运载火箭贮箱大型网格壁板镜像铣削三维工艺建模与快速编程理论方法

This project addresses the main challenges and requirements for the precision machining of net-patterned wallboards of storage containers of the new generation carrier rockets and heavy carrier rockets of our country, focusing on the aluminum net-patterned wallboards with complex features, which are of large size, thin wall and high accuracy requirement, in small batches and even single-piece production. The project aims at the key issues in the most advanced “Mirror Milling” technology, particularly the modeling and reuse of the machining process for large thin walls, including: 1) the definition and 3D modeling of machining features; 2) intelligent numerical control programming driven by machining features; and 3) real time automatic wall thickness compensation and pro-active control of the rigidness of the whole machining system. This project will carry out in-depth theoretical study, key technology development, experiments and real applications. New definition for dynamic machining features and 3D process modeling methods will be proposed, by considering the interactions between the cutting process, wall thickness inspection and the responsive support mechanism, and the characteristics during the whole milling process of net-patterned wallboards with mirror effect quality. By revealing the inter-constraining relationships among the structural and kinematical characteristics of the complex mirror milling equipment, the machining process, and the paths of the cutting tool, a full suite of methods will be proposed for the intelligent numerical control programming, wall thickness compensation in real time and pro-active rigidness control of machining system, all driven by the models of dynamic machining features. The expected achievements include the numerical control programming efficiency improved by more than 3 times, the machining accuracy of wall thickness within ±0.1mm and surface roughness within Ra 1.6. The outcome of this project would also provide theoretical and technological support necessary for the accumulation and reuse of machining process knowledge, which is critical to the intelligent machining of large-sized, thin-walled complex structural parts of small batches and even single-piece production.

本项目面向我国新一代和重型运载火箭贮箱网格壁板加工的重大需求，以大型薄壁、特征复杂、壁厚精确、小批量甚至单件生产的铝合金网格壁板为研究对象，瞄准“镜像铣”这一大型壁板加工的最先进技术，围绕工艺建模和重用的关键问题： 1）加工特征定义与三维工艺建模；2）加工特征驱动的快速数控编程；3）壁厚实时自适应补偿和工艺系统主动刚度控制，从理论方法、关键技术、试验与应用等层面展开深入研究。提出考虑镜像铣工艺中切削加工、壁厚检测和随动支撑的交互作用及其演变特性的动态加工特征定义与三维工艺建模方法，揭示镜像铣加工装备复杂结构和动力学特性、加工工艺与刀轨之间的约束关系，提出动态加工特征驱动的快速数控编程、壁厚实时自适应补偿和工艺系统主动刚度控制方法，编程效率提高3倍以上，达到壁厚精度±0.1mm、表面粗糙度Ra1.6的质量目标，为小批量甚至单件生产的大型复杂构件工艺知识积累、重用及智能加工提供理论和技术支撑。

铝合金网格壁板是新一代运载火箭和重型运载火箭中最重要的薄壁件，是火箭制造中数控加工量最大的零件，其壁厚精确控制是贮箱网格壁板高效加工的关键。镜像铣削技术是大型铝合金网格壁板加工的最新技术和发展趋势，而镜像铣的高精度加工工艺和快速编程技术亟需突破。经过分析和研究，本项目对以下三个问题进行了重点研究：.针对加工全过程工艺知识建模难题，建立了加工动态特征概念，提出并证明了可定义为同一加工特征的充分必要条件，进而建立了加工全过程工艺知识模型，为工艺知识建模由以零件为载体到以加工动态特征为载体的模式转变奠定了理论基础。.针对镜像铣多工艺约束引起的刀轨生成难题，提出了基于图像的加工特征刀轨自动生成方法，建立了驱动几何的图像表征以及刀轨的图像栅格化表征，将刀轨整体优化问题转化成图像处理问题进行求解，实现了网格壁板零件镜像铣削快速数控编程。.针对网格壁板镜像铣过程中的切削稳定性和加工变形问题，提出了机床全工作空间刀尖模态预测方法、瞬态铣削稳定性分析方法和镜像铣削时滞系统壁厚闭环控制方法，实现了网格壁板零件镜像铣削自适应加工。.项目在研期间，发表论文34篇，SCI收录30篇，其中1篇为2019年度ESI高被引论文。申请国家发明专利18项，获授权12项，获软件著作权3项。出版专著1部。获国家标准正式立项1个。受邀作国际学术会议大会主题报告1次，主办国际学术会议/研讨会5次。出版Special Issue 2部。研究成果在上海航天设备制造总厂等航空航天制造企业的多项运载火箭的网格壁板和箱底零件的加工中进行了应用验证。获国家技术发明二等奖1项（项目负责人第一完成人）、国家科技进步二等奖1项（子课题负责人第二完成人）、国防科技创新团队奖1项（项目负责人为核心成员）、四川省科技进步一等奖1项等。.项目负责人获国家自然科学基金杰出青年基金和国家自然科学基金创新群体项目1项（核心成员），被评为“长江学者”特聘教授。项目组成员1人被评为“长江学者”青年学者。培养博士研究生18名，毕业6人，1人获江苏省优秀博士学位论文，1人获上银优博佳作奖。培养硕士研究生26名，毕业12人。