注塑模的智能制造工具外文文献翻译、中英文翻译

Intelligent Manufacturing Process Tool For Plastic Injection Molding Aravind Kumbakonam, Terrence L. Chambers, Suren N. Dwivedi Department of Mechanical Engineering University of Louisiana, Lafayette Bill Best Ash Industries Aravind Kumbakonam, Terrence L. Chambers, Suren N. Dwivedi Bill Best Ash IndustriesAbstract This paper presents an overview of ongoing research aimed at the development of a Computer-Based Intelligent Manufacturing Process Tool, at the University of Louisiana at Lafayette. The Manufacturing Process Tool is a computer program, which would help the manufacturer in solving problems associated with Injection Molding. These problems include long process set up time, non-optimized cycle time, and poor control of the molding process. The Manufacturing Process Tool would eventually help the machine operator (who need not be an expert) in setting up, optimizing, and controlling the Injection Molding Process; thus maximizing the production rate on that particular Injection Molding machine. Introduction Plastic Injection Molding is the worlds most common method of producing complex commercial plastic parts with excellent dimensional tolerance. According to the C-mold design guide, 32% by weight of all plastics processed go through Injection Molding machines, making Plastic Injection Molding one of the most important manufacturing processes. It is seen that the final molded part quality is chiefly dependent on the type of material, mold design and the molding process settings. Once the material and the mold to be used are specified, the part quality basically depends on the molding process. The molding process is quite complex involving many variable process parameters like pressure, temperature and time settings. These process parameters have to be optimally set in order to improve part quality and maximize the production capacity of the Injection Molding machine. Educated and experienced individuals are required to set up and optimize such a complex process. These individuals control the molding process on a trial and error basis, which is usually time consuming. This method of controlling the molding process relies heavily on operator intuition and a few “rules of thumb,” which the operator develops over a period of time while working with different materials, pressures, temperatures and time settings. This paper presents an outline of ongoing research at the University of Louisiana at Lafayette involving the development of Intelligent Knowledge-Based Engineering Modules (IKEM) for . IKEM has different modules, namely: Parsing, Mold Design, Cycle Time and the Manufacturing module, which are linked to each other. This paper mainly concentrates on the Manufacturing module, which involves the development of an Intelligent Manufacturing Process Tool called “The Optimizer.” Figure 1 and 2 show the transfer of data between the different modules of IKEM.Figure 1.Intelligent Knowledge Base Engineering Modules For Plastic Injection Molding Figure 2. Data Flow Diagram The Optimizer captures non-deterministic knowledge in the Injection Molding process from an expert in this field, and also uses deterministic knowledge available in the form of relations, look up tables, etc. The Optimizer is written in Visual Basic, and would assist the machine operator in setting up, optimizing and controlling the Injection Molding Process and thus maximize the production rate on that particular Injection Molding machine. As shown in Figure3, The Optimizer helps the manufacturer in the set up of the molding machine, by giving him the initial optimal process parameter values. These values could be later fine tuned for personal benefits by the operator using his intuition and guesswork.Figure 3. Product Development Vs Time图3：产品开发与时间Knowledge-Based Engineering Modules An expert system, or a knowledge-based system, is defined as “a model and associated procedure that exhibits, within a specific domain, a degree of expertise in problem solving that is comparable to that of a human expert” Of the different kinds of expert systems available, which have their individual advantages and disadvantages, the “Rule Based” type of knowledge-based system is the one that is most commonly used. It basically consists of a knowledge base, containing a set of “IF-THEN” statements, called “rules.”The IKEM project deals with creating an expert system containing a set of IF-THEN rules collected from the human expert with the help of knowledge engineers. Two of the most important issues that are essential to the reliability of this rule-based type expert system are, the Knowledge Acquisition and Knowledge Representation. Knowledge Acquisition This is the initial approach wherein the knowledge engineers extract the rules from the human expert. This step is quite labor intensive, and has been considered the bottleneck in the expert system development process. Knowledge acquisition mainly depends on the skills of the knowledge engineer. His primary aim is to extract strategies or rules of thumb from the human expert(s) and transfer it to the knowledge base. Care has to be taken during the knowledge acquisition process, as it directly affects the knowledge representation scheme, later. Its seen that there are two basic types of Knowledge Acquisition: 1. Knowledge Acquisition directly from the human expert (non deterministic knowledge) 2. Knowledge Acquisition thorough previous cases, relations, look up tables (deterministic knowledge). The Intelligent Manufacturing Tool being developed intends to capture nondeterministic knowledge which is gained by experience in the field, as well as more deterministic knowledge available in the form of relations, look up tables, etc.A student knowledge engineer from the University of Louisiana at Lafayette has been working in conjunction with Ash Industries, Lafayette, which is primarily a Plastic Injection Molding plant. The knowledge engineer interviews the human expert at Ash Industry. He then organizes this extracted data in a logical fashion. This knowledge extracted from the expert is in the form of heuristics, or more precisely “rules of thumb.” The expert develops these heuristics or rules of thumb intuitionally and from his prior experience in this field. These “rules of thumb” act as the guidelines, using which; the molding process is operated for the most optimal product quality. A typical example of a set of “rules of thumb” or heuristics developed by the expert useful in the calculation of clamp tonnage is: Rule 1: IF part wall thickness = 0.04 inch And material = crystalline THEN clamp tonnage = 2.0 ton/ inch2 * Rule 2: IF part wall thickness = 0.04 inch And material = amorphous THEN clamp tonnage = 3.0 ton/inch2 Rule 4: IF part wall thickness 0.04 inch And material = amorphous THEN clamp tonnage =3.6 ton/inch2l Inch2 represents the cross sectional area of the total number of parts in the mold that are perpendicular to the nozzle of the injection molding machine.To this rule set we add a couple more rules, to determine whether the material selected is amorphous or crystalline. Using these six rules the expert system calculates the required clamp tonnage. Rule 5: IF mold shrink, linear flow rate* of the material = 12 THEN material = crystalline.Knowledge Representation Knowledge Representation is the second stage of the knowledge engineering process, wherein the knowledge acquired is coded into the Knowledge Base. The heuristics obtained by the knowledge engineer from the human expert are represented in the knowledge base using IF-THEN rules so that conclusions can be drawn by the expert system. These IF-THEN rules are then coded in VISUAL BASIC. A separate material database is created in Microsoft Access and then linked with the Visual Basic program. This is shown in Figure 4.Mold shrink, linear flow rate obtained from the material database.The Outputs The Optimizer gives out the most optimized values of different parameters affecting the Injection Molding process, which have to be controlled in order to ensure that a high quality part is produced in the most economical way. The outputs of The Optimizer could be confined to four different categories, namely: the Temperature, Pressure, and Time And Distance. Each of these outputs are represented in Figure 5 