熔滴过渡光谱信号模式识别分类器的设计

以熔化极气体保护焊电弧光谱信号作为样本,设计了一种熔滴过渡模式识别分类器.首先对光谱信号进行预处理并抽取了多个关键性的特征参数,通过降维分析得到一组新的特征向量.随后建立了相应的识别函数和最小距离法分类器.最后利用检验样本对分类器的性能进行了检验和评价.判别结果表明,利用该分类器能够较好地对MIG焊和C02焊熔滴过渡类型进行自动识别,具有较高的准确性和识别精度,为实现熔化极气体保护焊熔滴过渡自动控制奠定了基础.     

基于RES理论的岩体失稳模式判别及其智能实现

基于完全耦合模型解决复杂岩石力学问题的方法论——岩石工程系统理论(RES)的基本原理，构造了地下工程岩体失稳模式判别的智能预测模型，并采用具有自学习、非线性映射和数据挖掘功能的模糊自组织人工神经网络模型ANN来实现RES交互作用矩阵的编码。通过一个实例说明这种方法具有良好的实用性。     

面状缺陷超声TOFD法信号和图像的特征与识别

针对铝合金厚板焊缝中易于出现的几种典型的面状缺陷,研究了其超声衍射时间差法（TOFD—time of flight diffraction）的检测特征,对获得的A扫描信号和B、D扫描图像的特征进行了分析和解释,并对B、D扫描图像进行了线性化处理。结果表明,结合A扫描信号和B、D扫描图像的特征,能够有效地对面状缺陷进行识别、定位和定量。线性化处理技术有效地提高了图像的时间分辨力,使面状缺陷的信号和图像的特征更明显,进而使定位定量更为精确。     

Analysis of ultrasonic C-scan image features and image processing for spot welding

A focused ultrasonic transducer is used for precise, step-by-step, surface ultrasonic scanning inspection of spot welds. Two methods for generating characteristic data matrices of ultrasonic echo signals of the spot welds are established. One is based on the ultrasonic echo amplitude and the other is based on frequency spectrum. Both methods generate scanning inspection data arrays and provide clear C-scan images. Based on C-scan noise reduction, the Krisch edge detection operator for edge detection-a gray window transform-reflects the shapes of the spot welds and allows quantitative determination of their size. The method based on the ultrasonic echo amplitude provides a greater quantitative detection accuracy.     

薄板点焊超声检测信号特征分析与缺陷识别

运用低碳钢薄板点焊超声检测有限元仿真模型,对气孔、压痕过深、熔核过小、脱焊等四种缺陷类型的点焊检测超声仿真信号进行快速傅里叶变换得到其频谱图,并采用统计学方法分别提取了超声信号时域和频域特征值.通过分析超声在不同缺陷焊点内部的传播规律,以及特征值的变化规律,总结了点焊缺陷类型的识别方法.利用该方法对大量点焊试样超声检测试验信号进行缺陷识别并与金相试验结果对比分析. 结果表明,综合分析超声检测信号时域和频域特征值规律,能够有效地识别点焊缺陷类型.     

Evaluation of thermal sprayed coating using ultrasonic inspection by means of bottom echo back reflection

Thermal spraying technique is widely used in various mechanical parts as a surface reforming technique. However, as demand to maintain superior mechanical performance in harsh operating environment increases, the need for non-destructive evaluation method for thermal spray coating becomes more important. For this purpose, we thinned the thickness of the thermal sprayed coating by abrasion with blasting and used ultrasonic inspection by means of bottom echo reflection for effective measurement of abrasion quantity in thermal sprayed coating. The results obtained are summarized as follows. When the thickness of thermal sprayed coating becomes thin, the echo height increases. This is because thermal sprayed coatings absorb ultrasonic energy. Ultrasonic energy absorbed by Al₂O₃ is smaller compared with Fe-13Cr coating. Thermal sprayed coatings submerged in water have a lower echo height compared with air. As mentioned above, the thermal sprayed coating thickness can be estimated using ultrasonic inspection by means of bottom echo back reflection.     

