zek.at | Artikel drucken

Digitalisation set to play a central role in hydropower utilisation

Designed and developed decades ago, the architectural foundations of continental Europe’s energy infrastructure are still going strong. For the most part it was thermal power stations, ...

... supported by nuclear energy and hydropower facilities, that formed the essential backbone of the overall system. However, as fossil energy sources are gradually being phased out, the resulting changes require a full redesign of the entire system. A new system architecture will have to be set up to meet the European Green Deal goals, which call for complete EU-wide climate neutrality by 2050. Thermal power plants are to be reduced to a minimum, and their generating capacities are to be replaced mainly by wind and solar power stations. However, achieving these goals will certainly raise considerable challenges in terms of grid stability due to the highly volatile nature of these energy resources. New technologies are needed to deal with these challenges. In all probability, this is where the oldest form of renewable energy will enter the stage: hydropower – the one form of energy that is better suited than any other to ensure the required compensation on the grid. That said, it will take quite a lot of adaptations and technical innovation to make this possible. To get hydropower future-proof and ready for this demanding task, a high-profile network of researchers has been founded under the auspices of the European Energy Research Agency (EERA), which is dedicated to promoting current research at various locations through­out Europe. The coordinator of the crucial “Digitalisation” sub-programme is well-known, especially in Austria:  Dr. Eduard Doujak from TU Wien (the Technical University in Vienna), who took the time to give an interview to zek HYDRO.

zek: Dr. Doujak, some months ago the prestigious energy research network of the European Energy Research Agency (or EERA for short) was extended to include a new joint programme on Hydropower. Could you please tell us a bit about the background of this move and its main objectives.
Doujak: Most of the credit is due to the programme’s initiators from the Norwegian University of Science and Technology (NTNU). They had been lobbying for the programme to be included in the network. TU Wien was invited to participate along with several other universities. We thought a cooperation like that was a good idea and that it would open up some interesting perspectives. That’s how we became a part of the EERA. This research network is dedicated primarily to promoting the transformation of the European energy system into a climate-neutral enterprise by 2050. The umbrella organisation consists of a network of around 50,000 European researchers who are working in variety of relevant fields.

zek: Who can join this network as it stands now?
Doujak: It’s a research network that is open only to universities and research organisations.

zek: Let’s talk about hydropower: what is the structure of Joint Programme Hydropower with its six sub-programmes, and how is it organised?
Doujak: It’s quite simple: each of the six sub-programmes is headed by a coordinator and their deputy. The management board consists of only 15 persons, including the joint programme coordinator. They are located all over Europe, and they meet once a month in an online session.

zek: As one of the sub-programme coordinators you are responsible for Digitalisation. What does digitalisation mean in the context of hydropower?
Doujak: This is not an easy question to answer, as digitalisation has become a buzzword that can mean a lot of different things. That’s why we had to come up with a precise definition: we first defined digitalisation on the machine level, specifically with respect to sensor technology, data analysis, maintenance, and various metatrends. On the next-higher level the definition refers to the entire facility and ways of optimising a hydropower plant. Another step up we come to the system level, where we are dealing with questions of optimising entire chains of power stations on the grid. Apart from that, the term digitalisation is also relevant in context of environmental issues. Here computational models are becoming more and more important, for example in studying environmental impact or specific aspects like bedload transport. Today we have monitoring systems that can track bedload routes along an entire reservoir. In today’s new­er models this information is cross-linked with precipitation data. This allows us to make exact predictions where and how sediment is deposited. Finally, digitalisation also covers social aspects, with a focus on the operational workflow at a plant.

zek: What is the main focus on the machine level?
Doujak: On the machine level, it’s mostly about sensor technology: which sensors can be used, and what for? What is necessary to eliminate damaging impact mechanisms? We have put a lot of time, research and computation into studying these issues. Over the past two years alone we have invested a total of around 2 million CPU hours on studying transient operating ranges, including phenomena such as wake turbulence under partial load, machine start-stop behaviour, and load shedding. We are planning to present the results of our studies at the Viennahydro event this autumn.

