Marine Energy: Electronics overboard! - EMF 2026

[applause] >> Oh, well, thank you very much and good afternoon. And I think given the conditions today, it's a warm welcome whether you intended it or not. So, if anybody does get too hot and has to walk out in the middle of the talk, I won't hold it against you. So, uh as you've just heard, I'm going to talk a little bit about marine energy and what I mean by that. I'll go through some of the different devices that are out there and talk a little bit about some of the projects I've been involved with, most of which have involved Raspberry Pi's and things that are very familiar to EMF and in one case a grappling hook, which I think was the story I promised in the abstract. So, hopefully we'll get there in time. Um I did realize when putting this together, it could easily have been a 90-minute talk. I'm not going to make you sit through that, but if you've got any questions or if there's anything that you want to know more about afterwards, come and find me in the Q&A tent. So, to begin at the beginning, as it were, what is marine energy or at least what do I mean when I'm talking about marine energy? Well, as far as I'm concerned, it's anything where we're taking energy from the sea. Okay? And generally that's going to be to turn it into electricity, uh although not exclusively. Now, there's a lot of different ways we can do that. I'll tell you that wave and tidal energy, wave and tidal power, are the two that you're most likely to come across, most likely to see. Uh wave energy comes in quite a lot of varieties. We'll talk about it in a minute. Tidal splits quite nicely into tidal stream energy, where we're taking advantage of the flow of water from place to place, or tidal range, where we're taking advantage of that height difference between high tide and low tide. They're not the only ways of doing it. There are a couple of others that are perhaps a little less uh commercial uh that are being investigated. So, you can take advantage of the temperature difference between bits of water, particularly somewhere that's a little hotter than the UK normally is. You can use the temperature difference between different depths of water or different areas to generate power. And you can also do the same sort of thing by taking advantage of different salinities, different salt concentrations in the water. They're a bit fancy. They're a little bit different. I don't know an awful lot about those, too, but they're interesting bits of technology. I'm also giving you a bit of an honorable mention to offshore wind. The actual energy bit of offshore wind, I don't think counts as marine energy, but there's an awful lot that it has in common in terms of the structures, the moorings, and working in a marine environment. So, you'll often find it included on the list. Now, the next thing to go through is like, well, why does this matter? Why is it important? Well, as far as I'm concerned, the UK is surrounded by water. We're an island with other islands. Uh so, we've got an awful lot of coastline, an awful lot of sea we can look at. And water carries a lot of energy. I think at this point we're all pretty familiar with the idea of wind turbines and how they generate electricity. Water can carry 800 times the energy in the same space. So, with smaller, differently designed devices, admittedly, we can get a lot more energy out of similar amounts of space. It's not just the amount of coastline we've got that's an advantage for the UK here as well. We're home to the second largest tidal range on the planet. The Bristol Channel, Severn Estuary, going to get 10 m between high water and low water. You think that's over a little over 6 and 1/2 hours between one and the other. That's a lot of water that's got to move from one place to another, and a lot of energy that could be harvested. There's a lot of wave energy as well, particularly off the northwest of Scotland, waves coming all the way across the Atlantic. And all these are ways with which we could get sustainable, low-carbon, and perhaps locally produced power. There's an awful lot of assumption that these sort of devices are looking at putting electricity onto the grid, and that is something that's happening. That is already in place. But you can also look at using them to replace diesel in offshore communities and things like that. So, there's a couple of other places we can use it. So, to look at a few in a little bit more detail, tidal stream, and I will admit this is where my experience is a little bit more so. This is again using that fast-flowing water. And when we find good places to look at tidal stream energy, it's pretty much anywhere where water driven by the tides is going to be funneled through some sort of gap. And that might be around headlands. You can think of it as one half of a funnel. Uh between islands and sort of those channels. But also places like estuaries. So, I've mentioned the Severn uh in the Bristol Channel. The water can accelerate as it moves upstream. It's all very local and very specific. And it's one of those things where it's important to remember that space is three-dimensional in this context. So, it can as well as having the coastlines coming in, you've also got the