This study was supported by the Research Program funded by the SeoulTech (Seoul National University of Science and Technology).

1. Broker installation
NOTE: To monitor and record processing data via the Internet, a computer system that relays the data should be prepared. The system should be accessible from both the publishers and the subscribers as shown in Figure 2. Thus, it needs to have a public IP address that is known prior to any communication. An open MQTT broker called Eclipse Mosquitto is installed on the system13.
<ol>
	<li>Con...

<ol>
	<li>Shafiq, S. I., Szczerbicki, E., Sanin, C. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Proposition+of+the+methodology+for+Data+Acquisition,+Analysis+and+Visualization+in+support+of+Industry+4.0.">Proposition of the methodology for Data Acquisition, Analysis and Visualization in support of Industry 4.0.</a> Procedia Computer Science. 159, 1976-1985 (2019).</li><li>Dilda, V., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Manufacturing:+Analytics+unleashes+productivity+and+profitability.">Manufacturing: Analytics unleashes productivity and profitability.</a> McKinsey &amp; Company. , (2017).</li><li>Ismail, A., Truong, H. L., Kastner, W. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Manufacturing+process+data+analysis+pipelines:+A+requirements+analysis+and+survey.">Manufacturing process data analysis pipelines: A requirements analysis and survey.</a> Journal of Big Data. 6, 1 (2019).</li><li>Nwanya, S. C., Udofia, J. I., Ajayi, O. O. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Optimization+of+machine+downtime+in+the+plastic+manufacturing.">Optimization of machine downtime in the plastic manufacturing.</a> Cogent Engineering. 4 (1), 1335444 (2017).</li><li>Zhou, T., Song, Z., Sundmacher, K. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Big+data+creates+new+opportunities+for+materials+research:+A+review+on+methods+and+applications+of+machine+learning+for+materials+design.">Big data creates new opportunities for materials research: A review on methods and applications of machine learning for materials design.</a> Engineering. 5 (6), 1017-1026 (2019).</li><li>Rousopoulou, V., Nizamis, A., Thanasis, V., Ioannidis, D., Tzovaras, D. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Predictive+maintenance+for+injection+molding+machines+enabled+by+cognitive+analytics+for+Industry+4.0.">Predictive maintenance for injection molding machines enabled by cognitive analytics for Industry 4.0.</a> Frontiers in Artificial Intelligence. 3, 578152 (2020).</li><li>Mamo, F. T., Sikora, A., Rathfelder, C. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Legacy+to+Industry+4.0:+A+Profibus+Sniffer.">Legacy to Industry 4.0: A Profibus Sniffer.</a> Journal of Physics: Conference Series. 870, 012002 (2017).</li><li>Figueroa-Lorenzo, S., A&#241;orga, J., Arrizabalaga, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=A+role-based+access+control+model+in+Modbus+SCADA+systems.+A+centralized+model+approach.">A role-based access control model in Modbus SCADA systems. A centralized model approach.</a> Sensors. 19 (20), 4455 (2019).</li><li>Schleipen, M., Gilani, S. -. S., Bischoff, T., Pfrommer, J. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=OPC+UA+&amp;+Industrie+4.0+-+Enabling+technology+with+high+diversity+and+variability.">OPC UA &amp; Industrie 4.0 - Enabling technology with high diversity and variability.</a> Procedia CIRP. 57, 315-320 (2016).</li><li>Ni&#539;ulescu, I. -. V., Korodi, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Supervisory+control+and+data+Acquisition+approach+in+node-RED:+Application+and+discussions.">Supervisory control and data Acquisition approach in node-RED: Application and discussions.</a> IoT. 1, 76-91 (2020).</li><li>Perez-Lopez, E. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=SCADA+systems+in+the+industrial+automation.">SCADA systems in the industrial automation.</a> Tecnolog&#237;a en Marcha. 28 (4), 3-14 (2015).</li><li>Andersen, D. L., Ashbrook, C. S. A., Karlborg, N. B. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Significance+of+big+data+analytics+and+the+internet+of+things+(IoT)+aspects+in+industrial+development,+governance+and+sustainability.">Significance of big data analytics and the internet of things (IoT) aspects in industrial development, governance and sustainability.</a> International Journal of Intelligent Networks. 1, 107-111 (2020).</li><li>Kashyap, M., Sharma, V., Gupta, N. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Taking+MQTT+and+NodeMcu+to+IOT:+Communication+in+Internet+of+Things.">Taking MQTT and NodeMcu to IOT: Communication in Internet of Things.</a> Procedia Computer Science. 132, 1611-1618 (2018).</li><li>. Mythbusting the MES. Systema Available from: <a target="_blank" href="https://www.systema.com/blog/mythbusting-the-mes">https://www.systema.com/blog/mythbusting-the-mes</a> (2021)</li><li>Yeh, C. -. S., Chen, S. -. L., Li, I. -. C. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Implementation+of+MQTT+protocol+based+network+architecture+for+smart+factory.">Implementation of MQTT protocol based network architecture for smart factory.</a> Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture. 235 (13), 2132-2142 (2021).</li><li>Parian, C., Guldimann, T., Bhatia, S. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Fooling+the+master:+Exploiting+weaknesses+in+the+Modbus+protocol.">Fooling the master: Exploiting weaknesses in the Modbus protocol.</a> Procedia Computer Science. 171, 2453-2458 (2020).</li><li>Mishra, B., Kertesz, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+use+of+MQTT+in+M2M+and+IoT+systems:+A+survey.">The use of MQTT in M2M and IoT systems: A survey.</a> IEEE Access. 8, 201071-201086 (2020).</li><li>. pyodbc 4.0.34 Available from: <a target="_blank" href="https://pypi.org/project/pyodbc/">https://pypi.org/project/pyodbc/</a> (2021)</li><li>Ayer, V., Miguez, S., Toby, B. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Why+scientists+should+learn+to+program+in+Python.">Why scientists should learn to program in Python.</a> Powder Diffraction. 29 (2), 48-64 (2014).</li><li>Boyes, H., Hallaq, B., Cunningham, J., Watson, T. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+industrial+internet+of+things+(IIoT):+An+analysis+framework.">The industrial internet of things (IIoT): An analysis framework.</a> Computers in Industry. 101, 1-12 (2018).</li></ol>

