Injection Molding Process Control, Monitoring, and Optimization

Human history has been defined in terms of materials categories: the Stone Age, the Bronze Age, and the Iron Age. It is well accepted that we are now living in a polymer age. Since the 20th century, polymer materials, including plastics, fibers, elastomers, and proteins, have gradually appeared in almost every area of people’s everyday life, and there are a variety of applications in agriculture, industry, and even the defense industry. In all of the polymer materials, plastic is a major class.

Control of the injection molding process not only involves sequence manipulation, but also some key process variable regulation. The following discussion will mainly focus on the process variable control. Early injection molding controllers were mainly constructed using simple electrical components such as timing relays and switches. The control is thus naturally open-loop. As the molding process becomes more and more complex, manufacturers require greater accuracy and tighter tolerances. All of the variations and disturbances during the molding process must be properly dealt with, so closed-loop control becomes necessary, which also triggers the application of computer-based controllers to the injection molding machines. It is therefore essential to give a brief introduction to some basic control concepts.

In the injection molding process, raw plastic materials are fed to the machine through the hopper with a certain volume each cycle, are melted and transferred to the nozzle of the barrel, and then go through the injection, packing, and cooling stages to form the final product of limited amount. The whole cycle normally lasts for several seconds to minutes. This kind of production fits the definition of the batch process, as quoted below [33]:

Injection molding is a typical batch process; it has its own characteristics in comparison to a continuous process. The obvious differences between a continuous process and a batch process like injection molding are (1) a batch process has a finite duration, (2) a batch process repeats itself until the specified amount of product has been made, and (3) a batch process is processed by an ordered set of activities. These characteristics make the control schemes proposed for a continuous process ill-suited for injection molding. Modifications of the original control algorithms have to be made to cope with these features. To summarize the difference between injection molding and a traditional continuous process, the distinctive nature of an injection molding process has three aspects:

Multiplicity of the operation phase is an inherent nature of many batch processes; each phase has its own underlying characteristics, and process can exhibit significantly different behaviors over different operation phases. For example, an injection molding process can be roughly divided into three operation phases, injection phase, packing phase, and cooling phase; a fermentation process can be divided into lag phase, exponential growth phase, and stationary phase. Apparently, each operation phase has different process variable trajectories, running modes, and correlation structures. So, statistical modeling and online monitoring of multiphase batch processes should include not only the overall operation performance, but the specific phase characteristics.

The ultimate concern of batch process monitoring is to monitor the variation of end-product quality and find solutions to improve quality consistency. It is, however, difficult to monitor product quality for batch processes in an online way. The main reasons lie in that (1) end-product quality attributes are only available after a batch operation is finished, and (2) most quality attributes are difficult to measure instantly after a product is produced. Alternatively, SPM techniques are used to monitor product quality indirectly, inferring end-product quality from process behaviors, which have been introduced in Chapters 5 and 6.

The measurement or monitoring of polymer melt status inside an injection mold is important to the online quality control of an injection molding process. Many efforts have been made in this area. Most measurements of the material status inside an injection mold have been limited to that of cavity pressure [130], mold surface temperature [131], mold temperature [131], and mold heat flux [131]. Limited progress has been made with respect to the monitoring of the melt-front position. Yokoi et al. [132–135] might be the first who succeeded in observing the melt flow in a mold by visualization through glass inserts fitted on the mold wall. This technique was modified by Bress and Dowling [136] for the observation of the large part surface side. This kind of visual technique using glass inserts allows for direct observation of the melt flow status, typically using a high-speed camera. This kind of device provides good information for research, but it has limited usage for industrial applications because glass cannot withstand the high pressure required in the applications. Furthermore, the flow and heat transfer along the glass surfaces may be different from those along the mold metal surfaces.

The quality of an injection molded part depends on the characteristics of the material, the performance of the molding machine, the mold design, and processing conditions. For a given material, mold, and machine, the part quality is mainly determined by its processing conditions, among which injection velocity during the filling stage plays a particularly important role. Studies [161–170] have shown that injection velocity can affect the degree of orientation and residual stress, as well as surface properties of the molded parts. In view of its importance, advanced control techniques [172–174] have been developed for the control of injection velocity. With such advanced controllers, the injection velocity can accurately follow any reasonable profile. But for a given mold and material, how should the injection velocity be profiled to produce a good-quality part? A method for profiling the injection velocity based on a constant melt-front-velocity strategy will be presented in this chapter.

After a mold cavity is filled, the injection molding process switches from the filling to the packing-holding stage. Proper profile setting of packing pressure is essential to part quality. A complete packing profile is determined by three key elements: start point of the packing, end point of the packing, and packing pressure levels in between. Several methods have been proposed and developed for detection of the start point of the packing, which is also the end point of the filling stage. For instance, the fuzzy V/P transfer technique [225], which is a technique based on the online detection of the abrupt pressure change at the end of filling through the fuzzy method. The capacitive transducer, as described in Chapter 8, is also capable of detecting this transfer point. In this chapter, the packing profiling work focuses on the other two elements.

Besides the filling and packing stages, plastication is another stage that plays an important role in final part quality. But unlike the other two stages, which focus on profiling of the injection velocity or the packing pressure only, more than one variable has to be considered. The key variable includes barrel and nozzle temperatures, back pressure, and screw rotation speed. A thorough study on parameter setting for the plastication stage is described in this chapter.