【作者】 孟建兵；

【导师】 徐文骥；

【作者基本信息】 大连理工大学 ， 机械制造及其自动化， 2009， 博士

【摘要】 作为现代工业生产中的重要环节,清洗技术在各种金属制品表面处理过程中,扮演着极为重要的角色。清洗技术的选择和清洗工艺质量的好坏不仅涉及能源节约、技术安全、环境保护等重大问题,还直接影响到产品的性能和质量。与传统清洗技术和其它非直接接触干法清洗工艺相比,大气常压等离子体弧清洗作为一种新兴清洗技术,在满足金属零件清洗的需要,代表清洗技术高效、无污染、低能耗等发展趋势的同时,又具有自身的特点和优势,越来越受到国内外学者的高度重视。虽然目前已经就等离子体产生方式、气体种类的选择、清洗质量的评价等方面取得了一定成果,然而,由于大气常压等离子体弧清洗是一个多理论、多学科的交叉领域,涉及到复杂的热物理、化学反应作用过程,因而迫切需要对其清洗机理等关键问题进行深入研究。本文针对基体表面污染物种类以及分布方式的不同,分别对斑状颗粒污染物和层状致密污染物清洗时的能量耦合机制进行了研究。在分析颗粒污染物的粘结力、致密污染物的层裂应变基础上,利用传热学理论和有限元法分别建立了大气常压等离子体弧与上述待清洗工件间的三维热力耦合模型,并对界面清洗力和清洗应变等能量耦合作用进行了有限元分析,进而揭示了弧功率、扫描速度、基体厚度、污染物厚度等工艺参量对界面温度、界面清洗力、清洗应变的影响规律,从而为大气常压等离子体弧清洗过程中实验参量的合理选择提供了依据。利用导热微分方程和阿伦尼乌斯定律,针对厚度方向清洗界面移动、计算区域变化、边界条件非线性的难题,本文建立了大气常压等离子体弧清洗反应动力学模型并确定了频率因子、活化能等反应动力学参数,从而揭示了清洗百分比随清洗时间、金属零件表面氛围温度的变化规律,并进行了表面水滴接触角测量、X射线光电子能谱分析等相关实验验证。结果表明,工件表面污染物的清洗百分比随大气常压等离子体弧清洗时间的增大而增大,直至该污染物被彻底清除;此外,清洗对象等离子体弧热流作用侧的表面氛围温度对清洗百分比有着重要影响,清洗百分比随氛围温度的增加而增加。本文针对大气常压等离子体弧的能量分布显著影响清洗质量的特点,以及直接测量难以获得电流密度分布、温度分布等弧特性的问题,在建立流体动力学方程、MaXwell方程的基础上,通过有限元分析的顺序耦合法、物理环境法,研究了等离子体弧特性分布情况并分析了弧电流、气体流量、喷嘴悬距等工艺参量对等离子体弧特性的影响规律。通过建立基于CCD图像采集和等离子体弧光谱诊断技术的实验验证系统,对相同工艺参量作用下的数值仿真结果和光谱诊断结果进行了比较,其结果较好地验证了该方法的有效性。针对气体放电的雪崩效应、自磁压缩导致的等离子体弧能量集中在较小区域内以及热流密度分布存在较大梯度变化的问题,本文在构建基于外加横向交变磁约束的大气常压等离子体弧清洗能量控制装置基础上,相继建立了等离子体弧摆动幅度模型和热流密度分布数学模型;分析了励磁强度、工作电流、气体流量、喷嘴悬距、励磁波形等工艺参量对横向交变磁场作用下的等离子体弧摆动幅度和热流密度分布的影响规律,并进行了相应清洗实验验证。结果表明,横向交变磁场能有效控制等离子体弧的形态和热流密度分布,从而在保证清洗质量的同时,显著提高清洗效率。此外,弧电流和气流量越大,摆动幅度越小、内部梯度和中心处的热流密度越大;增大喷嘴悬距,则摆动幅度增大、而中心处的热流密度和内部热流梯度减小。本文针对大气常压等离子体弧清洗过程中工艺参数众多,不同工艺参数对清洗质量的影响程度不同以及有些参数之间存在交互性、相关性的问题,在大量清洗实验的基础上,借助主成分分析法和最小二乘支持向量机技术,建立了以表面水滴接触角为评价指标的清洗质量预测模型。在此基础上,利用人工鱼群算法对给定范围下的工艺参数组合进行了优化选择,并与正交试验极差分析法获得的较优参数组合进行了比较。结果表明,基于PCA-SVM的清洗质量预测模型的最大相对误差和MAPE误差分别为3.05%和2.605%;将优选后的工艺参数组合进行大气常压等离子体弧清洗实验,则能够明显降低水滴接触角值,即提高清洗质量。更多还原

