SSV Software Systems

Klaus-Dieter Walter | Meinrad Happacher,

The digital twin in practical use

A digital twin is not only a virtual instance of a physical object, but also a universally usable IT application for adapting the requirements of a wireless modem retrofit application, as in the practical example described.

© MangKangMangMee/stock.adobe.com

The following task is assumed in the practical example described: In a machine control system, there is a slot for retrofittable expansion cards to connect the respective machine to the ISDN telephone or GSM mobile network if required. This connection is intended to enable certain remote maintenance tasks. This also includes remote level monitoring of various operating resources. Alternatively, an individual telephone number can be called automatically to establish an ISDN or CSD data connection and transmit the current fill levels and other operating data using a configurable character string. However, the ISDN and CSD services required for operation have been or are now being discontinued by practically all providers. An alternative is therefore required to retrofit existing installations and equip new machines with an IoT communication module that can be used worldwide.

On the one hand, the new retrofit solution to be developed should fit directly into the existing slot without any technical changes to the machine and, depending on the machine location, use a mobile network or IoT satellite network for bidirectional communication. In terms of data technology, at least one data record should be transmitted daily from each machine equipped with this retrofit technology to the IT of a service partner. Other important requirements are

  • Minimal investment costs for the procurement and retrofitting of the required communication module in the communication slot of a machine.
  • Minimal operating costs for daily data transmission via a wireless connection.
  • Radio adaptability for worldwide operation, different locations, network operators and the associated national approvals.
  • Flexibility in terms of data adaptation to the IT systems of various service partners.

In this example, the machine controller communicates with the modem module via AT commands (extended Hayes command set). As these commands are stored in the respective control software and such software may not usually be changed, the IoT retrofit module must reproduce the required AT commands and convert them into the individual communication actions for the respective mobile radio or IoT satellite networks in order to transmit machine data to an IT infrastructure and possibly enable further tasks.

Advertisement

IoT wireless alternatives

© SSV

The major challenge for this retrofit task is the completely different machine locations, the associated different mobile networks and, if there is no corresponding terrestrial network at a particular location, the connection to IoT satellites in orbit with antenna technology similar to LTE.

If 4G mobile network coverage is available, an LTE-A modem (LTE+) can be used, for example. This enables data transmission rates of 500 Mbps and more. The LTE-A data throughput is suitable for live streaming of relatively high-resolution camera images as well as for high-performance remote machine access, comparable to wired DSL access. At other locations, however, there may only be an LTE-M radio network with a fall-back to LTE Cat NB1 (NB-IoT). This is still sufficient to transmit small amounts of data (e.g. sensor measurement data) without any problems. However, interactive remote access to the command line, remote desktop (VNC) or web-based user interface of a machine controller is not possible.

The most demanding detailed task at the moment would be the use of an IoT satellite network with LEO satellites (Low Earth Orbit; swarms of miniature satellites at orbital altitudes between 200 and 2,000 km). The interfaces and the provider-specific use of radio technology in this market segment require extensive expert knowledge (there are currently no standards for extraterrestrial IoT communication).

Other challenges include the possible data volume per month, the location-dependent time offset between transmission and reception times and data integration in IT applications. On the cost side, however, increasing competition is already having an impact. In the fast-growing market environment for providers of LEO satellite communication, most business models are currently in the positioning phase. This has an impact on the established pricing models for data connections with geostationary satellites, which until now have primarily addressed high-end M2M applications. In this respect, it is not easy to make meaningful comparisons of operating costs, especially as the details and number of satellites in orbit and the resulting differences in message transmission times also play a significant role. There are new providers that enable the operation of IoT devices with up to 750 messages and a maximum of 192 bytes of user data for USD 5 per month. However, further restrictions include specified daily limits for the number of uplink and downlink packets and a maximum total limit of 60 downlink messages per month (example: SpaceX subsidiary Swarm). An established competitor charges a monthly line rental fee of USD 15 per IoT device and USD 0.15 per data package with 50 bytes of user data. Larger packages cost twice as much (Iridium Short Burst Data, provider: Ground Control Communications). Calculated over the year, there are already significant differences in operating costs.

Integration twin in the cloud

The digital twin is configured on site when an IoT communication module is installed in a machine using a service app in the smartphone. This means that the respective twin instance knows the connection type (IoT satellite, LTE, NB-IoT). The app can also be used to perform various diagnostic and test tasks (signal quality of the wireless connection, connection test between retrofit module and machine controller, data connection to the cloud service).

© SSV

Due to the completely different WWAN variants, the data transmission bandwidths between the machine control system on the one hand and a cloud service on the other vary greatly. Taking into account the provider-related data volume limits and the special features of LEO satellite orbits and the visibility ranges that depend on them (depending on the location, data transmission is only possible at certain times within a variable time window), this leads to very different amounts of user data per day. However, the major advantage of such a modular solution is that it can be used to implement industrial IoT applications with global wireless coverage at an acceptable cost. Regardless of where the IoT retrofit module is installed, there is either LTE broadband or narrowband network access. Otherwise, an IoT satellite radio connection is used as an alternative.

In practice, it must be ensured that a standardized cloud interface exists for data access. In other words, the data user does not have to worry about whether the machine data is transmitted via a live data stream via LTE+, at hourly intervals via LTE-M or via a few daily NB-IoT or satellite data transmissions, which may also have been pre-summarized using special machine learning (ML) algorithms before being sent. As usual, the operator can access a website via a web browser or a remote desktop via a VNC client to view the current machine data image or equipment levels or to make new configuration data settings. With an LTE-A or LTE-M connection, the web or VNC server required for this could be located directly in the machine. With an NB-IoT or LEO satellite link, this is not practical due to latency. This problem can be solved with a digital twin: Web servers and websites or VNC servers along with remote desktop are always located in the twin instance within a cloud service platform. The actual data connection to the machine controller is handled by an agent that is configured in relation to the connection type used (LTE, NB-IoT or satellite) when the digital twin is instantiated. In such a solution, however, the digital twin should add a suitable "data quality indicator" that informs the data user whether the displayed machine data was obtained via live stream, interval-based transmission, time series data interpolation or possibly even through the use of special ML classification algorithms. The background to this is: with an LTE connection > CAT 9, high-resolution camera images can be transmitted in real time; if NB-IoT or an IoT satellite link is used, the image data should be analyzed directly in the sensor and only the result transmitted.

The author: Klaus-Dieter Walter is a member of the management board at SSV Software Systems.

© SSV Software

You can also find this article in the e-paper and further details on implementation in the next part of this series.

  • Xing Icon
  • LinkedIn Icon
Advertisement
Advertisement

You might also be interested in

Advertisement

Administration tray

Digital twin - concrete!

A defined standard for the asset administration shell is crucial for the interoperable and efficient exchange of digital twin data - so much for the theory. Nine partners showed what a specific use case looks like at the Hannover Messe.

read more...

Lenze

The digital twin as an accelerator

The digital twin is primarily associated with industry. However, the digital blueprint can be more than just a means of analyzing systems and processes. On the whole, the technology is a key to greater sustainability and climate protection.

read more...
Advertisement
Advertisement
Advertisement
Advertisement
Advertisement
Advertisement

OPC Foundation

OPC UA from the cloud to the sensor

OPC UA is also establishing itself as a standard for the exchange of process data between controllers. The next step is now to extend the concepts developed for the controller-to-controller use case for the controller-to-device and device-to-device...

read more...
Subscribe to our newsletter
Advertisement
Back to home