fact that the shape of the seabed, the bathymetry, can also become shallow. And you can get that sort of acceleration effect there as well. There are a few different types of tidal stream device out there, but mostly they end up looking like underwater wind turbines. So, to give you some examples, uh two real devices, these are not uh these are not renders. So, on the left-hand side there, you've got a device from a company called Andritz Hydro. You need to double-check the specific model in the picture, but that's about 18 m diameter on the rotor in the front of that one. So, this is not a small device. It's designed to sit on the seabed, generate electricity, and come to the surface as little as you can get away with. On the right, you've got a different approach. This is a floating platform. So, this was one that was being tested in Canada, hence the color scheme. Um and this has got six smaller rotors. And the rotors on this are about 6 m in diameter. Okay? Advantage of being on the surface, it's a lot easier to get there for work, for maintenance, and things like that. So, two different approaches both being pursued commercially at this point. So, moving on, we then have tidal range. And this has hit the news in the UK at various points. There's been a few different proposals. Tidal range uses that change in height of the tide instead of the speed of flowing water. So, if you'll indulge my inner physicist, if tidal stream is taking kinetic energy more directly with water flowing over the blades, tidal range is more about gravitational potential, taking advantage of that change in height over the tide. Mechanically, they're quite simple. You build a wall that's big enough to hold the water back until the height difference either side of it is large enough to drive some turbine. You let your water through, generate your electricity. And once things equalize, you wait until the height difference builds up again in the other direction, and let your water back through. So, these are lagoons, barrages, and there's a few examples of those around the world, and there's been a few proposals in the UK as well. And tech- from a technological standpoint, these look a lot like hydroelectric power. So, your typical sort of dam with turbines at the bottom, it's the same sort of idea. And so, a little bit of a diagram to hopefully illustrate what I was going on about in terms of the different water levels either side. Um And again, you're going to hit some midpoint at low and high water, but you can design a device that generates in both directions. So, that's the tidal side covered off. On to wave energy. Now, this is a little less my field, but very interesting, there's an awful lot of energy in the waves, and we're mostly talking about swell waves near the coast in this context, where the wind has driven over the surface to give us the waves there. There's a much wider range of devices for wave energy that are being investigated. Now, I've got three categories here, and they're not particularly exhaustive, but you can have devices that move with the waves, but you've got six different kind of directions you could react to. So, whether that's surge, sway, and heave are your three kind of linear motions, or roll, pitch, and yaw are your angular motions. You can design a device that would work with either, or a device that would work with more than one of these in combination. Or you might design a device that works with one and unintentionally reacts with the other, and that's a little bit of a bigger problem to deal with. The other torque devices you can get might be something that reacts to pressure change. So, you imagine the waves coming towards the shore, you've got a change in the height of the water. That change in height gives you a change in pressure below the surface. And you can then generate electricity from that either directly, and there's a lot of research going on looking at materials which give you kind of a capacitor effect that you can turn directly into electricity. Or you might use something that can inflate and deflate under pressure, pumping a fluid, often air, through a turbine. And that is actually a theme for a lot of marine energy devices. It's how do we turn energy that's from the sea into something that spins that we can hook up to a generator. Cuz taking a spinning object connected to a generator to make electricity is something we've been doing for quite a long time now. And so, it's a common theme in a lot of those designs. The last category is quite fun. Designed for something we'd call overtopping, which is, as it sounds like, you have your structure, might be fixed to a shore, might be floating. As the waves come up to it, they run up the device or break over the top, dump water into the top of the device, and as it empties through, drives a mechanism. So again, typically drive a turbine to give electricity. So, there's a few more options there. And there are to give you a little diagram of what some of these might look like. So, we have items that float on the surface and react to the motion of the waves. We might have what we call the point absorbers there that are designed to bob up and down in the waves that again react against something on the seabed. There's