and discussed in detail below.Temperature: Approximately 80% of the plastic products produced today are made of thermoplastics. Thermoplastics could be defined as “ plastic materials which, when heated, undergoes physical change1.” The different types of thermoplastics are: Amorphous materials, “ which basically soften as the temperature is increased and get softer and softer as more heat is absorbed, until they degrade1.” Crystalline materials, “ these dont have a softening stage but they stay firm until they are heated to a particular point at which they start to melt and later degrade if more heat is added1. ” Figure 5. The OptimizerConsidering the differences in the properties of amorphous and crystalline materials, we set the different temperatures in the Injection Molding machine. 1. Melt temperature: The temperature to which plastic material has to be heated before it is injected into the mold. Optimizing the melt temperature results in controlling the flow rate of the material, material degradation, brittleness and flashing.2. Barrel temperatures: The different temperatures to be set at the rear, middle and the front end of the barrel of the injection-molding machine. 3. Nozzle temperature: The temperature set at the machine nozzle, which is right in front of the heating zone (barrel) of the plastic. 4. Mold temperature: The temperature at which the injection mold has to be set to obtain a plastic part of high quality with a lower cycle time. The optimized mold temperature helps in obtaining reduced cycle time and better part quality having a glossy finish, less warp and less shrinkage Pressure: There are various pressures to be optimized in the Injection-Molding machine. 1. Clamp Pressure: The amount of pressure to hold the injection mold tightly against the injection pressure. The optimal clamp pressure prevents the mold from flashing due to less clamp tonnage. It even saves energy and the mold from collapsing due to high clamp tonnage. 2. Injection Pressure: The amount of pressure required to produce the initial filling (95%) of the mold cavity. The optimized injection pressure helps to attain a part of high quality, less shrinkage, less warp, and that is easy to eject. 3. Holding Pressure: The second stage of the injection pressure, and usually fills up the remaining 5% of the mold cavity. It is usually needed to hold the plastic in the mold, from flowing back into the barrel. 4. Back Pressure: The pressure exerted by the plastic on the screw spindle. Optimized backpressure helps in obtaining a part of better density and fewer voids.Time: Cooling time: It is the amount of time required by the plastic part in the mold cavity to solidify and get ejected safely. The optimized cooling time helps in achieving better cycle time. Distance: Mold open distance: The distance for the mold halves to open apart in order to eject the part safely. Optimal mold open distance is necessary for better cycle time.Conclusion The Manufacturing Process Tool, which has been discussed in this paper, is being developed at the University of Louisiana at Lafayette. When completed this Tool will be able to give the initial optimal process parameter values, which are crucial to start off the injection molding process. These values could be later fine tuned for personal benefits by the operator using his intuition and guesswork. References 1. Douglas M.B., Fundamentals of Injection Molding: Material Selection and Product Design Fundamentals, Vol. 2, Society of Manufacturing Engineers. 2. C-MOLD Design Guide. - A Resource for Plastic Engineers. 3. Dym. J. B., Injection Molds and Molding, 2nd edition, Van Nostrand Reinhold. 4. Xinming Jin, Xuefeng Zhu, “Process Parameters Setting Using Case-Based and Fuzzy Reasoning for Injection Molding.” Proceedings of the 3rd World Congress on Intelligent Control and Automation. June 28-July 2, 2000, Hefei, P.R. China. 5. Ignizio J. P., Introduction to Expert Systems: The Development and Implementation of Rule-Based Expert Systems, Mc Graw Hill. 6. Bob Hatch, On the Road with Bob Hatch: 100 Injection Molding problems solved by IMMs Troubleshooter, Injection Molding Magazine. 