IoT-enabled CAMP Framework

Get Complete Project Material File(s) Now! »

Chapter 3 IoT-enabled CAMP Framework

This chapter introduces the structure of proposed Cloud-based AM Platform (CAMP). The motiva-tion behind the system design is explained, followed by an introduction of the enabling technologies including IoT and CPS. After this, an overview of the proposed system is presented to introduce the platform’s stakeholders. The general work flow in product development processes is then summar-ized to identify the key functionalities required in the platform. Details of the architecture design and the product development process in the platform are also provided. Last but not least, the cyber-physical 3D printer is introduced as a foundation to realize the hardware access and sharing in the cloud. To realize a real-time, standard and secure communication, the information model for AM is developed, and the possibility of self-diagnosis is examined.


The key objective of this project is to develop a cloud-based environment for AM that supports rapid product development. From the literature review provided in Chapter 2, it was identified that cur-rent CM systems for AM are following a similar pattern to traditional manufacturing. Considering the current situation and unique characteristics of AM, a novel cloud-based AM platform is pro-posed which takes advantage of emerging IoT technologies. Based on this, cloud-based CPS can be developed to enable the sharing of distributed 3D printing resources. Together with knowledge and expertise, AM services in the cloud can cover all of the value-creation processes of a product life cycle, from design, process planning to fabrication and monitoring.

Enabling Technologies

As one of the most important enablers, the IoT plays a central role in improving the CAMP’s smart-ness. The concept of IoT can be traced back to 1995 [173]. With the rapid development of Radio-Frequency Identification (RFID), sensor devices, wireless networks, etc., the IoT has drawn much attention from both the industry and academia in recent years. According to the ITU standard [174], the IoT can be defined as a global infrastructure for the information society, enabling advanced ser-vices by interconnecting (physical and virtual) things based on existing and evolving interoperable information and communication technologies. More specifically, through the exploitation of identific-ation, data capture, processing and communication capabilities, the IoT makes full use of things to offer services for all kinds of applications, whilst ensuring that security and privacy requirements are fulfilled. The types of devices and their relationships with physical things are presented in Fig-ure 3.1. A physical thing can either directly connect to the networks, or indirectly connect through a data-carrying device. Using this kind of structure, the 3D printers can be projected into a virtual environment and better communication between printers and the cloud can be achieved.
CPS can be used to model physical things in computers. Lee defined the CPS as integrations of computation and physical processes [175]. Embedded computers and networks monitor and control the physical processes, usually with feedback loops where physical processes affect computations and vice versa. As an emerging subject, the CPS for 3D printers have shown some appealing func-tionalities. DebRoy, Zhang, Turner et al. provided an overview of current activities on cyber-physical 3D printers [176]. Wang, Kwok & Zhou proposed an in-situ droplet inspection and control system for the material jetting process [177]. Kim, Espalin, Cuaron et al. presented a method to detect material deposition status in the material extrusion process by measuring the motor current of the filament feed pump [178]. Eiriksson, Pedersen, Frisvad et al. developed an Augmented Reality (AR) interface to realize more intuitive control and monitoring [179]. AR has also been applied to detect geometric failures in material extrusion [180]–[182]. An in-situ monitoring of strain and temperature distribu-tions for material extrusion process has been proposed [183]. A CPS for process-level modelling and simulation was developed for the binder jetting process [184]. The information extracted from the physical twins can be used for monitoring and detecting failures. Briefly, they can be divided into two types: direct detection and indirect detection. Direct detection means sensors can extract real-time information directly from the part in print, e.g. the images used in AR-based geometrical measure-ment. Conversely, indirect detection means sensors extract the information from other sources rather than the printing part, e.g. measuring motor current for material deposition status detection [178].
Enabled by IoT, the CPS can be extended to the cloud via the internet [185].With these advanced technologies, various 3D printers can be connected to the cloud and communicate with other ma-chines and humans via the internet. Complex simulations and other computations can be efficiently conducted using distributed and scalable computing resources in the cloud.

