摘要：
In order to solve the problem of low accuracy of tool wear detection due to the poor quality of generated data under small sample problems, a deep learning model based on data enhancement and feature fusion is proposed. Firstly, in order to solve the problem that there is no quality evaluation standard in the training process of the traditional generative adversarial network (GAN), the K nearest neighbor algorithm is proposed to test the data generated by the GAN model for the second time. The improved GAN model can be automatically trained to get the optimal model according to the second test results. Secondly, in order to enhance the anti-interference effect of the model, a double-path parallel convolutional neural network (DPCNN) which combines with the characteristics of frequency domain and time-frequency domain is constructed to analyze the wear data. Furthermore, the hyperparameters of the model are optimized by Bayesian optimization algorithm (BOA). Finally, the effectiveness of this method is verified in the saw blade wear detection experiment. The results show that the performance of this model is better than other models, and the accuracy rate in the experimental detection reaches 100%.
作者机构：
[Yang, Kai; Wang, Xiangjiang] College of Mechanical Engineering, University of South China, Hu Nan, Hengyang;421001, China;[Yang, Kai; Wang, Xiangjiang] 421001, China
会议名称：
2022 International Conference on Mechatronics and Automation Technology, ICMAT 2022
摘要：
在进行核电站检修工作时，提前确定机器人在水室外的工作位置可以为机械臂进入人孔提供相对更大的活动范围和操作空间。为提高后续堵板作业的工作效率，通过建立蒸汽发生器人孔与机器人的相对空间位置数学模型，制定了一种基于TOF（Time of flight，飞行时间）激光位移传感器的定位方法，经实际操作可实现一次侧堵板的拆装工作。并通过机械臂末端位置对比实验得到该方法的相对重复定位误差为0.0845%，具有一定的可靠性。
作者机构：
[Weiwei X.; Nini S.; Xiangjiang W.] School of Mechanical Engineering, University of South China, Hunan, Hengyang, 421001, China;Hunan Provincial Key Laboratory of Ultra-Fast Micro-Nano Technology and Advanced Laser Manufacturing, Hunan, Hengyang, 421001, China;[Zhiyong C.] School of Electrical Engineering, University of South China, Hunan, Hengyang, 421001, China;[Xinling S.; Xinlin W.] School of Mechanical Engineering, University of South China, Hunan, Hengyang, 421001, China<&wdkj&>Hunan Provincial Key Laboratory of Ultra-Fast Micro-Nano Technology and Advanced Laser Manufacturing, Hunan, Hengyang, 421001, China
摘要：
为了研究激光切割UO2陶瓷芯块-316Ti不锈钢复合结构温度场的分布，考虑了材料的热物性参数与温度的关系，运用ANSYS有限元软件模拟了该复合结构的激光切割过程，通过ANSYS的Fluent模块实现激光热源C语言程序的UDF加载与编译。研究了不同激光切割工艺参数下的温度分布云图，结果表明：温度分布主要呈现为类彗尾状；最高温度集中在切割热源中心处，工件表面的最高温度随着激光功率的增加和切割速度的降低而升高；当激光功率过小或切割速度过大时，工件会因热量输入不足而切不透；当激光功率过大或切割速度过小时，切缝会因材料熔化区域变大而变宽。仿真结果可以对实际切割过程的工艺参数优化提供参考。 To investigate the temperature-field distribution of UO2 ceramic pellets-316Ti stainless steel composite structure during laser cutting processes, considering the relationship between thermal physical parameters and the temperature of materials, we simulated the laser cutting process for the composite structure using ANSYS finite-element software. User defined function loading and compiling of laser heat source C language program was realized using the Fluent module of ANSYS. The temperature distribution cloud maps were studied at different cutting parameters. The results show that temperature distribution is mainly expressed as comet-like state. Maximum temperature is concentrated at the center of the cutting heat source, and it increases with increasing laser power and decreasing cutting speed. When the laser power is too low or cutting speed is too high, the workpiece due to insufficient heat input will not be cut through. If the laser power is too high or cutting speed is too low, the kerf width widens due to the lager material melting area. The simulation results can be used to optimize the parameters of the actual cutting process.
