【作者】 孙宪萍；

【导师】 王雷刚；

【作者基本信息】 江苏大学 ， 材料学， 2008， 博士

【摘要】 挤压技术具有“高效、优质、低消耗”的优点,在技术上和经济上都有很高的实用价值。已广泛应用于机械、仪表、电器、轻工、宇航、船舶、军工等工业部门,已成为金属塑性成形技术中不可缺少的重要加工手段之一。但挤压工作状态复杂,模具寿命低,制约了挤压技术的应用和发展。随着新材料和新产品的开发,对挤压模具寿命提出了更高要求。模具寿命是一个综合性的技术问题,在挤压成形过程中,磨损是影响模具寿命的决定性因素,尤其在高温成形过程中,模具因磨损而失效的情况超过70%。模具的磨损不是材料的固有特性,与工况条件、模具材料、坯料材料、表面形貌、接触方式、润滑方式等多种复杂因素有关。模具的磨损是一个热(温度)-力(摩擦力)-化学介质(润滑剂)多因素相互耦合作用下的非线性动力学问题,涉及到塑性力学、摩擦学、金属学、化学、热力学等多门学科的交叉知识,至今未形成统一的理论体系。因此研究挤压模具的磨损机理及其控制方法,对提高挤压模具寿命,丰富和发展塑性加工摩擦学具有重要理论意义和现实意义。本文从挤压模具磨损的微观机理出发,探索摩擦力的形成机制和磨损计算方法,分析微凸体的瞬时摩擦温升,开展模具型腔等磨损优化设计和磨损控制方法研究,以全面提高挤压模具的寿命。具体研究工作如下:1.首先基于Hertz接触理论,引入塑性方程,建立微观粗糙表面弹塑性接触模型,利用粘着摩擦理论分析摩擦力形成机制,研究挤压过程中微观粗糙表面几何形貌对摩擦系数的影响;然后将Hertz弹性接触理论和热传导的基本理论相结合,研究两粗糙表面相对滑动过程中摩擦引起的微凸体瞬时温升分布情况。2.采用热力耦合有限元法计算由坯料塑性变形和坯料与模具间的摩擦引起模具型腔表面的本体温升,并与人工神经网络相结合,用有限元模拟软件分析的数据作为学习样本训练所建立的神经网络模型,以此模型预测挤压模具型腔表面的本体温升,以提高温升计算速度,为模具温升模型的建立奠定基础。3.将有限元分析、神经网络和遗传算法相结合,应用于挤压模具型腔优化设计。采用B样条函数插值描述凹模型腔轮廓形状,用有限元数值模拟获得型腔表面节点的应力场、速度场和温度场,基于修正Archard磨损模型计算型腔磨损深度,以此作为样本训练BP神经网络,建立模具型腔控制点与磨损深度之间的映射关系,再以等磨损为目标,采用序列二次规划法和遗传算法优化模具型腔轮廓形状。4.采用比拟实验研究在模具钢表面磁控溅射和离子镀TiN系列涂层的摩擦磨损性能,利用扫描电镜和三维形貌仪检测添加稀土元素Y对涂层耐磨性能的改进效果,并探讨其机理。更多还原

【Abstract】 Extrusion technology has the advantages of high efficiency, good quality and low-energy consuming, which has high practical value in the aspect of technology and economy. Extrusion technology has been widely used in machinery, instruments, electrical apparatus, light industry, aerospace, ship, military industry, and other industrial sectors, which has become one of the indispensable means of processing in metal forming field.However, working conditions are complicated in extrusion forming process and die service life is short, which constrain the application and development of extrusion technology. With the development of new materials and products, higher demands are proposed for die service life. Die service life is a comprehensive technology problem. And wear is the predominant factor which affects the die service life, especially at high temperature the failure is caused by wear in over 70% cases during extrusion forming process. Die wear is not the natural characteristic of material, but related to working conditions, die and workpiece material, surface monograph, contact type, lubrication mode and other complicated factors. Die wear is a nonlinear dynamic problem of multi-factor coupling, which involves thermal (temperature) - force (friction) - Chemical (lubricant). The wear is related to plasticity, tribology, metallography, chemistry, thermodynamics and other subjects of cross-knowledge, and has not formed a unified theoretical system. Therefore, the research of wear mechanism and wear control method for extrusion die have important theoretical and practical significance for improving the die service life, enriching and developing the tribology of plastic processing.In the aspect of the micro-mechanism of die wear, friction mechanism and calculation method of wear were explored, and the transient friction temperature rise of asperity was analyzed. Wear optimum design of extrusion die cavity and wear control method were researched to improve die service life. Specific research work is as follows:1. Based on Hertz contact theory, the model of microscopic elastic-plastic contact was established and friction mechanism was analyzed by adhesive friction theory. Effects of geometrical monograph of microcosmic rough surface on friction coefficient were researched. Then, combining the elastic contact Hertz theory and the fundamental theory of thermal conduction, the transient temperature rise distribution on asperity, which was caused by friction during relative movement between two rough surfaces, was calculated.2. Body temperature rise on die cavity surface caused by plastic deformation of workpiece and friction between die and workpiece was calculated by finite element method (FEM). Combining artificial neural network, the established neural network was trained with the simulation results as learning samples. The model was used to predict the body temperature rise of die cavity surface and improve the calculation speed of temperature rise, which lays the foundation for the establishment of temperature rise model of die cavity surface.3. Finite-element method, BP Neural Network and genetic algorithms were combined together to optimize extrusion die cavity. The method of B-spline function interpolation had been used to describe extrusion die cavity profile. The temperature, pressure and velocity field of nodes on the cavity surfaces were gained by FEM simulation. Wearing depths of extrusion die profile were calculated by modified Archard theory. The results were used as samples to train BP neural network, so that nonlinear mapping relation between reference point of die profile and wearing depth was established. In order to obtain uniform wearing depth, sequential quadratic programming and genetic algorithms were applied to optimize cavity profile of the extrusion die.4. Friction and wear properties of die surfaces, which were treated by magnetron sputtering and ion coating TiN, were studied by analogy experiment. The improving effect and mechanism of the wear resistance of coating by the adding rare earth element Y were investigated by scanning electron microscope (SEM) and 3-D topography instrument.更多还原