On rereading my previous post, ‘Rebooting marine renewables’, it occurred to me that it is mostly about why the technologies need to be ‘rebooted’, and not about how it should be done. It was already too long by the time I got to that bit and I just wanted to finish it off as quickly as possible. I think, therefore, that it would be worthwhile expanding on this point, as it’s actually the most important part of the whole thing.
To recap the story so far, a number of very large machines have been built and deployed but have not succeeded in consistently generating electricity with high levels of availability for long periods of time. Wave has failed spectacularly to do this, whereas three full-scale tidal devices have managed to achieve availability between 40% and 50%, if measured over a judiciously chosen time period, but much lower if measured over their entire time in the water. This is nowhere near good enough to proceed to commercial farms. Despite this there are several plans to build commercial arrays, including one that is in an advanced stage of construction, but that doesn’t alter the basic facts.
Marine renewables people have a habit of comparing its development to that of wind energy. They argue that the two technologies’ physical resemblance implies that their development paths will also resemble each other and that marine energy will follow the same trajectory that wind energy followed, only later.
This is just wishful thinking. Of all the technologies that could be the template for marine energy, wind was chosen, not for any sound scientific reason, but because it is successful. Wind is simply the technology that they want it to be like. Unfortunately, wanting something to be true doesn’t make it true, no matter how badly you want it.
If there is any reason at all (which there isn’t) for thinking that the development trajectory of a new technology is likely to resemble that of any particular old one, then it must be that the selection pressures that the technologies face, and their ability to adapt to them, are the same. All electricity generation technologies face the same selection pressures, but wind energy has been better able to adapt to them than marine energy has. Wind turbines are located in a less aggressive environment—the air—and are easier to access for maintenance and repair. These differences are bigger and more relevant than the superficial similarity that they both extract energy from a moving fluid.
There is, however, another ‘emerging’ energy technology whose development has some similarities to that of marine energy and which, I think, can teach it a valuable lesson, provided that you don’t push the analogy too far. That technology is nuclear fusion. Yes, you did read that correctly. I say that not just to make the obvious cheap (but valid) jibe that, for both technologies, commercial reality is always a constant, or even increasing, n years in the future or that, because its fuel is extracted from seawater, fusion is often described as ‘a limitless supply of energy from the sea’. The important lesson is what the fusion community did when they realised that they weren’t going to be doing mass deployment any time soon.
They didn’t pretend that they were still on course for commercial reality and try to ‘engage investors’ with a series of desperate-sounding stunts such as ‘new technology assessment processes’ and ‘supply chain gateways’. They identified what the precise problems were and went back into (or, rather, didn’t come out of) the R&D lab to solve them. (If you aren’t familiar with fusion’s early years there’s an excellent documentary entitled Britain’s Sputnik that was broadcast on BBC Radio 4 in January 2008.)
As an outside observer with only public domain information to go on, it looks to me like the wave and tidal devices that have been deployed so far have simply fallen victim to the destructive power of the marine environment. Wave devices are most vulnerable because they have to be deployed in the splash zone, which is the most destructive part of that environment, but tidal devices are also subject to wave action and to turbulence. In this respect wave and tidal technologies differ only in degree, not in kind, and the same solutions may apply to both of them.
Reliability modelling studies that have been published on wave and tidal devices seem to indicate that, using currently available components, they face an uphill struggle to achieve sufficient levels of reliability for commercial operation.
Reliability engineering is a mature discipline with its own professors, learned societies and practitioners. It would be astounding if this existing body of knowledge had not been applied to its maximum extent in the design and operation of the devices that have been deployed so far. Indeed, it would be scandalous if more than £450M had been spent without any of it going on an aspect that is critical to the projects’ success.
It must therefore be the case that careful systems engineering can’t push these machines’ reliability any higher. That is, the fundamental limits of components, materials and structures have been reached. These limits are where they are because demand for components, materials and structures with much higher levels of reliability than those already available is insufficient to stimulate research into how to make them. Most industries are managing just fine without them. Wave and tidal, on the other hand, can’t.
My suggestion, therefore, for how to ensure that wave and tidal eventually succeeds is to stop throwing good money after bad on near-market support activities and redirect it into fundamental research into how to make components, materials and structures that, when deployed in the splash zone or in strong turbulent currents, have sufficiently low failure rates to enable devices to just sit there and generate for a long time without breaking.
This is most likely to require laboratory-scale research into the fundamental processes that are involved in component failure and searching for ways to slow them down or stop them. Just as the nuclear industry had to develop materials that could be deployed in high levels of radiation without becoming embrittled, so the marine energy industry needs to develop materials that can be subjected to three-dimensional random dynamic stress patterns over tens of millions of cycles without cracking or deforming.
Another idea is to have a university-based project to build a medium sized, say 1/4 scale, generic tidal turbine, install it in a location with a strong current, operate it until something breaks, take it out of the water, conduct a thorough post mortem to establish the root cause of what went wrong, re-design whatever broke to prevent it happening again, redeploy and operate until something else breaks, and continue this cycle until the device can operate for three years without requiring unscheduled maintenance. When that has been achieved, they could progress to 1/2 scale and repeat the process, eventually ending up with a full scale machine that can operate for several years without breaking. This would be difficult to do for wave because there is no generic wave device, and so they would be faced with the problem of how to choose the right one.
Whatever form it takes, the solution is that these technologies need to go back into the lab for more research. This must be done by academic institutions to ensure that it is carried out with full scientific rigour and that the results are published in full. There should be no role for funding bodies that keep their results secret and regard them as their own intellectual property. There also needs to be a diversity of funders, so that a good idea that is rejected by one funder stands a chance of being approved by another. No organisation should be allowed to set itself up as a gatekeeper to the only route to access public funding.
It is likely that the solutions will come from unexpected directions, and at unexpected times. It may even be that what will deliver success to wave and tidal energy is general progress in the engineering and materials sciences, and that the best thing to do is simply to wait for that to happen before having another go. On the other hand, it would be good if research designed to enable wave and tidal energy benefited other areas in the same way that motor racing R&D has improved the safety and efficiency of ordinary cars.
All of this will take a long time, and its outcome is not certain. Some people may say that we can’t wait that long, that developing these technologies is a matter of urgency, but that would be misunderstanding the situation. Firstly, it has already taken 18 years to get where we are now, not counting the wave programmes of the 1970s and 80s. We have just passed Pelamis Wave Power’s posthumous 18th birthday and in September it will be 18 years since the start of MCT’s Seaflow project. Having waited that long already, another ten or twenty years doesn’t seem excessive. Secondly, as explained in my previous blog post, more of the same isn’t the answer. If government agencies continue to buy into the groupthink that the sector is forging ahead, then it will indeed always be jam tomorrow.
This is only an initial set of ideas, but it would be good if it resulted in a realistic discussion.
Remember you heard it here first folks!
© Copyright 2016 Howard J. Rudd all rights reserved.