基于时频判别特征的焊缝超声检测缺陷分类

根据超声检测信号的瞬变特性,针对焊缝检测的缺陷分类问题,提出用判别追踪算法提取缺陷信号的局部时频判别特征,并结合概率神经网络实现了焊缝超声检测信号的缺陷分类.在提取时频判别特征时,提出考虑新选原子与已选原子的相关性的判别基提取方案,以降低特征之间的冗余,使提取出的特征能更有效地鉴别不同类别的缺陷.用该方法对一电子束焊缝试块中的缺陷进行了分类,结果表明,时频判别特征适合超声信号的缺陷分类,并能有效地抑制晶粒噪声的影响,考虑判别原子间相关性后可获得更高的分类正确率.     

Fuzzy pattern recognition of AE signals for grinding burn

Grinding burn is a common phenomenon of thermal damage that has been one of the main constraints in grinding difficult-to-machine materials. Grinding burn damages materials and degrades properties, by causing tensile residual stresses or microfractures in the workpiece surface. Numerous methods have been proposed to identify grinding burn. However, the main problems of current methods are their sensitivity and robustness. This paper describes a new method of grinding burn identification with highly sensitive acoustic emission (AE) techniques. The wavelet packet transform is used to extract features from AE signals and fuzzy pattern recognition is employed for optimising features and identifying the grinding status. Experimental results show that the accuracy of grinding burn recognition is satisfactory.     

Bayesian decision-based pattern recognition on spectrum signal of metal transfer modes

A software system of pattern recognition on spectrum signal of metal transfer mode has been developed using Visual Basic under Windows environment. On the basis of the coincidence relation between the spectrum signal and metal transfer mode, according to the geometrical pattern feature of the spectrum signal, several key characteristic parameters are extracted. The correspondent recognition function and a minimum distance classifier have been constructed based on Bayesian decision theory. The results show that using this system, the metal transfer mode of MIG, MAG, CO2 welding can be recognized automatically which provides the basis for automatically controlling of the metal gas arc welding metal transfer.     

Non-destructive investigation of sluices using radar and ultrasonic impulse echo

Radar and ultrasonic impulse echo have been successfully applied for the assessment of two almost 100 year old sluices. The interior condition of the side walls has been investigated with radar enabling the detection of detachments of the faced brickwork as well as the determination of moisture content. With ultrasonic impulse echo working joints were found in the concrete slab of the sluice heads. The results of these investigations gathered comparative data to reduce the number of cores which had to be taken.     

Weld seam recognition algorithm based on HSI space or pipeline ultrasonic flaw inspection

In order to realize automatic weld seam tracking for pipeline ultrasonic flaw inspection, an image processing algorithm based on HSI color space was presented. Since the color tones of weld seam were different from the parent metal, weld seam images were transformed to HSI color space. In the HSl colar space, the weld seam and base metal area can be apparently distinguished. By using this image processing algorithm, the edges and centerline of pipeline weld seam can be correctly extracted. An industrial application system was developed based on the image processing algorithm, and the image processing time is less than 70 ms and the accuracy of weld seam recognition is better than 2mm.     

焊接坡口特征设计与焊缝特征提取方法

分析了数字化CAD/CAPP/ROBOTICS集成焊接对焊接坡口设计的要求,设计了满足在设计环境和工艺规划环境双向数据流动和共享的焊接坡口特征设计的总体方案,提出了基于坡口特征库的典型焊接坡口造型和改进的基于CAD/CAM平台二次开发程序驱动的通用焊接坡口特征造型实现方法,给出了基于坡口特征设计解决坡口信息提取及焊缝特征识别与提取的方法,并实例验证了坡口特征造型方法的正确性.     

Training multilayered perceptrons for pattern recognition: a comparative study of four training algorithms

This paper presents an overview of four algorithms used for training multilayered perceptron (MLP) neural networks and the results of applying those algorithms to teach different MLPs to recognise control chart patterns and classify wood veneer defects. The algorithms studied are Backpropagation (BP), Quickprop (QP), Delta-Bar-Delta (DBD) and Extended-Delta-Bar-Delta (EDBD). The results show that, overall, BP was the best algorithm for the two applications tested.     

