面向三维激光熔覆修复的工艺规划研究

发布时间：2018-06-12 15:48

本文选题：激光熔覆修复 + 工艺规划　；参考：《南京航空航天大学》2017年硕士论文

【摘要】：激光熔覆作为一项新型的绿色再制造技术,以成形组织性能好、适用范围广、对环境污染小等优点在金属零件的修复中得到了广泛的应用。工艺规划是实现激光熔覆修复自动化的基础,合理的工艺规划可以大大提高修复的质量和效率。但目前激光熔覆修复缺乏完整的工艺规划体系,修复中工艺参数设置缺乏理论依据,使得修复质量无法保证,修复自动化难以实现。为了提高修复质量与效率,实现激光熔覆修复自动化,本文对激光熔覆修复的工艺规划进行了相关研究,主要内容包括:(1)针对预处理后零件表面缺损特点,提出了一种基于机器视觉的三维缺损特征提取方法。首先根据缺损区域二维图像边界与三维特征点坐标,对零件缺损底面的空间边界进行重构和定位,再结合缺损表面信息,实现对零件三维缺损特征的提取。(2)根据零件缺损特征,对激光扫描路径进行规划,分析了熔覆层形貌特征对熔覆成形的影响,建立了单道熔覆层特征参数决策模型,在此基础上研究了激光熔覆修复的路径规划与运动实现问题。(3)针对熔覆过程的复杂性,采用粒子群优化(PSO)的反向传播(BP)神经网络建立熔覆工艺参的预测模型,通过仿真测试和熔覆试验验证了所建预测模型的可靠性,实现了在样本范围内对特定单道熔覆层特征的熔覆工艺参数决策。本文针对表面损伤零件的激光熔覆修复,建立了一套修复工艺规划体系,并围绕着修复路径规划和熔覆工艺参数决策设计了多道激光熔覆试验,验证了修复工艺规划方法的合理性与可行性,表明所提方法可以有效的指导缺损零件的激光熔覆修复工作,为激光熔覆修复的自动化实现提供了理论依据。
[Abstract]:Laser cladding, as a new green remanufacturing technology, has been widely used in the repair of metal parts because of its advantages of good forming structure and properties, wide application range and low environmental pollution. Process planning is the basis to realize the automation of laser cladding repair, and reasonable process planning can greatly improve the quality and efficiency of restoration. But at present, the laser cladding repair lacks a complete process planning system, and the process parameters set in the restoration are lack of theoretical basis, which makes the repair quality can not be guaranteed, and the repair automation is difficult to realize. In order to improve the quality and efficiency of laser cladding and realize the automation of laser cladding repair, the process planning of laser cladding repair is studied in this paper. The main contents include: 1) aiming at the characteristics of surface defect of pretreated parts. A 3D defect feature extraction method based on machine vision is proposed. Firstly, according to the coordinates of 2D image boundary and 3D feature point of the defect area, the space boundary of the bottom surface of the defect is reconstructed and located, and then the 3D defect feature of the part is extracted according to the defect feature by combining the defect surface information. The laser scanning path is planned, and the influence of the morphology of the cladding layer on the cladding forming is analyzed, and the decision model of the characteristic parameters of the cladding layer is established. Based on this, the path planning and motion realization of laser cladding repair is studied. Aiming at the complexity of cladding process, a predictive model of cladding process parameters is established by using PSO (Particle Swarm Optimization) and back Propagation (BPN) neural network. The reliability of the prediction model is verified by simulation test and cladding test, and the decision of cladding process parameters is realized in the sample range for the characteristics of a specific single cladding layer. In this paper, a repair process planning system is set up for repairing surface damage parts by laser cladding, and multi-channel laser cladding test is designed around repair path planning and cladding process parameters decision. The rationality and feasibility of the repair process planning method are verified. The results show that the proposed method can effectively guide the laser cladding repair of defective parts and provide a theoretical basis for the automatic realization of laser cladding repair.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TG665

【参考文献】