Automation and industrial communication are buzzwords for decades. They mean something quite different when you look at the 80s, 90s, 00s, 10s, today ... Where are we today? Not really far away from the 80s.
In February 1985 I attended the first time the GM MAP Team in Detroit (MI) - it was a cold week:
This was my first trip to the USA. At that time I did not expect to come back to the US for more than 130 times ... almost all trips related to standardization: MAP, MMS, UCA, IEC, IEEE, ...
The MAP (Manufacturing Application Protocol) project and especially the MMS (Manufacturing Message Specification) standard where the first combined attempt to define a single set of international standards for manufacturing automation systems. As you may know: they failed - because they were far too early.
MMS (ISO 9506) defines many services that have been smiled at. But if you read today (2019-02-23) what experts in the OPC/UA World are looking at - then you wonder how it was possible in the 80s to define most of the basic services the industry is looking for TODAY:
Since the MAP days we have learned some crucial lessons:
The French novelist Andre Gide nailed it when he wrote, "Everything that can be said has been said, but we have to say it again because no one was listening."
According to my 50 years of experience as a technician, the most crucial challenge in automation is this: People of different application domains (control center, RTU, protection, PLC programming, robot controlling, communication, security, engineering, maintenance, ... telecomms, internet, web, ...) DO NOT LISTEN TO EACH OTHER!!! If one expert of a specific domain talks - no one from the other domains is listening!
Talk together and have a look at what people have said and done even decades ago! It may be better than what you were told. It may save you hours and days and weeks ... of struggling.
In February 1985 I attended the first time the GM MAP Team in Detroit (MI) - it was a cold week:
This was my first trip to the USA. At that time I did not expect to come back to the US for more than 130 times ... almost all trips related to standardization: MAP, MMS, UCA, IEC, IEEE, ...
The MAP (Manufacturing Application Protocol) project and especially the MMS (Manufacturing Message Specification) standard where the first combined attempt to define a single set of international standards for manufacturing automation systems. As you may know: they failed - because they were far too early.
MMS (ISO 9506) defines many services that have been smiled at. But if you read today (2019-02-23) what experts in the OPC/UA World are looking at - then you wonder how it was possible in the 80s to define most of the basic services the industry is looking for TODAY:
- Client/Server
- Selfdescription
- Read/Write/Report
- Two-Way-DataExchange (like RPC)
- Standard Configuration
- Semaphore
- Event Management
- Journaling (Logging)
- ...
Since the MAP days we have learned some crucial lessons:
- In addition to Client/Server we need Publisher/Subscriber (as defined some 15 years after the MAP project in IEC 61850; GOOSE and Sampled Values)
- In addition to ISO/OSI Transport we need TCP/IP ... done in IEC 61850.
- We need many semantic models ... as the many Hundred Logical Nodes in IEC 61850, e.g., for electrical measurements MMXU or Temperature Supervision STMP, ...
- Standardized system configuration is key for any future automation system ... as defined in SCL (IEC 61850-6) for energy systems.
Fieldbusses are understood today as the "maximum credible accident". Heinrich Munz (Lead Architect Industry 4.0 at KUKA) says in the just published special issue ot the magazine "tsn & opc ua 2019" (www.computer-automation.de) on page 12: "Jeder Gerätehersteller muss die Anschaltung und das Engineering jedes seiner Produkte an mehr als zehn unterschiedliche Feldbusse entwickeln und pflegen - ein betriebs- und volkswirtschaftlicher Super-GAU." [Each vendor has to develop and maintain hardware and engineering tools for each of his products to be compliant with more than 10 different fieldbusses - economically a maximum credible accident.]
My personal resume after reading through the special magazine is this:
My personal resume after reading through the special magazine is this:
- The third fieldbus war started some years ago and is expected to go on for many years.
- The standard series IEC 61850, IEC 62351, IEC 61968/70 (CIM), IEC 61400-25, ... provide most of what OPC-UA and TSN are looking for.
- It is likely that the providers of traditional and Ethernet-based Fieldbusses will migrate during the next years to OPC UA and TSN.
- OPC UA and TSN will be implemented and used - why not?
- In the meantime the energy domain is already using and extending the semantic models, applying the needed services and feeling happy with the standardized configuration language.
- What else do you need?
According to my 50 years of experience as a technician, the most crucial challenge in automation is this: People of different application domains (control center, RTU, protection, PLC programming, robot controlling, communication, security, engineering, maintenance, ... telecomms, internet, web, ...) DO NOT LISTEN TO EACH OTHER!!! If one expert of a specific domain talks - no one from the other domains is listening!
Talk together and have a look at what people have said and done even decades ago! It may be better than what you were told. It may save you hours and days and weeks ... of struggling.
Posts
No comments:
Post a Comment