A Petri-net based distributed monitoring system using PIC microcontrollers

https://doi.org/10.1016/j.micpro.2004.08.003Get rights and content

Abstract

This paper describes the development of a distributed system for process and condition monitoring. The developed system is based on the Petri-net concept, and uses PIC microcontroller technologies in its implementation. The resulting distributed system has data communication features such as Fieldbus and Internet connectivity thus enabling the deployment of a flexible and low-cost structure for monitoring purposes. The proposed system architecture is discussed in detail, considering both the hardware and software elements involved. The implementation of the Petri-net approach in a microcontroller environment is also presented. Finally, the benefits and limitations of such a system are considered, using some initial results obtained in an experimental implementation.

Introduction

Microcontrollers are being developed with increasing processing power and flexibility. This has encouraged their use in areas where computer based systems previously represented the best available choice [1], [2], [3], [4]. One such area is process and condition monitoring (PCM). This work outlines developments using microcontrollers that will offer low-cost alternatives to previously deployed PCM systems. It is hoped that the availability and functionality of these systems will stimulate their deployment in areas that have previously been reluctant to consider their use due to the high costs involved.

PCM will be increasingly used to meet demands for lower cost and more effective manufacturing [5]. The complexity represented by modern manufacturing systems calls out for the use of online and real-time monitoring technologies to enable critical parts of the process to be assessed and to ensure that adequate actions can be taken in a planned way [6], [7]. The use of Internet based (remote) monitoring strategies is also becoming a key factor when considering global manufacturing [8]. This relatively new scenario has led machine manufacturers to review the tasks related to the provision of support to their customers, forcing them to provide new machine developments with features that include, for example, the ability to undertake remote fault analysis. However, cost is still a factor that has prevented the use of PCM technology on a larger scale [9]. It is proposed here that Microcontrollers, supported by technologies such as industrial networks, can provide online and real-time capabilities at a much lower cost.

Many processes may be described as a logical sequence of events. This has led the authors, among others, to the development and use of Petri-nets as a tool for PCM [10], [11], [12], [13]. Details of the work previously undertaken by the authors can also be accessed via the IPMM research group web page [14].

The new approach presented in this work is centred upon the development and implementation of a Petri-net based monitoring structure employing PIC microcontrollers rather than the previously utilised PCs. The paper considers, in Section 2, the configuration of the PIC based system and the monitoring and connectivity aspects developed to support this new implementation. This section also introduces the concepts of the Petri-net based approach to process monitoring. An outline of the way in which the elements of the Petri-nets are utilised is described in Section 3. This section also considers the operation of the monitoring system using a CNC machining centre toolchanger as a Case Study. The paper finally considers, in Section 4, the advantages associated with these developments and the direction that future research may take.

The aim is to offer a low-cost alternative for process and machine monitoring thus making the technology more accessible. Additional benefits arise since these developments are based upon microcontroller technology that does not demand a large number of peripheral components and can be installed and operated close to the process.

Section snippets

System configuration

The primary requirements of the system are to follow the process events and store them in a database for further analysis. To achieve this the Monitoring System was developed based upon the structure shown in Fig. 1.

The Monitoring Module captures the process signals and operates the Petri-net representing the process, while recording the relevant events. The Connectivity Module provides an Internet link that is necessary to enable process events to be delivered to a remote location. The

Petri-net monitoring system operation

This section outlines the capabilities of the Petri-net based system by considering the role played by Petri-net transitions and places in facilitating the deployment of a powerful and flexible monitoring tool. The operation of the system is then presented using a machining centre toolchanger as a case study.

Conclusions and further work

The implementation of a distributed monitoring system has been described. It represents a low-cost alternative for PCM, employing a PIC microcontroller to carry out monitoring functions. Flexibility is assured by the employment of Fieldbus technology, which provides a way of deploying Monitoring Modules alongside or within the process, depending upon the circumstances. Internet connectivity enables online data storage, allowing the use of high-level tools for data analysis. Since this process

Acknowledgements

M. R. Frankowiak is a PhD student supported by CAPES, a Brazilian Federal Agency for Post-graduation Education.

References (21)

There are more references available in the full text version of this article.

Cited by (17)

  • Microcontroller-based process monitoring and management using embedded Petri-nets

    2012, Robotics and Computer-Integrated Manufacturing
    Citation Excerpt :

    To further facilitate this approach an output transition was defined, allowing the initiation of an hardware output which was capable of triggering events external to the Petri-net. The hardware architecture supporting this implementation has been previously described [1,23]. The Petri-net structure was implemented using a microcontroller that was capable of acquiring and analysing up to twenty four separate digital signals, four analogue signals and two pulse inputs.

  • Artificial neural network implementation in single low-cost chip for the detection of insecticides by modeling of screen-printed enzymatic sensors response

    2010, Computers and Electronics in Agriculture
    Citation Excerpt :

    The activation function was computed in a Lock-up Table (LUT). A distributed system for process and condition monitoring based on PIC microcontroller was developed by Frankowiak et al. (2005). The system was based on Petri-net concept.

View all citing articles on Scopus
View full text