Evaluation of temper embrittlement by ultrasonic attenuation

Research conducted up to the present time indicates the possibility of predicting the toughness of a material by utilizing ultrasonic attenuation characteristics. However, practically all of the cases concerned involve the investigation of changes in toughness brought about by making large changes in the microstructure of the material. However, the temper embrittlement encountered in Cr---Mo steels and the like are accompanied by large degradations in toughness with virtually no changes in microstructure and tensile strength. For this reason, it has been considered difficult to evaluate temper embrittlement based on ultrasonic attenuation characteristics. In the research reported here, sensitivity was enhanced by increasing the ultrasonic frequency (to 50 MHz) and the correlation between ultrasonic attenuation characteristics and temper embrittlement characteristics has been experimentally clarified.     

电子束焊缝超声无损检测声场特性及回波特征

针对复杂结构的电子束环焊缝的检测问题,提出斜入射水浸聚焦超声C扫描成像的检测方法.研究了工件内二次焦点深度随轴向液程的变化关系,横波和纵波检测时缺陷波的识别问题,以及折射角和焦点张角对检测灵敏度的影响.结果表明,工件内二次焦点深度的试验测量值大于理论计算值,随着轴向液程的增加,测量值和计算值误差增大.利用横波检测根部未焊透缺陷,无表面回波干扰,缺陷波容易识别.在工件上声波入射面较小的情况下,焦点张角越小,检测灵敏度越高.     

Assessment of modification level of hypoeu- tectic Al-Si alloys by pattern recognition of cooling curves

1 Introduction Due to their excellent castability and good mechanical properties, Al-Si alloys are the most important and most widely used casting aluminum alloys. A further increase in mechanical properties can be achieved by means of a modification treatment with Na or Sr. Without this treatment, the eutectic silicon crystallises into an acicular or lamellar morphology with poor properties; when a modifying element is added, it becomes a very fine and fibrous silicon network resulting in a…     

超声波探伤辊道的改进

HFW超声波探伤生产线主要是对热处理后的钢管进行探伤,保证合格钢管的正常生产。探伤机组是通过两组旋转辊道来保证钢管的匀速运送。为了适应不同产品规格,旋转辊道需要转到相应的角度,要求各组辊道的角度偏差小于正负0.5°。原设计万向轴连接的辊道无法满足辊道的角度偏差要求,通过拉杆式传动实现各组辊道的并列连接,进而消除了各组辊道的累积误差,使辊道的角度偏差控制到要求的标准范围内。     

基于超声信号和图像融合的焊缝缺陷识别

超声无损检测已被广泛用来检测材料内部的缺陷,然而对缺陷性质的识别始终是检测的难点,为此研究了一种基于超声信号和图像融合的焊缝缺陷识别新方法.该方法充分利用检测数据,通过对缺陷回波信号特征与缺陷形态特征的数据融合,实现了焊缝缺陷的有效识别.利用自主研制的超声成像手动检测系统对含有气孔、夹渣、裂纹、未焊透和未熔合五类典型焊接缺陷的焊件进行了检测,分别提取缺陷的超声回波信号特征和缺陷图像的形态特征,构建神经网络实现超声信号和图像特征的数据融合.结果表明,该方法实现了多类缺陷的识别,提高了缺陷识别率,有助于焊缝质量评定.     

中厚板超声波探伤缺陷及控制技术

介绍了钢板超声波探伤的缺陷类型,对连铸板坯轧制中厚板后出现超声波探伤不合格的钢板进行取样、金相、电镜分析,找出引起中厚板探伤不合格的主要因素是钢中氢含量、铸坯中心偏析以及夹杂物;通过采用RH真空处理、钢水洁净度控制、铸机动态轻压下、加热工艺优化、控轧控冷工艺优化及板材堆垛缓冷等关键工艺技术,使钢板内在质量得到改善,探伤合格率明显提高。     

Al/Ni异种金属超声波点动焊接工艺及界面组织分析

对1060铝合金和N4镍合金异种金属进行了超声波点动焊接,优化了焊接工艺,分析了接头界面组织,焊接接头组织微观形貌和力学性能.结果表明,超声波焊接能够实现1060铝合金和N4镍合金异种金属的有效连接,能够得到韧性和强度都很高的焊接接头;确定了铝镍异种金属超声波焊接的焊接工艺参数为焊接压力25.2~36.0 MPa,焊接时间75~85 ms;当焊接压力为32.4 MPa,焊接时间为85 ms时,抗剪强度超过铝侧母材.焊接接头界面的XRD和EDS分析结果表明,接头界面存在由Al,Ni两种元素互扩散而形成的2 μm厚的反应扩散层.