The authors declare no conflicts of interest.

By following the protocol presented, the processing data can be monitored and recorded without expensive IT solutions such as the MES. The IoT technologies can make it easier to acquire and deliver data from conventional machines. It has been shown that the message-based protocol, MQTT, successfully serves as a platform for data communication for polymer processing lines. Moreover, additional data can be flexibly measured and transmitted together. The additional publishers employed in this work were the Raspberry Pis. No...

In the wave of the 4th industrial revolution, the acquisition and monitoring of various processing data have become important tasks1. In particular, improving the manufacturing process using process data and establishing efficient operation plans will be an important goal of all manufacturing facilities2,3. Downtime can be greatly reduced if an alarm can be sent out of the factory or if predictive maintenance can be performed in time4. Recently, many efforts have been made for data analyses in polymer processes5,6. However, it is not easy to conduct these tasks due to the difficulties in acquiring such data from the existing systems7. The hierarchical structure of the control and instrumentation makes the data acquisition and communication difficult.
First of all, it is not possible to obtain data from different machines with different manufacturing dates. It is difficult to realize communication between different machines since this requires interoperability between different fieldbuses in proprietary formats. In this way, communication methods and data formats are kept private. This helps one easily maintain data security but keeps users dependent on the machine builder for the services and future developments. The recent control computers including human-machine interface (HMI) attached to polymer processing machines are mostly Windows-based these days but are loaded with software created in a proprietary development environment. It is possible to use programmable logic controllers (PLCs) from different companies to communicate with the sensors or actuators, but in many cases, the upper supervisory control and data acquisition (SCADA) system is dependent on the control computers8. This practice has caused numerous protocols, fieldbuses, and control systems to compete in the market. Although this complexity has been alleviated little by little over time, many types of fieldbuses and protocols are still in active use.
On the other hand, communication between control devices and SCADA has been standardized by the Open Platform Communications United Architecture (OPCUA)9. Moreover, communication between SCADA and the Manufacturing Execution System (MES) has also been mainly made through OPCUA. In such a tight hierarchical structure, it is not easy to freely extract data for process monitoring and analysis. Usually, data has to be extracted out of the SCADA or MES10. As mentioned earlier, these systems are vendor-specific, and the data formats are rarely open. As a result, the data extraction requires substantial support from the original information technology/operational technology (IT/OT) solution vendors. This can hinder data acquisition for monitoring and analysis.
In a film extrusion line, the control PC is supervised by a SCADA system11. The SCADA system is operated by a computer program that cannot be easily modified. The computer program might be editable, but the editing is quite expensive and time-consuming. To easily monitor and analyze the processing data, the data should be accessible from any location. To monitor the processing data away from the site, the computer program should be capable of streaming the processing data to the Internet12. Moreover, an open free method reduces the expenses for the data acquisition13. This approach allows data analysis to be performed even in small factories that cannot afford to invest in commercial IT solutions14.
In this study, a message protocol based on the publisher-subscriber model is employed. The Message Queuing Telemetry Transport (MQTT) is an open and standard protocol that enables messaging between multiple data providers and consumers15. Here, we propose a system that acquires, transmits, and monitors data using MQTT for existing manufacturing facilities. The system is tested in a film extrusion line to verify the performance. The data from the original controller are transmitted to an edge device via the Modbus protocol. Then, the data is published to the broker. In the meantime, two Raspberry Pis publish the measured temperatures and illuminance to the same broker. Then, any device on the Internet can subscribe to the data, followed by monitoring and recording it as shown in Figure 1. The protocol in this work shows how the whole procedure can be done.

