【作者】 杨德明；

【导师】 许振明；

【作者基本信息】 上海交通大学 ， 环境工程， 2011， 硕士

【摘要】 废弃电路板是废旧电子电气产品的重要组成部分,因其含有大量金属,如铜、铝、铁、镍、铅、锡、锌、金、银、钯、铑等,具有重要的回收价值。同时,还含有重金属、有机物等有毒有害物质,处理过程中存在环境风险。因此,废弃电路板的回收和处置从资源和环境的角度来说有着重要的意义。废弃电路板物理法资源化工业生产线包括四个部分:多级破碎、物料分级、静电分选、除尘设备,尽管工艺工业化的过程中已经解决了很多问题,但是在不同工艺之间的配合上面还是存在一些问题。如果粉碎和分级系统不能得到很好的控制,整个生产线将会面临几个问题:效率过低、金属非金属未有效解离、颗粒过破碎、破碎不充分、堵料和设备过热等。为此,本文研究了废弃电路板冲击粉碎动力学特性,实验分析了物料参数对粉碎效果的影响,并根据生产线实际状况建立了破碎分级循环过程模型以及神经网络粒度预测模型,为废弃电路板破碎分级生产线的工业化应用提供了优化和控制新方法。对粉碎后的电路板颗粒应用不同的分布模型进行粉碎分布特性的分析,发现罗辛-拉姆勒(R-R)分布能更精确地描述废弃电路板冲击破碎的分布规律,说明此模型可应用于预测废弃电路板冲击破碎的分布特性。根据粉碎过程及粒度分布状况建立了精度高,适用性强的一级粉碎动力学模型,对电路板冲击粉碎的生产过程有一定的指导意义。对影响电路板冲击粉碎性能的影响因素进行了分析,得出主要的影响因素。对进料颗粒粒径与进料质量对电路板冲击粉碎特性的影响作了实验分析,得出在描述废弃电路板颗粒冲击粉碎粒度分布的R-R模型中c、m值的变化规律。采用将粉碎颗粒粒级矩阵化的方法,为废弃电路板资源化生产线建立了破碎分级矩阵模型,并与实际生产线对比,确认了模型的有效性。运用MATLAB软件,将建立的模型应用于生产线的稳定状态分析。结果表明,3次循环即可让生产线达到稳定状态,生产线有着很好的应变性和很小的波动性。将建立的模型应用于不同系统总进料量分析,发现各部分物料量与进料量有着非常好的线性关系,该模型为生产线过程控制自动化奠定了基础。针对废弃电路板破碎分级粒度控制进行神经网络原理的理论分析,选用BP神经网络作为建模类型,使用MATLAB设计并训练了BP神经网络,并证明网络的收敛性。对仿真结果进行误差分析,表明这一神经网络模型能较好地模拟实际生产过程,为控制器的设计奠定了基础。更多还原

【Abstract】 Waste Printed circuit boards (PCBs) are recognized as an important part of waste electronic products, it contains a lot of valuable metals, such as copper, aluminum, iron, nickel, lead and tin, zinc, gold, silver, palladium, rhodium and others,so it has important recovery value. At the same time, it also contains heavy metals, organic and other toxic or harmful substance, the treatment process contains environmental risk. Therefore, the recycle and disposal of waste PCBs have momentous significance from the perspective of resources and environment. Chemical and mechanical methods are two traditional recycling processes for waste PCBs. The whole technology of waste PCBs physical recycling industry line contains four parts: multiple scarping, material screening, multiple-roll corona electrostatic separator, and dust precipitation. Although many problems in the industrialization of technology were solved, there are still some problems blocking the integration of different technologies. If the grinding and classification cyclic system is not well controlled, the whole process will face problems such as metals and nonmetals insufficient dissociation, particles over-pulverizing, incomplete comminuting, material plugging and equipment fever.Therefore,this paper studies the kinetics characteristics of waste PCBs impact crushing, and material parameters on crushing effect was analyzed with experiment. According to the actual conditions of production line, crushing and classification cyclic model and neural network size prediction model were established, it provides a new method for optimization and control of the industrial application of waste PCBs production line.With crushing matrix method, a grinding and classification matrix model for waste PCBs automatic production line was built. By contrast with actual production line, the validity of the model was verified. With software of MATLAB,the model was applied for steady-state analysis of the production line, it was found that only 3 cycles are needed for the equilibrium of the production line, which indicated that the production line have a strong adaptability. The established model was applied for analysis according to different weight of feeding material, material flow in each part presented a linear relationship with the feeding material. The model provides a good foundation for automatic process control of the production line.According to waste PCBs crushing and classification granularity controlling,analysis the principle of neural network theory,and BP neural network is choose as modeling type, MATLAB is used to design and train the BP neural network, and the convergence of the network is proved. The result of simulation error analysis shows that the neural network model can simulate the practical production process, which laid a foundation for the controller design.According to crushed PCBs particles, different distribution model are applied to analysis distribution characteristics, it is found that R-R distribution can more accurately describe the waste PCBs impact crushing regularity distribution, it illustrate that the model can be applied to predict the impact crushing waste PCBs distribution characteristics. According to the crushing process and particle size distribution, first level crushing dynamic model with high accuracy is established, which has certain directive significance to waste PCBs crushing production process. Factors influence the PCBs crushing performance were analyzed, and got the main influence factors. The feeding material weight and particle size on the impact crushing characteristics were experimentally analyzed, and got c, m value change rule of R-R model which describe the waste PCBs particles impact crushing distribution.更多还原