2018 International Conference on Intelligent Systems (IS)
Advanced CPS Service Oriented Architecture for Smart Injection Molding and Molds 4.0 Hector R. Siller University of North Texas Denton, Texas, United States of America
[email protected]
David Romero Tecnológico de Monterrey Monterrey, Mexico
[email protected]
Ricardo J. Rabelo Federal University of Santa Catarina Florianopolis, Brazil
[email protected]
Elisa Vazquez Tecnológico de Monterrey Monterrey, Mexico
[email protected] Abstract—This paper explores current advances towards Smart Injection Molding, and introduces the concept of Smart Molds or Molds 4.0. It presents a case study of the manufacture of a medical device, including the experimental set-up of an instrumented mold, as well as general recommendations on how to implement a Smart Manufacturing vision in the plastic Industry 4.0. Furthermore, it includes a proposal of an Advanced Cyber-Physical System (CPS) Service Oriented Architecture (SOA) for real-time monitoring and data analytics of a smart microinjection molding process and for smart molds instrumentation as a way to realize such smart vision. Keywords—Smart Injection Molding, Smart Molds, Molds 4.0, Cloud Manufacturing, Cyber-Physical System, Industrial Internet of Things, Service Oriented Architecture, Industry 4.0.
I. INTRODUCTION Recent advances in real-time monitoring and control of manufacturing processes, thanks to sensor-data, advanced data analytics and computer processing power, have underpinned the realization of Smart Manufacturing and Industry 4.0 visions [1] [2]. This provides more reliable ways to support related issues like “digital quality management” [3] and “zero-defect manufacturing” [4]. Nevertheless, limited attention has been put into the development of Smart Injection Molding Processes in spite of its high manufacturing complexity and its significant contribution to the industrial economy in the fabrication of value-added products [5]. At the shop-floor level, according to Kenig et al. [6], the desired elements in an intelligent control of injection molding are: (a) a reliable, real-time, measurement of the process parameters, (b) a process model that describes the relationship between the process parameters and the part properties, and (c) teaching capabilities of the control system to ensure that it can identify deviations from process limits, and their effects on the quality of the part. Hence, Smart Injection Molding can be defined as a sensing adaptive-controlled injection molding process for producing parts by injecting molten material into a mold cavity in a real-time monitored and controlled production environment, where the molded part will cool-down, solidify and hardness into the shape that has conformed to the contour of the mold cavity. Such control variables may include screw speed, mold pressure, injection temperature, and take-outtime. Moreover, integrated product-process quality control in injection molding towards zero-defects requires more than a
978-1-5386-7097-2/18/$31.00 ©2018 IEEE
smart manufacturing process. It requires a Smart Mold, which can be described as a sensorised mold with self- and processmonitoring capabilities supporting the digital quality management of molded parts and its own predictive and preventive maintenance (i.e. asset care) based on the total control of pressure, temperature, vibrations, cycles, integrity, hot runner and cooling parameters of the intelligent object. Other requirements for Smart Molds and Smart Injection Molding include: (a) faster processing of the molded part, (b) finished parts out of the mold requiring no further work, (c) faster changeover of molds to accommodate a greater variety of molded parts, (d) automated materials handling and processing capability, and (e) data gathering and storage of quality records for each part produced from a specific mold [7]. Regarding the essentials of Industry 4.0 vision, Molds 4.0 represent an environment where injection-molding machines are seen as autonomous equipment that can collaborate with other manufacturing equipment via computer networks and with the enterprise information systems (e.g. MES) for reliable, faster and smarter decision-making. Such Smart Equipment is regarded as a ‘service-provider actor’ or ‘sensing-smart agent’ with the capability to seamlessly interoperate with other and required manufacturing resources to more efficiently establish logical teams to better cope with the requirements of the manufacturing process value chain. All the required information and control flows are integrated and coordinated by a wider and multi-layered systems architecture based on the principles of Service Oriented Architecture (SOA) [8] and the benefits provided by cloud-computing and computing pervasiveness. All this enables data networking with other manufacturing information systems in order to support higher productivity and quality performance of the whole production system. A Smart Injection Molding Machine and a Smart Mold are then seen as enablers for a “Mold Industry 4.0”. Lastly, the experimental set-up refers to the development of an Advanced Cyber-Physical System (CPS) SOA Architecture for real-time monitoring and data analytics of a smart injection molding process, and the instrumentation of its corresponding “Smart Mold” with piezoelectric sensors for data acquisition, both set-ups aiming to draw general recommendations in terms of sensors technology, data acquisition system, and signals conditioning.
II. TOWARDS SMART MOLDS OR MOLDS 4.0 Piezoelectric technologies have been introduced in several applications for acceleration, force and pressure sensing [9]. Furthermore, traditional devices used are mainly accelerometers, dynamometers and acoustic emission detectors, which have been used in thousands of research works in machine tools condition monitoring and other automation applications [9]. However, it is recognized that the recent introduction of temperature and pressure “piezoelectric sensors” for conditionbased monitoring could lead to fostering a Smart Injection Molding Process. In this particular area, it is important to test first the reliability of signals by performing additional validations with the aid of “Finite Element Analysis”. The work presented in this section is intended to exemplify the engineering behind the construction of sensorised molds as the first step of the construction of a test-bed for Molds 4.0. The selected case study was a mold adapted to hold four cavities and fabricated for tooling microinjection-molding machines [10]. The plastic part to be injected is a locking ligation system, which is often used in surgical procedures for the treatment of several medical conditions and injuries. Fig. 1 shows the engineering design of the cavities, including runners, gates and the digital representation of the part to be injected. The characteristics of the Mold 4.0 can be seen in Fig. 2, in which the general geometric and functional features are the following: four cavities, sliders for enabling the injection in hidden areas, runners, ejection pins, gates and plates [10].
After the selection of critical variables, the next step is the installation of piezoelectric sensors inside the cavity. The selection of suitable sensors depends on the complexity of geometric elements of the part and on their adaptation to the small scales of these geometries. The characteristics of piezoelectric sensors selected for this application are showed in Table I. TABLE I. CHARACTERISTICS OF PIEZOELECTRIC SENSORS Range Overload Sensitivity Linearity, All Range Thermocouple Type K Operating Temp. Range (general) Temp. at the Cavity
Bar Bar pC/bar
0 – 2000 2500 -4,891
%FSO
