民用大飞机15-5PH不锈钢铣削工艺研究

发布时间：2018-10-13 11:36

【摘要】：15-5PH不锈钢作为一种新型不锈钢材料,以其优良的机械性能、耐高温和耐腐蚀性能,被广泛应用在飞机驱动装置、机身主梁及起落架等关键结构件。其优良力学性能也预示着该材料的难加工性。在15-5PH不锈钢加工过程中,由于其导热系数低和弹性模量小等,易出现变形抗力大、切削温度高、刀具磨损破损快以及加工硬化等现象,从而导致15-5PH不锈钢的加工质量和加工效率低下、加工成本过高。因此,研究15-5PH不锈钢的切削性能,优化其工艺参数,开发具有自主知识产权的专用数据库对于提高加工效率、控制加工成本和加工质量具有重要意义。本文针对企业对于切削数据库系统的功能需求,选取粗加工、圆弧面和腹板三类特征开展基础实验研究、切削结果建模研究和切削参数优化研究。再将所有实验数据和实验结果录入到切削参数库系统,并调试运行。针对15-5PH不锈钢粗加工过程中,刀具磨损和破损严重以及刀具寿命低下等问题,选用单因素实验设计方法,开展刀具磨损和刀具寿命实验。观察和分析不同条件下刀具的磨损和失效形式,分析各切削用量对刀具寿命的影响机理和规律。采用BP神经网络算法,进行刀具寿命建模,建立刀具寿命的预测模型,实现刀具寿命的经验预测。并借助粒子群优化算法,以切削效率为目标,进行粗加工铣削参数优化。针对曲面粗糙度,选用正交试验设计方法,开展15-5PH不锈钢圆弧面工件精加工实验。研究分析铣削参数对圆弧面表面粗糙度的影响程度和规律。同时,采用偏最小二乘法和运动学分析法,分别进行表面粗糙度的经验建模和理论建模。通过对比预测精度更高的理论预测模型,以理论预测模型为基础,结合粒子群优化算法,以切削效率为目标,进行圆弧面精加工铣削参数优化。针对腹板特征件,选用正交试验设计方法开展腹板精加工实验,研究分析切削参数对腹板表面粗糙度的影响程度和规律。其次,运用BP神经网络算法构建表面粗糙度的经验预测模型。最后,基于粒子群优化算法,结合预测模型,以切削效率为目标,进行腹板精加工铣削参数优化。最后,将各类特征件的实验参数和结果、预测模型以及优化算法和优化结果录入到航空切削参数库系统,为数据库系统的核心功能提供理论、数据和技术支撑。
[Abstract]:As a new type of stainless steel material, 15-5PH stainless steel has been widely used in aircraft driving device, fuselage main beam and landing gear for its excellent mechanical properties, high temperature resistance and corrosion resistance. Its excellent mechanical properties also indicate that the material is difficult to be machined. In the process of processing 15-5PH stainless steel, due to its low thermal conductivity and small elastic modulus, it is easy to appear such phenomena as high deformation resistance, high cutting temperature, fast tool wear and damage, and work hardening, etc. As a result, the processing quality and efficiency of 15-5PH stainless steel are low and the processing cost is too high. Therefore, it is of great significance to study the cutting performance of 15-5PH stainless steel, optimize its technological parameters and develop a special database with independent intellectual property rights to improve the processing efficiency, control the processing cost and process quality. In this paper, according to the functional requirements of the enterprise for cutting database system, we select three features of rough machining, arc surface and web to carry out basic experimental research, modeling of cutting results and optimization of cutting parameters. Then all the experimental data and experimental results are inputted into the cutting parameter database system and debugged and run. Aiming at the problems of serious tool wear and breakage and low tool life during rough machining of 15-5PH stainless steel, single factor experimental design method is used to carry out tool wear and tool life experiment. The wear and failure modes of cutting tools under different conditions were observed and analyzed, and the influence mechanism and law of cutting parameters on tool life were analyzed. The BP neural network algorithm is used to model the tool life and the prediction model of the tool life is established to realize the empirical prediction of the tool life. With the help of particle swarm optimization (PSO) algorithm and aiming at cutting efficiency, the milling parameters of rough machining are optimized. According to surface roughness, orthogonal design method is used to finish machining 15-5PH stainless steel round arc workpiece. The influence of milling parameters on the surface roughness of arc surface is studied and analyzed. At the same time, the empirical modeling and theoretical modeling of surface roughness are carried out by partial least square method and kinematics analysis method respectively. By comparing the theoretical prediction model with higher prediction precision, based on the theoretical prediction model, combining with particle swarm optimization algorithm and aiming at cutting efficiency, the milling parameters of arc finish machining are optimized. According to the characteristics of web, the orthogonal design method was used to carry out the experiments of web finishing, and the influence of cutting parameters on the surface roughness of web was studied. Secondly, an empirical prediction model of surface roughness is constructed by using BP neural network algorithm. Finally, based on particle swarm optimization (PSO) algorithm and combined with prediction model, the milling parameters of web finishing are optimized with cutting efficiency as the goal. Finally, the experimental parameters and results of all kinds of features, the prediction model, the optimization algorithm and the optimization results are inputted into the aeronautical cutting parameter database system, which provides theoretical, data and technical support for the core functions of the database system.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：V261.23

【参考文献】