Transcription

• Speaker #0

  Hi everybody, good morning, good afternoon to you, wherever your place on the world. I'm happy to welcome you to this new edition of the UAO Goes Live. And just to give you a bit of recap, we have been introducing the organization, we have been receiving users of the UAO technology. We have been doing some interviews, some calls. Last time we were with the company Stahl, which is one of the founding members of Universal Automation Network to understand their choices and why joining this, our organization and listening a bit more on their products. And today we now go to the, let's say, the third category of members of Universal Automation Network, which is academics. I'm happy today to welcome on stage with me the professor Valery Vyatkin. Hi Valery

• Speaker #1

  Hello Greg, hello everybody

• Speaker #0

  Hey thank you for joining us today on this UL Goes Live. I understood and I see from your background that you are already all set with the 4099 technology. Maybe before we go to the topic, can you explain Explain us a bit more what is your background?

• Speaker #1

  Okay, my background is applied computer science, but I'm doing all my life applications of computer science in industrial automation and for the last 20 something years I've been involved with this IEC 61499 standard in exploring how it can benefit Automation users. That's what we do in our lab.

• Speaker #0

  Great. And And what about the Factorido Future that I see behind you?

• Speaker #1

  Yes, we call this laboratory at Aalto University. We call it Aalto Factory of the Future. It is our research playground where we explore enablers of flexibility for future factories. Among these we explore mainly software technologies, IT technologies, and wireless communication. So we see this as a critical mass for changing how automation systems and manufacturing systems are designed.

• Speaker #0

  Okay, yeah, interesting. That's great. So basically seeing how could the automation of the future and the factory of the future look like basically in your lab?

• Speaker #1

  Yeah, we are trying with some niche technologies as I said. Modularity, very smart modules, autonomy. Autonomy in terms of energy, in terms of communication and of course on top of it intelligence. So how to design, how to structure the intelligence of future automation systems.

• Speaker #0

  Great stuff coming out. From what I understood from today's topic, you want to discuss with us about cloud commissioning using 64099?

• Speaker #1

  Yes. One big thing in automation systems development is their commissioning. It actually takes a lot of effort to commission automation systems. And we often even don't know how much effort does it take. take and it's unpredictable because at the last moment something happens and then commissioning engineers spent days and months on site just to get the system finally working and this is a this is a like a black hole where we believe that we can we can a little bit reduce this effort great

• Speaker #0

  great so i think you can go on with with klovic I see.

• Speaker #1

  Yes, okay so shall I start?

• Speaker #0

  Yeah go ahead.