<table border="1" fo:keep-together.within-page="1" fo:keep-with-next.within-page="always">
	<tbody>
		<tr>
			<td>Edge Device</td>
			<td>Adavantech</td>
			<td>UNO 420</td>
			<td>Intel Atom E3815 Fanless</td>
		</tr>
		<tr>
			<td>Film Extrusion Machine</td>
			<td>EM Korea</td>
			<td>Not Available</td>
			<td>For production of 450 mm film</td>
		</tr>
		<tr>
			<td>Pydroid</td>
			<td>IIEC</td>
			<td>Not Available</td>
			<td>Android Devices</td>
		</tr>
		<tr>
			<td>Python3</td>
			<td>Python Software Foundataion</td>
			<td>Not Available</td>
			<td>Windows, Linux</td>
		</tr>
		<tr>
			<td>Raspberry Pi 4</td>
			<td>CanaKit</td>
			<td>Not Available</td>
			<td>Standard Kit</td>
		</tr>
	</tbody>
</table>

data communication based on mqtt in a polymer extrusion process

This work aims at building a flexible data communication structure for a polymer processing machine by employing a publisher-subscriber based protocol called Message Queuing Telemetry Transport (MQTT), which is operated over TCP/IP. Even when using conventional equipment, processing data can be measured and recorded by various devices anywhere through an Internet communication. A message-based protocol allows flexible communication that overcomes the shortcomings of the existing server-client protocol. Multiple devices can subscribe to the processing data published by source devices. The proposed method facilitates data communication between multiple publishers and subscribers. This work has implemented a system that publishes data from the equipment and additional sensors to a message broker. The subscribers can monitor and store the process data relayed by the broker. The system has been deployed and run for a film extrusion line to demonstrate the effectiveness.

It has been found that the data shown in the HMI and measured by the Raspberry Pis were monitored and recorded in the subscribers as shown in Figure 5. As presented in the video, the processing data are logged into the database.
<img alt="Figure 1" class="xfigimg" src="/files/ftp_upload/63717/63717fig01.jpg" /> 
Figure 1: Outline of the data transmission using the MQTT protocol.<...

Watch this Scientific Journal Video about Data Communication Based on MQTT in a Polymer Extrusion Process at JoVE.com

Data Communication Based on MQTT in a Polymer Extrusion Process

This work proposes a flexible method for data communication between a film extrusion system and monitoring devices based on a message protocol called Message Queuing Telemetry Transport (MQTT).

This work aims at building a flexible data communication structure for a polymer processing machine by employing a publisher-subscriber based protocol ...