【Abstract】 Metal parts are widely used in aircraft, automobiles, ship manufacturing, and etc. Their quality and reliability are determined by the countermeasures of each link in parts processing. As an indispensable link in modern industrial production, cleaning technology plays an important role on the treatment process of metal surface. Besides it can reflect the quality management level of an enterprise, the choice of cleaning methods and the control of cleaning quality not only are involved in energy conservation, technical safety, environmental protection and other major issues, but also have a direct impact on product performance and quality. As a new cleaning technology, atmospheric pressure plasma arc (APPA) cleaning can meet needs of metal parts cleaning and development trend of cleaning technology, such as high-performance, non-polluting, low energy consumption and etc. Furthermore, compared with traditional, other non-direct contact and dry-cleaning technology, APPA has attracted increasing attention of domestic and foreign scholars. Although some achievement have been achieved in a few aspects such as generation methods of plasma, choice of working gas and evaluation methods of cleaning quality, there still exist some problems including cleaning mechanism that are required further theoretical and methodological researches, due to APPA cleaning being a multidisciplinary research field, which merged the thermo-physical effect and chemical reaction.According to substrate surface for the types of pollutants, as well as the characteristics of distribution, a three-dimensional coupled thermo-mechanical model about APPA acting on the cleaned metal parts has been established, by virtue of the theory of heat transfer and finite element method. In this thesis, Energy coupling mechanism of APPA cleaning speckle particle pollutants and layered dense pollutants have been studied respectively. Moreover, the effect of process parameters such as APPA power, scanning speed, substrate thickness and contaminant thickness on the interface temperature, interface cleaning force and cleaning strain has been revealed, so as to provide theoretical basis for the choice of process parameters during atmospheric pressure plasma arc cleaning metal parts.According to the mobile cleaning interface, changeful calculation region and non-linear boundary conditions along the thickness direction, in this thesis, a mathematical model of reactive kinetics in the metal surface contaminant cleaning using APPA has been developed, by means of thermal conduction differential equation and Arrhenius equation. Afterwards, reactive kinetics parameters such as activation energy and pre-exponential factor are calculated. On this basis, the intrinsic relationship between contaminant removal percentages, removal rate and influencing factors such as cleaning time and ambient temperature of APPA cleaning are revealed and validated with related experiments results. The results indicated that contaminant removal percentages increase with cleaning time increasing until the contaminant is drastically cleaned by APPA. Furthermore, the ambient temperature of APPA on the contaminant surface affects the removal percentages strongly. The removal percentages increase with the increase of the ambient temperature. To avoid the damage of metal substrate surface because of higher temperature and ensure the removal rate of the contaminant, the appropriate temperature which lies between the contaminant decomposition temperature and damage temperature of metal substrate.According to the fact that energy distribution of APPA having a significant effect on cleaning quality and arc properties including current density distribution and temperature distribution being difficult to achieved by direct measurement methods, a three-dimensional axisymmetric mathematical model, including the influence of the swirl exiting in the plasma torch, has been developed to describe the heat transfer and fluid flow within a combined plasma arc with magneto-hydrodynamics equations and Maxwell equations. In this model, a mapping method and a meshing method of variable step-size are adopted to mesh the calculation domain and to improve the results precision. To overcome a problem from the coexistence of non-transferred arc and transfer arc and the coupling between electric, magnetic, heat flow and fluid flow phenomena in the combined plasma arc, a sequential coupling method and a physical environment approach are introduced into the finite element analysis on jet characteristics of the combined plasma arc. Furthermore, the jet characteristics of combined plasma arc such as temperature, velocity, current density and electromagnetic force are studied; the effects of working current, gas flow and the distance from the nozzle outlet to the anode on the distributions of temperature, velocity and current density are also revealed. Compared with the collection and diagnosis on the combined plasma arc by CCD, the results show that the simulated value appears to be in good agreement with measured value under the conditions of the same process parameters.According to the problem that the energy distribution of plasma arc being limited in smaller region and the temperature gradient of plasma arc having more variation in the arc column owing to the avalanche effect of gas discharge and the constraint of plasma arc by its own magnetic field, an external transverse alternating magnetic field is applied to APPA to create a plasma for cleaning a large metal surface. On the basis, two mathematical models are developed to describe the oscillating amplitude of the plasma arc root along the metal surface and the heat flux density distribution of plasma arc on the metal surface, respectively. The behavior of plasma arc under the external transverse triangular alternating magnetic field imposed perpendicular to the plasma current is discussed, and the effect of process parameters such as working gas flow rate, arc current, magnetic flux density and the distance from the nozzle outlet to the anode workpiece on the form and heat flux distribution of plasma arc are also revealed and validated with related experiments results. The results show that it is feasible to control the shape and heat flux density of APPA with the external transverse alternating magnetic field, which can expand the region of plasma arc thermal treatment and flatten the heat flux density upon the workpiece, so as to ensure APPA cleaning quality and to simultaneously improve the cleaning efficiency of APPA cleaning. Furthermore, the oscillating amplitude of plasma arc increases and the heat flux density gradient upon the workpiece decrease with the magnetic flux density enhancing. However, an overly strong magnetic field results in the plasma arc unstable. Under the same magnetic flux density, less gas flow rate and arc current, more distance from the nozzle outlet to the anode cause the oscillating amplitude to increase. Contrarily, the more gas flow rate and arc current, the more heat flux density peak increase. Moreover, more distance from nozzle outlet to workpiece descends the heat flux density peak.According to the fact that APPA cleaning being a complicated and non-linear process, cleaning quality being influenced by process parameters and the interaction of different parameters being complex, based on many experiments on APPA cleaning, an intelligent predictive model of the non-linear relationship between cleaning quality (water contact angle of cleaned surface) and process parameters is established with the research of Least Squares Support Vector Machines (LS-SVM) and Principal Component Analysis (PCA). Afterwards, Artificial Fish Swarm Algorithm (AFSA) is introduced to optimize established control model of cleaning quality and to obtain the optimum process parameters under a certain range of these parameters. The results indicate that it is feasible to apply PCA-SVM and AFSA in forecasting the cleaning quality and determining the process parameters of APPA cleaning. The maximum relative error and MAPE error are 3.05% and 2.605%, respectively. It is obvious to reduce the water contact angle of cleaned surface and to improve cleaning quality of APPA cleaning with above optimum process parameters.更多还原