• Speaker #1

  Okay thank you Greg. So today we're going to talk about cloud-based virtual commissioning environments that we develop at Aalto Factory of the future and this is a purely software project. But first where we are in southern Finland, Aalto University, You see our beautiful campus and a factory of the future is located in this nice garage. inside of which we play with advanced automation technologies. It's all fully IEC 61499 automated, and we have different devices there, different modules developed in-house, and we use advanced control devices all to confirm and investigate benefits of this technology for industry. But when it comes to commissioning 61499 systems specifically. 61499 means distributed automation. So how do we commission such systems? We develop the software using integrated development environment, which is a software tool that's running on our computer. And then we connect this engineering station to a network of program logic controllers. The magic of IEC 61499 is that we can deploy one program that is a block diagram to many computers, to many PLCs connected by network. So we can do that and see how this code runs on the real machines. But in case if we want to save machines for unexpected failures during the debugging process, we may also want to deploy code to real PLCs, but still connected to some simulation model that again runs on our computer and against that simulation model we can execute the code and see if everything looks to be correct in the simulation. So there is a traditional way how people do commission 61499 systems. So In ISA 61499, in the supporting tools, there are soft PLCs, which are basically some kind of analog of PLC that is running on your computer. And you can instead of deploying code to real PLCs, deploy it to those soft PLCs and again connect it to simulation model and make sure that everything works fine. There is a gap between this result, result of this type of commissioning and between the real deployment to real plant because soft PLCs are very approximate way of representing real PLCs. They usually have unlimited resources and they are not configured exactly like your PLCs in the field. they may not have all the libraries all the other way around. They may have too many libraries and you wouldn't see any incompatibilities. And if you have a really big system with dozens and even hundreds of PLCs, it will be quite hard to deploy those soft PLCs on the same computer where you engineer the system. I will use a small case study to demonstrate clothing. But first, let me run you through this case. That's an example of a very simple process control system where we have a tank which is filled with liquid. And that's typical for many process control systems. We control level there. We control temperature there. And maybe we are cooking something inside these vessels. And we imagine that we have a modern automation of this process control system with a number of devices. We may have smart sensors, we may have smart valves, smart pumps there and all actuators with embedded controllers. So we have a network of control devices and that is what is our hardware automation structure. Then we have a software. So, we have developed for this simple system, we have developed a software application using ISO-61499. You see the function block diagram that is network of components and here different components playing different roles. All together it's an automation system of this water, we call it water control system. It includes advanced control with a cascade of two PID controllers. where we can use different controllers in place of each master and slave. We can, here there is a recipe controller represented as a state machine. Here is an HMI implemented inside a function block. So one function block diagram in IC61499 can cover all this operational technology, information technology, IoT, and digital twins. And then having developed this, we deploy it to network devices. So we tell each function block where to go. And this is our demo case, which we will use to illustrate what Clovic means. But OK, so suppose you have developed this and now you want to commission it using existing tools without Clovic. So what would you do on your computer? You will open engineering development environment, which is in this case EcoStruxure Automation Expert tool. You have there, you see that we have a distributed automation system with six devices and we will deploy, we will start six soft PLCs on the same computer. On the same computer, we are running also a simulator of our system. And we are running HMI that is part of our 61499 code. Now everything is running. We have deployed code to PLCs. We start the simulation tool. In this case it's Factory I/O. And we see that everything works. Everything seemed to work correctly. And six controllers communicate with each other. provide data to HMI screen and they exchange signals with a simulator, and the dynamic simulator imitates the process. So this is local virtual commissioning using soft PLCs. But as I said these devices, these soft PLCs which we are running on the same computer, they do not have exact information about how many are your modules will be inserted in that plc what will be the memory limitations what are library configurations besides to do this type of commissioning you would need to have a license for uh this advanced simulator tool and it may not scale well if you increase number of devices or increase the system so what's the difference of clothing in the clothing You go to Clovic website, through which all the services of Clovic are accessible in a web browser. So you open the engineering development environment in the web browser. You open the simulator also in the web browser and most importantly in the web browser, you initiate that deployment in the cloud of Clovic. images of exact PLCs which are used in your system configuration. So all these containerized PLC images can be instantiated in the cloud and after that you can start the process of commissioning where everything is running in the cloud. You just have a view of that through your web browser. So here is a live demo where we have On the screen we have the Clovic website. We create first instances of PLCs. We have six PLCs. In this case, we used Iceblock PLCs. And you see that we created six images and they are now up and running. We can see the I/O status. We start the dynamic simulator of the process. Now everything runs in the cloud and we see through this Clovic website, we see how we can drag and see how HMI works, what changes in the physical process. So this is a way of executing it in the Clovic environment. So a little bit of internals of Clovic. Clovic system scans your system configuration, which is an XML and it discovers what type of PLCs you are using. And then it creates instances of PLC images in the cloud after which it starts all the necessary components and lets you to commission the system. So it's not only convenient and scalable, but also it is a collaborative environment in which commissioning engineers located in different parts of the world can do together the commissioning process before going on site. So what is the benefit of using Clovic? We anticipate that with a Clovic system if you take an average project of a large-scale automation system, like for example a refinery with about 190 programmable logic controllers and over 40,000 IOs. Commissioning of such a system can be estimated to 150 amendments and cost over 2 million euros. A big part of this effort and of these expenses are related to people being on site and this could be very remote locations and it could take a long time. With Clovic, we aim at reducing significantly these expenses. So for example, for this 190 PLCs project, we estimate that we could save up to six months of commissioning time and save about 1.7 million euro of money by shortening the commissioning and reconfiguration time. Besides, shift to ISA61499 as a platform provides further benefits to end users of automation because now you have freedom to choose different compliant devices best on the market available at the moment and this could lead to further reductions in your costs. Okay, so that's In a nutshell, our project. It is a research project at Aalto University, but this project aims at software product development. If our results are promising, we aim at some commercialization after this research project ends. Today, we are looking for industrial partners with whom we can do pilots and we can take their 61499 projects and demonstrate how this can be commissioned using Clovic environment. So please contact us using this email: clovic@alta.fi. And I'm looking forward to your questions. Thank you very much for your attention.

• Speaker #0

  Thank you, Valérie. It was really great and I think it gives us a good opportunity to think Yeah, when we are doing industrial projects, where to be more efficient, where to save money, how we can simulate as much as possible. And your project is exactly in that direction, to have more of the preparing and saving in the upstream phase so we can then better develop that. That's good You told us that it's a project which is currently ongoing. Do you have like a timeline on when, let's say, this project is supposed to end or something?

• Speaker #1

  Yes, this project will continue. The current funding and the current forum will continue one year more. So it will end in the end of June next year. And now we are talking to several partners members of Universal Automation who already have experience with 61599 and we are trying to run their systems and their developments inside Clovic. This we see as a win-win approach.