The protocol in this work allows building a flexible data communication structure for a Polymer processing machine by employing a publisher subscriber based protocol code MQTT. Even when using a classical equipment processing data can be observed and recorded by various devices anywhere in the internet. The protocol facilitates data communication between multiple publishers and multiple subscribers.We have implemented a system that publishes processing data from a existing extrusion line to the subscribing devices through a broker device. The data from a legacy extrusion controller has been interpreted in an device and sent up to the broker. Addition devices for ambient temperatures at two different locations and of during processing also publish the data to the broker.Then the broker relays the data to subscriber devices which are interested in those data. To display and record the data. The subscriber system has been designed and built.Since all the code for the participating devices have been written in Python. The code can be reused on the devices with different operating systems. Finally, the system is deployed and tested for line to demonstrate the validity First broker installation.To begin configure the MQTT broker system so that it may monitor and record processing data over the internet. In order for a broker device to relay such data it must be accessible by both publishers and subscribers. Connect a computer system to the internet with a public IP address so that the broker can be accessed by both publishers and subscribers.Then on the computer install, open broker software such as Eclipse Mosquito use a test tool like MQTT lens to examine the MQTT brokers operability. Second, main publisher preparation. Now let us prepare the main publisher device.As previously mentioned this computer publishes the machine data via MQTT to the broker. Legacy data should be interpreted and repackaged to be sent out. This can be usually done by RS-485 or ethernet.The connection in the hardware level should be verified depending on the bus type. The considered extrusion machine sends out data via Modbus through an ethernet port, to acquire data from the extrusion machine and publish them. Place a computer as the main publisher in the machine site.On that computer install Python three as the software environment. Then install Pi Modbus to enable Modbus connectivity and communication. Examine the Modbus function codes of the extrusion controller and connect it to the main publisher.Fully identify the data and the associated address and the Modbus codes from the machine using a Modbus tool such as Modbus poll or Q mod master on the main publisher. Then write a Python code on the publisher that retrieves the data from the extrusion controller. In addition, combine data stream via PCIE USB RS-232 and RS-485 from other devices.Import paho dot MQTT dot client and implement the code to connect and publish data to the broker. Third, additional publisher preparation. In addition to the main publisher let us employ additional IOT devices to acquire and publish ambient temperatures and film illuminance.To do so. Two Raspberry Pi devices have been employed. Each device publishes the measured data to the broker as the main publisher does.Place the devices near the sensor locations. Then install Python three on the devices and implement the code to acquire the sensor data. The sensor data are transmitted by I2C for the illuminance sensor and by GPIO for the temperatures.Reuse the previous Python code to publish the data. Thanks to the device independence of Python. The code for windows can be reused for the Raspberry Pis.Fourth, subscribers set up. Now let us explain how to subscribe to the processing data. As previously stated, any devices on the internet may receive the data via the broker.Again, the data can be processed and visualized by a Python code. On an internet connected device install a suitable Python environment depending on the device and the operating system. For example, on an Android device Pydroid 3 should be installed instead of Python 3, then import both paho dot MQTT dot client and paho dot MQTT dot subscribe to connect to and receive data from the broker.Then build a user interface as required based on Pi QT5. Since implementation of this part is quite lengthy and can be laborious. The details are not further described here with this code display the incoming data on the GUI.Also know that existing applications such as MQT tool in app store can receive the data. Fifth, data logging. To record the data during monitoring the Python should be able to access a database.In this work the data is written to a Microsoft access file. Considering the data scale and software availability. Select a subscriber device to record the data.Then import pyodbc in the Python code to access the database. For recording the processing data, send a query to the database by the Python code. To retrieve the recorded data send another query to the database.Once the data is retrieved, the data can be analyzed by restructuring the data as needed. A table readable in a spreadsheet can be instantly built. Sixth, deployment.After all the items are developed the Python codes should be deployed onto each device. The mode of connection wired or wireless is not important but it has to be secured that each device should be able to access the broker. This means the broker plays as a gateway on the border between the internet and internet for the security purpose.Here to maintain better security connect the extrusion controller the main publisher and the additional publishers to the internet. Then connect one ethernet port of the broker to the internet and the other to the internet to monitor and record the processing data. Connect subscribers to the internet as required.Seventh, execution. To test the whole system, start the extrusion line. Next turn on all the computer devices.Then start the broker software Mosquito, and run all the Python codes. Representative results. As a result of executing the designed and implemented system, processing data are monitored and visualized.The data can be shown in a variety of ways including graphs and tables, even on a mobile device the data can be monitored and recorded. Moreover, the data is recorded and retrieved in order to be analyzed. Conclusion It has been found that the data shown in the HMI and measured by the Raspberry PIs are monitored and recorded in the subscribers.By following the protocol presented the processing data can be monitored and recorded without expensive IT solutions such as the MES.

data-communication-based-on-mqtt-in-a-polymer-extrusion-process

Research

JoVE Journal

Engineering

3.2K Views.  Seoul National University of Science and Technology (SeoulTech). The protocol in this work allows building a flexible data communication structure for a Polymer processing machine by employing a publisher subscriber based protocol code MQTT. Even when using a classical equipment processing data can be observed and recorded by various devices anywhere in the internet. The protocol facilitates data communication between multiple publishers and multiple subscribers.We have implemented a system that publishes processing data from a existing extrusion li...