zek: Where CFD calculations are concerned, this sounds as if there is still a concern that the sheer volume of data might become unmanageable.
Doujak: Of course, massive data volumes like that cannot be stored locally. So you have to plan ahead and decide which insights you are interested in, and then you capture only the relevant parameters for what is called post-processing. We couldn’t possibly store all of the parameters. That would quickly exceed our entire data storage capacity.

zek: Are things like High-Performance Computing – HPC – or Artificial Intelligence – AI – relevant in this context?
Doujak: Yes, absolutely, although you have to be careful to make a distinction here: HPC is being used already, for example, for numerical simulations in computational fluid dynamics (CFD), and for processing large volumes of sensor data.  AI is used to varying degrees in processing and interpreting the sensor data. Today numerical simulations are next to impossible without HPC, and in future this technology is likely to become just as impor­tant for sensor data processing. It’s all about parallel processing of measurement data under real-life conditions – that is, live. AI (specifically machine learning) is a currently a hot topic on the machine level. In the long run, of course, AI might be worth considering for numerical simulation as well, but that’s still far off, I think.

zek: The term “digital twin” has been mentioned a lot recently. Can you tell us how it applies to hydropower?
Doujak: When we are talking about “digital twins”, it’s important, once again, to define exactly what we mean by that term and how we want to use it. On the industrial level, for example, it means a virtual 3D model, which these days you can even explore with VR goggles in some cases. On university level, digital twin technology is defined as an interconnection between the real world with an artificial, virtual one. This way, we can more easily understand and avoid problems before they even occur. You can run simulations to test various operating modes, and once you have fine-tuned a process in the model, you can transfer it back to the real-life system. The hydropower industry has taken notice of the benefits and the extensive future potential. For example, some manufacturers are already applying this technology to their production processes.

zek:  Looking at the big picture, how is this going to affect the hydropower industry?
Doujak: It’s quite impossible to answer this question definitively. That said, in my personal estimation hydropower will move further towards grid stabilisation to stay profitable, especially in our geographic region. On a global scale, we probably have to look at future application areas  from a broader perspective. For example, one of the Horizon 2020 research projects in central Asia calls for the erection of a demonstration and research facility. It will take on a series of tasks, from irrigation and high-water protection to bedload management, including final energy production and grid stabilisation: a very broad range of tasks covering the latest developments.

zek: If hydropower is indeed headed toward grid stabilisation, what are the implications from the operator’s perspective?
Doujak: In terms of machine technology, it would require a higher degree of flexibility. This means unlocking the entire operating range. And doing that won’t be without its consequences. To study this aspect more thoroughly, we took a closer look at a specific machine and did the calculations over a wide load spectrum to see what happens under fluctuating conditions. In this particular case we found that the machine’s ageing process is accelerated by a factor of six when it is operated on the grid under such volatile conditions. That said, let me stress that this finding cannot be generalised and that it’s still early days as far as our investigations are concerned. We’ll need to do more research on various different types of machine before we can come to a more general conclusion.

zek: What would it take to get the machines future-proof and ready to take on these conditions?
Doujak: It would take changes to the basic machine design, with sturdier turbines. This has already been done for some of the more recent pump storage power plants. These turbines have thicker blades and a more rigid profile. The problem is that this reduces efficiency. Looking at older facilities, there’s the question what are they to do? How can the facility be kept up and running economically? That’s the key question.
Once the operator knows that the facility will deteriorate more quickly by a factor of x when it is operated that way, they can offset this loss against the profit generated. But it they don’t know that crucial x, what are they to do? How can they know with any degree of certainty that they will be profitable at the end of the day? It will take further extensive research to find reliable answers to these questions. We’ve got a lot of work ahead of us.

zek: What would you say are the advantages of hydropower research being included in the EERA?
Doujak: The main idea is that it allows us to develop holistic solutions. To do so, it is certainly an advantage if you can plan future projects collectively. It’s also a matter of integrating hydropower into a broader framework. That’s why we are trying to enable a sort of research collaboration between the individual joint programmes. Especially in digitalisation even very different areas of research can greatly benefit from each other. In view of this, I believe that we are currently at a very interesting and exciting stage in the process.
zek: Dr. Doujak, thank you very much for the interview.

© zek.at