current, voltage, torch position and travel speed) the welding process can be adjusted by a feed-back loop from the process parameters to maintain the desired stable process state. This can be achieved with little operator intervention. Figure 3 Schematic diagram depicting the principle of introduction monitoring of the welding process The dynamic nature of welding means that data acquisition and processing, must be rapid enough to extract useful information before any major change occurs in the welding process. A two-step real-time radiographic analysis involving a fast search for defective regions followed by fine identification and location of defects has achieved this requirement. Real-time radiographic images have been used in the control of arc welding conditions in butt-joint welds. A combined approach using real-time radiographic images of the weld immediately behind the pool was used for weld penetration and quality control. In-production 371678 sensing has been used to determine the quality of both gas metal arc (GMA) and gas tungsten arc (GTA) welding processes. This technique allows detection of weld pool geometry and weld defects in real time. The technique evaluated the solidified weld metal behind the electrode. Two types of discontinuities were detected: incomplete sidewall penetration and porosity. These discontinuities were distinguished from sound welds using an expert-system technique with a success rate of 92%. Unfortunately the expert-system algorithm was unable to discriminate between the discontinuity types as successfully. Porosity was identified correctly 70% of the time and incomplete sidewall fusion was correctly identified 63% of the time. But et al, have shown that CTA and resistance spot-welding can be monitored ultrasonically and are making progress in developing an in-production monitoring system for resistance spot-welding. Using a piezoelectric transducer and complaint presents the possibility of contamination of the weld by the coupling medium, obviously impractical for production. To overcome this problem, a non-contact ultrasonic system has been developed. The system developed uses a pulsed Nd: YAG laser for ultrasound generation and an electromagnetic acoustic transducer (EMAT) for ultrasound reception. Two other non-contact transducer are currently under investigation. The fist technique involves the simultaneous observation of the infrared (IR) and the ultraviolet (UV) radiation from the welding process using dual wavelength fiber-optic sensors. The radiation is produced both from the hot melt pool and from the plasma produced by the beam/vapour interaction. The technique has been successfully employed to indicate disturbances encountered in laser welding. The second technique uses in-production processing of video images. Encouraging results have been reported that provide weld joint area and bead centerline cooling rates in GMA welding. This information is then used by a fuzzy logic controller and an artificial neural network to modify process parameters. The increasing demands of high production rates and greater weld quality at lower costs will necessitate the strengthening of the bond between the technologies of welding and nondestructive inspection. In achieving these goals, in-production monitoring of the welding process will increase in importance and may well become indispensable. Conclusions and future work In recent years some exciting developments have occurred in NDT techniques for weld inspection. Major advances have been made in several fields particularly in ultrasonics, electromagnetic methods for both crack sizing and residual stress measurement, and in on–line monitoring of the welding process. Many of the advances in NDT techniques have been driven by today’s increasing pressures for cost-effective weld inspection. Cost effectiveness is linked to factory such as reliability, sensitivity speed and coverage of NDT techniques. The need for greater reliability, speed and coverage has resulted in the increasing use of automated ultrasonic systems for weld inspection particularly in the nuclear application. Rapid development has occurred in a range of non-contact NDT techniques which should improve speed of inspection in the future, as will the continuing development of neural networks for automated data processing. Worldwide, the need exists to implement these advance in NDT technology in weld validation codes. Clearly there is no value in replacing an acceptable inspection method for the sake of technological sophistication alone. However, substantial improvements are possible by the incorporation of advanced concepts such as Time-of-Flight-Diffraction in weld inspection. In conclusion, many challenges remain in NDT of welds, particularly in minimizing inspection costs without prejudicing structural integrity. These challenges are best met by close cooperation between welding engineers and NDT experts, so that best practice is achieved based on a knowledge of not only NDT but also weld manufacture, fracture mechanics and structural mechanics.