7. Mok S.L, Kwong. C.K,Lau. W.S “ Review of research in the determination of process parameters for plastic injection molding.” Advances in Polymer Technology, V 18, n 3, 1999, p 225-236. 8. Shelesh-Nezhad. K, Siores, E. “Intelligent system for plastic injection molding process design.” Proceedings of 1996 3rd Asia Pacific Conference on Materials Processing, Nov 12-14 1996, Hong Kong, Hong Kong, p 458-462 9. Yeung.V.W.S., Lau.K.H., “ Injection Molding, C-MOLD CAE package, Process Parameter Design and Quality Function Deployment: A case study of intelligent materials processing.” Published in Journal of Materials Processing Technology. ARAVIND KUMBAKONAM Mr. Kumbakonam is a graduate student of the Mechanical Engineering Department at the University of Louisiana at Lafayette. He had done his B.S. form Bangalore Institute of Technology, Bangalore, India. His areas of research include Design, Solid Modeling, Artificial Intelligence and Supply Chain Management. SUREN N. DWIVEDI Dr. Dwivedi is the Endowed chair Professor of Manufacturing in the Mechanical Engineering Department at the University of Louisiana at Lafayette. His research interests include Integrated Product and Process Development (IPPD), Concurrent Engineering, Manufacturing Systems, and CAD/CAM. TERRENCE L. CHAMBERS Dr. Chambers is an Assistant Professor and the Mechanical Engineering/LEQSF Regents Professor in Mechanical Engineering at the University of Louisiana at Lafayette. His research interests include design optimization, artificial intelligence. He is a member of ASME and ASEE, and is currently serving as the Vice-President of the ASEE Gulf-Southwest Section. Prof. Chambers is a registered Professional Engineer in Texas and LouisianaBILL BESTMr. Bill Best is currently working as a plant manager at Ash Industries in Lafayette, LA. He is an expert in the field of injection molding having an experience of more than 40 years in this field.注塑模的智能制造工具机械工程部，路易斯安那大学摘要本文介绍了正在在拉法路易斯安那州大学研究的旨在建立一个电脑化的智能制造过程工具。制造过程工具是一种有助于制造商解决与注塑相关问题计算机程序。这些问题包括：开始成型的时间，非优化的周期时间和控制不良的注塑过程。制造过程中工具将有助于机器运行(不一定是专家), 优化和控制注塑成型过程。从而获得最大限度地生产效率,特别是注塑机 。注塑模是目前世界上最常用的方法用于制造复杂的商业塑料部件并具有良好的尺寸公差。依照 C-模子设计指导,被处理的所有塑料的重量的32% 通过注入成型机器,制造塑料的注入成型是最重要的制造工艺之一。最终的成型质量则主要取决于材料的类型, 模具设计及成型过程的设置。一旦材料和模具确定,零件质量基本上取决于成型过程。成型过程是一个相当复杂的,涉了及许可变的工艺参数,如压力,温度和时间设定 。为了提高零件质量,最大限度地提高注塑机生产能力，这些工艺参数必须被设定。这样一个复杂的过程要求用学历与个人的经验建立和完善。这些独立的成型工艺控制是在试验和误差的基础上, 这通常是很耗费时间的。这些控制成型过程的方法主要依靠操作者的直觉和一些经验法则，这些“经验法则”是操作者在过去一段时间,用不同的材料,压力,温度和时间设定 得到的。本文概要列出了目前正在拉法洲路易斯安那州大学进行的涉及基于注塑成型过程知识工程模块(ikem) 智能化发展的研究。ikem有不同的模式,即:句法分析,模具设计,周期时间和制造模块,这些是相互关联的。本文主要集中在制造模块 ，其中涉及开发一种智能调节工艺的过程工具,所谓的 优化器。图1和2显示了不同ikem单元的数据传递。图1：基于工程塑料注射成型的智能知识库图2：数据流程图优化器的注塑过程从这方面的专家捕捉不确定性知识, 并且以查看表小册子等形式使用确定性知识。优化器是写在VisualBasic中,将协助机器操作建立, 优化和控制注塑成型工艺,从而最大限度地生产效率,特别是注塑机。如图3所示,优化器帮助制造商设定成型机的初步优化工艺参数值。为得到最大效率，这些参数值可以由操作者用自己的直觉和猜测进行微调。基于知识设计模块一个专门的系统,或以知识为基础的系统,定义为模式和相关程序, 在一个特定的区域, 某种程度上的解决问题是相等于一个人类专家，各类专家系统是有用的,都有各自的优点和缺点。基于知识系统的基于规则这一项就是最常用 。它基本上由包含基于知识的含有“IF-THEN”的语句，称作规则。ikem该项目涉及建立一个专门系统包含一套if-then规则，这些规则是专家在知识工程师的帮助下建立的。最重要的两个问题是这一基于规则专家系统, 获取知识和知识表达的可靠性。知识的获取这是知识工程师最初从人类专家提取规则的办法。 这一步是劳力密集型,并一直被视专家系统的发展过程的瓶颈。 知识的获取,主要是靠知识工程师的技术。 其主要目的是提取策略或人类专家经验法则,并把它移交到知识库。必须慎重获取知识的过程,因为它直接影响到后面的知识表达计划。这里有知识获取的两个基本类型: 1. 知识直接从人类专家获取(不确定性知识)2. 获取知识是从先前的案件中的关系,查找表(确定性知识)。智能工具制造业正在研制捕捉从这方面经验得到的不确定性的知识,以及更多确定性知识的小册子形式的关系,查找表等。 拉菲特路易斯安那州大学一个学生知识工程师一直在工业灰努力工作,这主要是注塑厂。 知识工程师访问的灰产业方面的人类专家。然后,他组织从逻辑上提取这个数据，这种从专家提取的知识是以的启发形式或更精确的经验法则 规则。专家开发这些启发形式或经验法则和从他在这个领域的经验。这些通则作为指导方针用于得到最优化产品质量成型过程。 一组典型的例子经验法则 或启发形式的发展在计算卡式吨位是有用的: 规则1: IF部分壁厚=0.04英寸And物质=结晶THEN钳重量=2.0吨/ 英尺2* 规则2: IF部分壁厚0.04英寸And物质=结晶THEN钳吨位=2.6吨/ 英尺2规则3: IF部分壁厚=0.04英寸And物质=非晶THEN钳吨位=3. 吨/ 英尺2 规则4: IF部分壁厚0.04英寸And物质=非晶，THEN钳吨位=3.6吨/ 英尺2 英尺2代表零件在模具中垂直于注塑机的喷管的总断面面积 这个规则给我们一对另外的规则,以确定选定材料是无定形或晶体。 运用这六项规则专家系统可以计算所需钳吨位。规则5: IF模收缩,材料的线性流速*=12 THEN =物质的结晶.知识表达知识表达是知识工程过程的第二阶段, 知识以编码形式编入知识库。启发形式是由知识工程师从人类专家在知识库用if-then规则才能作出结论的专家系统获取。这些if-then规则被编在VisualBasic中。一个单独的材料数据库系统被创建在Microsoft Access中,并且与VisualBasic程序相关联。如图4所示。图4.访问 VisualBasicLinkage注塑模收缩,从材料数据库索取线性流速产出优化器给出了影响注塑成型工艺的各种参数的最优化值, 这是在生产中必须要加以控制以确保获得高质量的零件最经济方式。优化器的产出限于四种不同类别,分别是:温度,压力,时间和距离。 每个产出如图5所示并且下文有详细的讨论。 图5 优化器.温度:现今生产的大约80%的塑料制品是由热塑性塑料组成。 热塑性可以界