盘型激光焊接状态多传感信息融合分析

发布时间：2018-08-16 17:19

【摘要】：针对大功率盘型激光焊接状态,研究一种基于支持向量机的多传感信息融合分析方法.使用紫外、可视和红外波段的两个高速摄像机同时获取激光焊接过程中金属蒸气、飞溅和熔池动态图像.通过模式识别技术提取焊接过程多传感信息特征及进行数据主成分特征分析,并以焊缝宽度变化作为衡量焊接状态稳定性的参数.运用支持向量机融合各特征,通过网格搜索和粒子群算法优化支持向量机参数,建立基于支持向量机的多传感信息融合模型.结果表明,支持向量机多传感信息融合方法能够有效预测焊缝宽度变化趋势,为大功率盘型激光焊接状态的实时监控提供试验依据.
[Abstract]:A multi-sensor information fusion analysis method based on support vector machine (SVM) is proposed for high power disk laser welding. Dynamic images of metal vapor, spatter and molten pool during laser welding were obtained by using two high speed cameras in ultraviolet, visible and infrared bands. The multi-sensor information feature of welding process is extracted by pattern recognition technology and the principal component feature of the data is analyzed. The variation of weld width is taken as the parameter to measure the stability of welding state. The support vector machine (SVM) is used to fuse the features, and the support vector machine (SVM) parameters are optimized by mesh search and particle swarm optimization (PSO), and a multi-sensor information fusion model based on support vector machine (SVM) is established. The results show that the multi-sensor information fusion method based on support vector machine can effectively predict the change trend of weld width and provide experimental basis for real-time monitoring of high-power disc laser welding status.
【作者单位】：广东工业大学机电工程学院;广州番禺高勋染整设备制造有限公司;
【基金】：国家自然科学基金资助项目(51175095) 广东省协同创新与平台环境建设专项资助项目(2015B090901013) 广东省重大科技专项资助项目(2014B090921008) 广州市科学研究专项资助项目(1563000554) 佛山市科技创新专项资助项目(2014AG10015)
【分类号】：TG456.7

【相似文献】