【作者】 黄云；

【导师】 梁锡昌；

【作者基本信息】 重庆大学 ， 机械工程， 2009， 博士

【摘要】 核电高压容器材料本身的难加工特性导致其磨削效率低,成本高,另外加工过程不能添加冷却润滑的特殊要求使得加工过程极易引起表面烧伤及其它表面损伤等不良现象产生,同时这些容器构件体积与重量巨大,磨具需要适应不同空间姿态调整和进给,加之工业发达国家对研发高效砂带磨削装备完成上述核电容器多道磨削工序的方法及其相关工艺参数严格保密与技术垄断,导致我国对核电容器制造过程中涉及上述关键磨削工序均采用人工进行砂轮修磨与抛光,其加工效率和精度都极低,且工人的工作环境极端恶劣和危险程度极大,这些难点严重制约了我国核电装备国产化制造进程。本文从这种现状出发,基于砂带磨削的“万能”和“冷态”特性思路,提出了一种核电高压容器高效磨削工艺方法,研制了新型强力高效砂带磨削自动化加工装备,创新性地提出了基于BP神经网络的核电容器材料高效强力砂带磨削工艺参数预测与优化方法,并将其应用于核电高压容器的高性能磨削,对实现我国核电高压容器国产化制造瓶颈的突破起到了重要作用。本文主要完成了以下几项工作:1.阐明了核电高压容器高效磨削所涉及的关键基础问题,搭建了高效强力砂带磨削工艺基础试验平台。同时利用有限元方法建立了强力砂带磨削过程中接触轮与工件接触模型;确立了砂带单颗磨粒切刃切除材料理论模型并开展了强力砂带磨削去除材料试验,探讨了不同磨削参数对核电容器材料去除率的影响,为实现现场磨削参数工艺优选调控提供理论及应用的研究依据。2.利用现代理化测试分析手段,率先完成了对两种国产核电高压容器典型材料因科镍690合金钢和20Cr-10Ni高效砂带磨削的综合测试分析研究工作。首次明确提出在砂带磨削核电容器材料进入稳定磨损期内磨具磨损形式是磨耗磨损和部分的破碎磨损并存的新观点,这一新观点对正确认识强力砂带磨削过程磨粒磨损的本质及其形成机理具有重要的启示作用。3.研制了实现核电高压容器强力磨削用的多功能高效磨削自动化装备样机,结合相关单元及辅助装置研制,并利用CAD/CAE手段进行该机系统化结构优化设计。同时为确保该新型磨削加工装备复杂控制功能,开展了高效强力砂带磨削加工装备的相关数控系统开发研究,为后续项目研究实现多坐标联动的高效强力砂带磨削智能控制奠定了完备的机电硬件基础。4.完成了基于BP神经网络的核电容器材料高效强力砂带磨削工艺参数预测模型建立,重点对因科镍690合金钢的磨削工艺参数进行了预测分析,砂带粒度、砂带速度预、磨头进给速度、磨头施加压力预测值与实测值间的误差均在可接受范围,大大缩短了装备样机进行现场应用磨削调试的时间,其BP工艺预测模型的建立对解决国产核电高压容器磨削工艺现场应用具有重要的指导意义。更多还原

【Abstract】 Nuclear power high-pressure vessel materials have the characteristic of difficult process, which leads to low grinding efficiency and high cost, and another special requirement of the process can not add the cooling lubricant, which would make the surface of the workpiece to be burned and generate other negative phenomena easily, such as the surface damage and so on. At the same time, the nuclear power high-pressure vessel’s size is very big and it’s weight is huge, so it needs to adapt to different spatial attitude adjustment and the feed to grinding tool, combined with the industrial developed countries monopolize the technology which could efficiently complete the grinding vessel including the abrasive belt grinding equipment and the process parameters. These factors resulted in the China’s nuclear vessel’s process by manual grinding and polishing which involved in the above key grinding processes with grinding wheel, and simultaneously the processing efficiency and accuracy are very low, and the working environment is extremely poor and dangerous, these difficulties seriously restrict the localization of China’s nuclear power equipment manufacturing process. In this paper, it presented a high-efficient grinding method to the nuclear power high-pressure vessel, and simultaneously it developed a new type of heavy and high-efficient automated processing of the abrasive belt grinding equipment, which is based on the characteristics of abrasive belt grinding "universal" and "cold state", creatively presented a efficient forecasting parameters and optimization methods to the heavy abrasive belt grinding process for the nuclear power high-pressure vessel materials which based on BP neural network, also it applies to the high-performance grinding for nuclear power high-pressure vessels, plays an important role in the breakthrough of the China’s nuclear power high-pressure vessel localization manufacture bottlenecks.In this paper, completed the following tasks:1. Clarified the critical infrastructure issues to the nuclear power high-pressure vessel which involves in high-efficient grinding; built a heavy and high-efficient test platform for the basis research of the abrasive belt grinding process. Simultaneously, established a contact model with contact wheel and the workpiece in the process of abrasive belt grinding which based on the finite element method; established the cutting-edge theoretical of the removal of material to a single abrasive grain and conducted a heavy abrasive belt grinding to the removal of material testing, and then discussed the impact of the nuclear vessel material removal rate to the different grinding parameters, provided theoretical and applied research to the optimization control parameters for the realization of grinding process in industrial field.2. By using of modern physical and chemical test analysis methods, first completed comprehensive test and analysis research work of high-efficient abrasive belt grinding to the nuclear power high-pressure vessel made of two kinds of typical materials Inconel 690 alloy and 20Cr-10Ni. For the first time, explicitly presented the new idea that the form of abrasive belt wear are made from both grinding abrasion and parts of grain broken wear when abrasive belt grinds nuclear power vessel materials into the steady wear period. The new idea plays an important role in correct understanding the nature of abrasive grain wear and formation mechanism in the heavy abrasive belt grinding process.3. Developed multi-function high-efficient grinding automation equipment prototype to grind nuclear power high-pressure vessel, combined with the relevant units and auxiliary equipment, using CAD / CAE to make the grinder systematic design and structure optimization. At the same time, to ensure the new type of grinding equipment complex control functions, developed the numerical control system for high-efficient heavy abrasive belt grinding equipment. To lay a complete hardware basis for follow study multi-axis high-efficient heavy abrasive belt grinding by intelligent control.4. Established the nuclear power vessel materials high-efficient heavy abrasive belt grinding processing parameters prediction model based on BP neural network with emphasis on grinding processing parameters prediction analysis of the Inconel 690 alloy. The deviation between the predicted and measured results of abrasive belt grain size, abrasive belt velocity, grinding head feed velocity and grinding head pressure are within an acceptable range, greatly reduced the grinder debug time in industrial grinding field. The establishment of BP prediction model has important guiding significance to solve nuclear power high-pressure vessel grinding field application.更多还原