Study on the thermo-effect of nugget growing in single-phase AC resistance spot welding based on the calculation of dynamic resistance

According to the welding current and electrode voltage detected in resistance spot welding, the dynamic resistance was calculated and the nugget growing thermo-effect in process of nugget forming was studied. The results showed that the process of nugget growth in resistance spot welding can be divided into three stages, which were initial stage, growing stage and stable stage. The welding current provided energy for nugget growth, which showed a high linear relationship with mean dynamic resistance and more prominent polynomial relationship with dynamic resistance heat. The dynamic resistance heat was an important evaluating indicator to the nugget growing thermo-effect, which was closely related to the characteristics of nugget forming. As the dynamic resistance heat was increased, the nugget diameter and tensile shear strength of spot weld were raised and showed a more prominent polynomial relationship. According to the calculation of dynamic resistance heat, these relationships provided a method for the nondestructive test of nugget quality features.     

Characterization of nugget nucleation quality based on the structure-borne acoustic emission signals detected during resistance spot welding process

Acoustic emission signals detected during the resistance spot welding of aluminum alloy were studied in order to assess the characterizations of welding process, the characterizations of the effect of welding parameters to nugget nucleation and the characterizations of the nugget quality by the analysis to acoustic emission signals. The results showed that the physical phases of nugget nucleation can be characterized by the acoustic emission signals detected during the resistance spot welding process. The effects of welding current and current duration to nugget nucleation can be characterized by the characteristic parameters of acoustic emission signals. The characteristic parameters of acoustic emission signals had a better relevance to nugget dimensions and weld strength, which made it possible to measure or predict the weld strength by the characteristic parameters of acoustic emission signals detected during the resistance spot welding process.     

基于实例的学习矢量量化神经网络诊断方法及其应用

基于CBR良好的可扩充性、可移植性和神经网络极强的分类能力,提出了基于实例的学习矢量量化神经网络诊断方法。该方法应用于机械故障诊断系统中,可以减小实例搜索空间,提高实例检索效率。论述了系统的设计方法和应用步骤。     

Online monitoring and control of particle size in the grinding process using least square support vector regression and resilient back propagation neural network

Particle size soft sensing in cement mills will be largely helpful in maintaining desired cement fineness or Blaine. Despite the growing use of vertical roller mills (VRM) for clinker grinding, very few research work is available on VRM modeling. This article reports the design of three types of feed forward neural network models and least square support vector regression (LS-SVR) model of a VRM for online monitoring of cement fineness based on mill data collected from a cement plant. In the data pre-processing step, a comparative study of the various outlier detection algorithms has been performed. Subsequently, for model development, the advantage of algorithm based data splitting over random selection is presented. The training data set obtained by use of Kennard–Stone maximal intra distance criterion (CADEX algorithm) was used for development of LS-SVR, back propagation neural network, radial basis function neural network and generalized regression neural network models. Simulation results show that resilient back propagation model performs better than RBF network, regression network and LS-SVR model. Model implementation has been done in SIMULINK platform showing the online detection of abnormal data and real time estimation of cement Blaine from the knowledge of the input variables. Finally, closed loop study shows how the model can be effectively utilized for maintaining cement fineness at desired value.     

代价敏感卷积神经网络：一种机械故障数据不平衡分类方法

机械设备实际工作过程中正常样本丰富、故障样本匮乏,卷积神经网络在处理这种分布不平衡的数据时对少数类的识别率很低。为解决上述问题,提出一种代价敏感卷积神经网络,首先经过多层卷积和池化运算学习原始监测数据中的机械设备本征性能状态知识;其次通过全连接层将本征性能状态知识映射为机械设备健康状态;最后利用代价敏感损失函数为少数类样本赋予较大的误分类代价,实现对不平衡的机械故障数据的有效分类。为验证所提方法的有效性,使用具有不同不平衡比的刀具数据集和轴承数据集,利用代价敏感卷积神经网络以及主流的分类算法分别测试其对于不平衡数据的分类性能。实验结果表明,所提方法对不平衡数据集中的少数类样本识别率相对于传统卷积神经网络提升了22%以上。     

A new method for measuring subgrade settlement in high-speed railway by using a linear CCD

With the rapid development of high-speed railway in China, measurements and controls of subgrade settlement have attracted much attention in recent years. A novel method to measure the subgrade settlement is proposed and a device has been developed. The device mainly contains a point light source and a point-position measurement device that uses a linear CCD to detect the position change of the point light’s image. Any position change of the point light source can be measured and the settlement for one point can be obtained remotely. The multi-point settlements can be obtained by using a series of such devices. Some methods such as self-calibration of the magnification, automatic background subtractions have been adopted to enhance the stability, adaptability and accuracy of the measurement system. The device has been used in some China high-speed railway sites and it has the advantages of high accuracy, low-cost and easy installation.     

The use of pattern recognition for classification of stochastic specïmens by a set of standard specimens

This paper describes an empirical method for the experimental assigning of specimens to a set of standard specimens. The procedure was developed especially for stochastic scenes. Typical applications of such a procedure are found in material sciences in assigning specimens to a standard series. The assignment is based on feature vectors obtained from analyses of the specimen texture by means of erosion or opening. In cases where the arrangement of particles is characteristic of the specimen, dilation or closing are used to obtain a feature vector. To increase the sensitivity of the method described the use of more than one structuring element is recommended in the analysis. Besides the classification of an unknown specimen the procedure provides information on the separability of the standard specimen and the significance of assignment.