【作者】 杜力；

【导师】 刘飞飞；

【作者基本信息】 江西理工大学 ， 机械电子工程， 2007， 硕士

【摘要】 如今商品的竞争就是质量的竞争,制造企业所面临的挑战就是看能否保证连续生产出高质量的产品。因此,企业迫切需要一种实施质量控制的智能辅助工具,同时这种智能辅助工具及其所应用的质量控制方法又必须同现代生产条件和生产模式的转变相适应,才能发挥真正的作用。在现代制造条件下,先进制造技术、先进加工设备以及先进检验设备的大量采用,对现场质量控制提出了更高的要求;同时随着市场需求的变化,多数企业又不约而同地把生产模式由“批量生产”模式逐步调整到“多品种、小批量生产”模式,大批量的SPC方法在小批量的生产中已经不能适用。针对这些问题,本文设计了一种嵌入式的工序质量SPC系统,并进行了硬件和软件的开发,同时研究了适于嵌入式系统的小批量SPC方法及其控制图识别方法。首先,本文分析了当前多品种小批量生产制造模式下,实施统计过程控制所面临的困难,探讨了一种适合在嵌入式系统中应用的小批量SPC方法。接着,对控制图的识别方法进行研究,提出一种基于BP神经网络的控制图识别方法,设计了该方法在嵌入式系统中应用的实现方案。最后采用ARM9内核的S3C2410芯片作为中央处理器设计了嵌入式工序质量系统,使系统具有较高的运行速度和数据处理能力;同时为使监控系统能够接受多种接口形式的在线检测数据,系统配置了RS232、USB、RJ45、IIC等通信接口;而且系统还配置了LCD和键盘接口,以便于人机交互。系统的软件设计采用了微软的Windows CE操作系统,运用Emebedded Visual C进行了质量控制软件的开发,质量控制软件具有Windows风格的图形用户界面、过程能力分析、控制图分析、实时诊断、质量数据的储存和管理、数据自动采集以及与网络通讯等功能,构成了完整的工序质量SPC系统。经应用验证表明:本文研究的嵌入式工序质量SPC系统具有功能完善、结构紧凑、使用方便并具有良好的系统可扩展性;无论从技术上,还是从经济性,都是一种可行的实用系统。更多还原

【Abstract】 Now the competiton of merchandise is the competition of the quality.The mass challenge confronted with by the enterprises is to make high quality product.Therefore, the enterprise requires one kind of the intelligence assistant implement urgently to carry out qaulity control for machining operations, further more, the implement and the quality control method of it for the manufacturing process must adapt to the transition of the morden manufactur condition and the morden manufacture pattern. Under modern fabrication condition, a great quantity of advanced technology and manufacture equipment along with detecting device have been adopted, so high requirement to the quality control for the machining operations has been brought forward; as the chang of market demand the production pattern of most enterprise has transitted to small batch from mass batch, the traditional means of SPC is not suitable for the quality control of small batch. Aiming at these problems, the paper designed a kind embedded system of spc for machining operation ,the hardware and software of it have also been developed, meanwhile the paper proposes a quality control method for the small batch production in the ebedded system, and studys the way of pattern recognition of abnormal control charts in embedded system also.Firstly, the paper analyzes the difficulty of the application of SPC in the small batch production, and discuses a method of SPC in the small batch production. Then the paper studys the means of pattern recognition of abnormal control charts, put forward a method of pattern recognition of abnormal control charts based on BP neual network, and contrives its application method in the ebedded system. Exponentially weighted moving average (EWMA) and fuzzy algorithm for the input samples are also developed to improve its recognition accuracy. Lastly, the paper designes a kind embedded system of SPC for machining operation by micro-process unit S3C2410 with ARM9 kenerl, the system is with high performance and with high speed configured with communication interface of RS232、USB、RJ45、IIC;LCD, keyboard and mouse is also configured to benefit the interaction between humen and computer.The system adopts the Windows CE.NET as its operation system, the quality control software is developed by using embedded visual c++ language, the software is with graphic user interface and it is with the functions of control charts analysis,process capability analysis, real-time diagnosis, automation data acquisition, data storage and management along with the capability of intranet communication, which consist a integral SPC system for machining operation.The result manifests that the system is perfect in function, integral and tight in structure, it is safe, low cost, easy to use and extendable.The system is feasible and steady, it is excellent not only in technology, but in economy also.更多还原