【作者】 赵朋；

【导师】 周华民；

【作者基本信息】 华中科技大学 ， 材料加工工程， 2009， 博士

【摘要】 本文将人工智能方法与在线检测技术相结合,提出了一种系统高效的塑料注射成形机(简称注射机)工艺参数在线优化与检测方法,本文的研究内容具有十分重要的工程意义和广阔的应用前景。从塑料注射成形过程的系统特征出发,考虑到工艺参数与塑料制品质量之间的非线性、强耦合性和时变性的关系,本文针对实际生产中工艺人员的试模思路,结合人工智能技术在处理弱理论、强经验领域问题的优势,综合运用实例推理和模糊推理技术建立了一种描述注射机工艺参数设置与优化全过程的混合智能模型。实际生产中,非优化的工艺参数将会导致成形周期变长,制品质量稳定性变差,准确获取注射机的各项运行信息是优化成形工艺参数从而提高制品质量的前提。本文开发了一套便携式注射机实时数据采集系统,并将其用于注射机工作过程的在线检测,研究了注射机的运行特征,并基于采集的数据对保压参数进行了优化,使注射机达到最佳工作状态。塑料注射成形是一个复杂的动态变化过程,伴随着温度和压强的剧烈变化,在实际加工中在线跟踪检测塑料熔体的演变规律具有十分重要的意义。本文在详细分析超声检测原理的基础上,利用超声波在异质界面的反射与透射行为以及对温度、压强的响应特性,对注射成形全过程进行了在线检测并对超声信号作出了合理的解释与分析。结合以上研究,本文采用嵌入式编程方法实现了运行在注射机控制器内的工艺参数智能优化系统,该系统能够在试模过程中通过信息反馈自动调整和优化工艺参数,自动修复成形缺陷,以最少的试模次数获得优质的成形制品。实际生产案例表明,该智能系统能成形不同塑料、不同复杂程度和不同浇注系统的塑料制品,与基于经验的“尝试法”相比,大大缩短了工艺设置周期,提高了产品质量,降低了生产成本。更多还原

【Abstract】 In this paper, a new process parameters optimization approach for plastic injection molding machine based on artificial intelligence method and online detection technology is proposed. The study in this paper is of great significance and wide application prospect.Based on the system characteristics of the injection molding process and considering the nonlinear relationship between the process parameters and the quality of the molded parts, a hybrid intelligent model employing case-based reasoning and fuzzy inference has been constructed according to the molding personnel’s thinking during the molding trial-runs and with a view to the characters and advantages of artificial intelligence method. The model can be used to determine the initial process parameters and optimize the process parameters on-line.In practice, non-optimized process parameters results in long cycle time and unstable quality of products. Acquiring all the running information accurately is the base of optimizing the process parameters. In this paper, a portable real-time data collection system for injection molding machine has been developed. The developed system has been used to study the running characteristics of injection molding machine, and based on the collected data, the packing parameters have been optimized.Injection molding is a non-isothermal, non-steady process during which the temperature and pressure change greatly. On-line detecting this behavior is of importance for improving final product performance during processing. In this paper, ultrasonic technique has been employed to monitor the injection molding process. Based on the reflection and transmission behaviors at interface as well as the response for temperature and pressure, ultrasonic technique can provide the information on filling, solidification and detachment during injection molding process.On the basis of above achievements, embedded programming method has been adopted to construct an intelligent system for optimizing the process parameters of injection molding machine. The developed system integrates with the molding machine directly and can be effectively used to set up the process parameters and correct them on-line. Experimental studies show that the intelligent system is applicable to any complex mold geometry and any polymer, and compared with "trial-and-error" method, the system can quickly determine suitable process parameters to produce perfect products at minimum cost.更多还原