System Overview

The CAMP is an environment that merges physical 3D printing resources with cyber world, and, consequently, provides 3D printing services by employing the underlying physical resources. There are three major stakeholders in the cloud – the service provider, the service requester and the plat-form provider (Figure 3.2). Service providers connect their printers and other printing resources to the cloud platform, and upload resource capabilities and know-how. Service requesters explore the options they have and find most suitable solutions for their tasks and make orders. The platform provider provides the cloud computing and storage facilities and construct the platform, where ser-vice providers can publish their resources and service requesters can find and use these resources.
Enabled by IoT, various printers are able to be directly connected to the cloud so remote control and monitoring can be achieved. Smart functionalities can also be developed based on this. In this en-vironment, a printing service is defined at a lower level as a specific configuration of a printer, along with its control and monitoring capabilities. This new definition of AM as a Service (AMaaS) enables customers to customize not only design models, but also printing processes in detail to achieve de-sired results. Knowledge shared in the cloud helps customers make reasonable designs and choose appropriate printing configurations/services for their requirements.

 roduct Development with AM

The aim of this platform is to support not only prototyping, but also end-use product development. For prototyping, the challenge is simply to find the optimal process configuration and print the de-signed model. However, For end-use product development, both design and process related factors need to be considered simultaneously to find the optimal global solution. This is a typical DfAM process.

DfAM Procedure

To help customers develop their products using AM, it is critical to understand the general DfAM procedures. Similar to conventional DFM, the procedure of DfAM is described as in Figure 2.4.

AM Information Flow

Understanding the information flow in AM processes helps us to identify the key inputs to a 3D printer and how to assist customers define these inputs. Communication between customers and suppliers in the cloud can then be improved. ISO17296-4 standard describes the basic data flow from a product idea to the actual part [186]. Accordingly, the general information flow as a basis for platform architecture design is defined in Figure 3.3.
Users’ inputs in the product development process are mainly in the process planning stage and design stage. Firstly, they need to come up with the design model based on their requirements. At the design stage, capabilities of AM processes need to be taken into account so that the design model is printable and able to maximize the utilization of AM resources. Secondly, they need to define the printing process’s key factors. At the process planning stage, users need to consider preferred materials, the AM technologies and printers available, and the parameters that are able to achieve the desired properties. Both the design and process planning stages require sufficient knowledge of AM to make informed decisions. After that, the sliced contour data will be generated and the printer will start printing. It is also important to allow customers to obtain the real-time status of the printer at the printing stage. If something unexpected happens, they can identify and fix it as soon as possible.

Architecture of the CAMP

In a typical CM platform, there are four layers: physical resource layer, virtual resource layer, ser-vice layer and application layer [31]. In this system, a similar architecture is adopted as shown in Figure 3.4.

Physical Resource Layer

There are two types of physical resources: hard resources and soft resources [38]. A classification of AM resources based on the work of [33] is presented in Table 3.1. With the application of sensors and networks, physical resources such as 3D printers can be connected to the cloud using IoT tech-nologies. In this way, real-time monitoring and control are able to be realized by customers. With real-time data, AI technologies can be applied to achieve self-diagnosis in the printing process. Much of the soft resources need to be transformed to appropriate digital formats so the platform could process and manipulate them based on the customers’ requirements.

1 Introduction 
1.1 The NZPA Programme
1.2 Setting the Scene and Identifying the Challenges
1.3 Objectives and Scope
1.4 Thesis Synopsis
2 Literature Review 
2.1 Additive Manufacturing
2.2 Distributed AM Services
2.3 Design for AM
2.4 AM Process Selection
2.5 Literature Summary
2.6 Research Gap and Motivation
2.7 Research Methodology
3 IoT-enabled CAMP Framework
3.1 Introduction
3.2 Enabling Technologies
3.3 System Overview
3.4 Product Development with AM
3.5 Architecture of the CAMP
3.6 Product Development Process in CAMP
3.7 IoT-enabled Cyber-Physical 3D Printer
3.8 Framework Summary
4 Knowledge Management System
4.1 Introduction
4.2 Bayesian Networks
4.3 Methodology
4.4 Confirmation Test
4.5 Model Development
4.6 Illustrative Example
4.7 Knowledge Management System Summary
5 Decision Support for Service Selection 
5.1 Introduction
5.2 Hybrid MCDM Framework
5.3 Option Navigation
5.4 Ranking
5.5 Illustrative Example and Comparison
5.6 Service Selection System Summary
6 Implementation and Case Study 
6.1 Architecture of Prototype System
6.2 Setup of Cyber-Physical 3D Printer
6.3 Service Selection System Development
6.4 Case Study
6.5 Implementation and Case Study Summary
7 Conclusions and Future Work 
7.1 Recap of the Research
7.2 Scientific Contributions
7.3 Limitations
7.4 Recommendations for Future Work

A Cloud-based Additive Manufacturing Platform to Support Rapid Product Development

Related Posts