• Speaker #0

  Yeah, and I mean this is not only your call, it's not only limited to your members but as well to others in the industry who would like to try the technology and be able to do simulation on the cloud. Yes, that's a very good opportunity. I don't know if I see any questions coming from the chat. I don't see any questions for now, which means that you have been pretty clear on the explanation, which is nice. We also know that it's currently the, let's say, the starting of the holiday season for some of the, almost a viewer, so maybe there will be a bit less of activity. Nonetheless, for me, what is interesting is as well to show that the whole ecosystem of Universal Automation Network, which includes the users, the technology vendors and the academics, is quasi working together to bring the technology forward. And this is what you actually show with the Clovic project. So that's quite interesting. So...

• Speaker #1

  That's true, yes. The ecosystem is very important. Okay, we have a question.

• Speaker #0

  This is what we are trying to push forward. And I see here a question from Kodanda. Thank you for Kodanda. So that's great, Valérie. Just a generic question. How old is the 649? And when 61131 is being so widely used, why it took so much time for 649 to be considered old?

• Speaker #1

  after so many years yes you know i i've been in the 61499 exactly 25 years into related development so it's quite old and it started you know uh by the same team who developed 61131 in right after they commissioned 61131 and it went on on autonomous uh like uh sailing, they started looking into the future. So they anticipated the same experts from leading automation companies such as Siemens, Rockwell Automation, ABB, Omron and others, right? And Schneider Electric also. So they all anticipated that this technology will be needed because the new generation of automation systems, distributed automation systems are coming. So it was unpredictable when distributed automation systems would be be like in not now in a mature state and they are now seem to be in a very mature space so internet of things technologies wired and wireless communication so all these are now in place another thing is it's like a chicken and egg thing so only with mature software tools you can hope that the technology will be adopted by industry. But to develop material software tools, it takes a lot of effort. It's a very expensive thing. So you cannot in like one year, for example, create a software tool which will be used by real industry. So it's natural process. The first edition of the standard was standardized in 2005 and in 2005, Isagraph has released first commercial version of 659 implementation and it was the first round uh that was a first chance for industry to to learn about it and start doing real life projects with that and in 2007 next control was formed as a company it took 10 years uh to next control to come up with really material then it was acquired by schneider electric and now it's it's a second round i would say so it takes some some kind of spiral development, but eventually we get to the level where the tools are good, it's supported by a variety of hardware. Now we have universal automation and with that critical mass industries such as you name it like Exxon Mobil or Shell so they engineers of this industry discovered this technology and said oh wow that's a great technology. And everything works and everything is convenient. And it's so promising. So that's why it took a bit long, yes. But the technology did not... It wasn't frozen. It was developing all this time. So it's not the same how it was 25 years ago.

• Speaker #0

  So yes, it needed... first product to be available in the market. So you name Isaac Raft, you name the Next Control product. We have 4Diag as well, which is there as an IDE and others as well, which are being developed as we speak. And for sure, it just takes time for everything. And I mean, that's why organizations such as ours, such as universalautomation.org are there to kind of push the technology on the market and give some backings with the help of, for sure, users, technology vendors, and academics. So that's our way of... of life. So I see some other comments. I see a great initiative, Dr. Vietkin, so that's good. I see Konanda saying, okay, understand. Alexander, that's what Universal Automation is all about. Give it a real go. Yeah. Thank you, Alex. And thank you for being with us today. I see another comment from Konanda. I agree that soft PLC have unlimited features, but as you know the traditional controllers are well protected from cyber threats and the interference from external networks. How easy is it to bring soft PLC to that stage?

• Speaker #1

  Well in our case, our images of PLCs that we are executing in our cloud environment, in a cloud-like environment, they have exactly same security features. as real PLCs because they are using the same runtime of Universal Automation. And this runtime already has sufficient security features like authentication, for example, when it connects to engineering tool. So it's nothing different from a real PLC in that sense. And yes, you're totally right that security now is... has never is of greatest importance so it's takes it takes a lot of effort by software developers to address this good so i think we are reaching the the end of this session i

• Speaker #0

  thank you valerie for making it today and and presenting us and sharing with us uh the information about clovic and cloud commissioning which is a great topic which we have not yet not yet tackled, but we hope that the viewers could think about it and see if there are some pilot projects for you to kind of as well test the technology on a real base.

• Speaker #1

  Thank you very much Greg, thank you very much for arranging this, it's my pleasure.

• Speaker #0

  Yeah no worries, it's always a pleasure and for the viewers thank you for watching this LinkedIn live so We are not entering into the holiday season, so the next LinkedIn Live should be in September, unless we have some last-minute changes which could always happen. So wishing you a very good summer break for those of you who have some and don't hesitate to contact us either drop your message from the event or you have the email of Valérie and of Klovic. So please and easy to reach out. Thank you, and we see you at the nice end of the day. Bye-bye.