| 摘要: |
| 电弧增材制造是一种重要的增材制造技术，其质量控制和缺陷诊断是保障产品精度和性能的关键。该技术近年来在制造业中得到广泛应用，特别是在高性能材料的制造和修复方面。通过电弧作为热源，该技术能够实现快速、高效的金属部件增材成形。目前，电弧增材制造面临的挑战包括成形质量控制、过程监测与缺陷诊断等问题，本文针对电弧增材制造的正常焊道、不连续焊道、塌陷焊道，通过对采集的电信号进行分析，以实现缺陷分类。首先，对原始信号通过变分模态分解（VMD）方法分解，对信号的时域特征、频域特征以及信息熵特征进行特征提取，挖掘出与缺陷类型相关的关键特征。随后，采用特征融合方法，提高了模型的准确性和泛化能力。最后，利用基于支持向量机（SVM）分类算法与随机森林（RF）分类算法对比分析，对成形良好、不连续、塌陷不同缺陷进行了有效分类。 |
| 关键词: 电弧增材制造 电信号 缺陷分类 |
| 基金项目:中央高校基本科研业务费专项资金（3092****008）。 |
|
| Fault Diagnosis Method of Arc Additive Forming Based on Electric Signal |
|
Zhou Yuejun Guo Yiming Xiao Mingkun Pei Shuaiwen
|
|
School of Mechanical Engineering, Nanjing University of Science and Technology, Nanjing 210094
|
| Abstract: |
| Arc additive manufacturing is an important additive manufacturing technology, and its quality control and defect diagnosis are the key to ensure product accuracy and performance. The technology has been widely used in the manufacturing industry in recent years, especially in the manufacture and repair of high-performance materials. By using the arc as a heat source, the technology enables rapid and efficient additive forming of metal parts. At present, the challenges facing arc additive manufacturing include forming quality control, process monitoring and defect diagnosis, etc. This paper analyzes the collected electrical signals for normal, discontinuous and collapsed welding passes of arc additive manufacturing to realize defect classification. First, the original signal is decomposed by variational mode decomposition(VMD) method, and the time domain features, frequency domain features and information entropy features of the signal are extracted, and the key features related to the defect types are excavated. Then, the feature fusion method is used to improve the accuracy and generalization ability of the model. Finally, the classification algorithm based on support vector machine(SVM) and random forest(RF) are compared and analyzed, and the different defects with good shape, discontinuity and collapse are effectively classified. |
| Key words: arc additive manufacturing electrical signal defect classification |