been some paddle type devices that have been tested. And these last ones I think are quite a clever little bit of physics in terms of how they operate. So, by having this kind of overhanging wall, as the water outside comes up and down, the water inside is going to try and do the same, pushing air through the turbine to generate electricity. And there's been a few of these tested as well. So, that's a fairly quick overview of where some of this energy comes from. Now, let's talk a little bit about what I've been involved with and the work I've been doing, and then I can get on to the tech of the stories. Uh in terms of the work we do as a research group at the university, we do computer modeling, trying to better understand how things behave, do measurements, often to back up that computer model or to improve it. And we also try and do testing, taking devices and putting them into a real environment. Okay, so taking them at sea or close to, and seeing how they perform against what we expect. Your computer model's only as good as the assumptions you've made and the data you've got to begin with. So, we build on that. But what sort of data are we interested in? And some of them you be able to take a good guess at. If we're looking at something like tidal stream, we're going to want to know what the flow speeds of the water are, both in terms of its average speed, give us some idea of how much power we can generate, but also things like turbulence. If you've been on an aircraft where you've experienced turbulence, that's a fluctuation in the plane's movement. You get the same thing with bits of the structure of a marine energy device, and that might affect how much energy you can convert to electricity at that point in time, but it also starts to think develop things like fatigue and how that might impact on the lifetime of these devices. So, it's a level of detail that we're interested in looking at, particularly as we already got these in the water. We might be interested in the surface behavior. So, if we're looking at the waves, couple of ways we can measure that. You can put a device that's going to float on the surface, measure its motion, and it can report some detail about what the waves are that have made it move around. Or you can use things like radar. You pick radar off the surface and use some information there. Or we can look at what's going on with the surface currents, and this was some work a colleague of mine did previously. Um we fly a drone over a tidal area, take some good quality video of the surface, and then from that work out what the currents are doing at the surface. You get a sort of 2D map of that top layer. Doesn't tell you anything about what's going on lower down. So, if you're going to be putting a turbine deeper in the water, that might not be enough information. But it gets you a large amount of data quite quickly. We might be interested in measuring tidal ranges. I would say we tend to gather that information cuz we're doing something else. The UK's quite lucky. We've got quite a lot of historic tidal range data that's already available. So, we will measure it, but we're not often depending on those measurements uh for the work we've been doing at least. And if we're testing a real device, well, we want to know where it is. Is it staying where we put it? Uh has it wandered off a lot? Do we need to phone someone to go and get it? Or is it moving as we expect it to? And I've said position and motion there as two slightly different categories. The position might be slightly longer term, whereas the motion might be how it's reacting as it generates power sort of second by second. That's a slightly more fine-grained detail. If we're testing something where we're generating electricity or trying to, then we're going to want to know how much power have we generated or could we generate? And the last category is one that I think anyone who's been involved in a project that maybe hasn't gone quite to plan will value. It's the what was that information. Something unexpected happens. There's a lot of value in knowing when did it happen so you can go and look at the data later. Whether that is that was a strange noise. I don't think we should have heard that. Or actually someone's just gone screaming past in a motorboat with a huge amount of wake. Let's look at the data and see if that affected what was going on. And both of those are examples that I've had when doing real tests. So, some things we might look at measuring. A lot of data, maybe a lot of sensors and devices that we have to get to talk to each other. Now, to try and get to some of the stories of some of the tech, two different devices that I've been involved with over the last few years. One of them is a full-scale but small-capacity turbine. So, what I mean is it's not designed to generate an awful lot of power. Um and it's an open turbine design, and by that I mean, if you want to go online and grab the CAD and have a look at it, it's all freely available. Okay. And that's the RRS or Remote River Energy System. Did a lot of testing with it in Pembrokeshire, and I'll show you a picture of that in a minute. We also have one which I think is more technologically interesting. And that's what we call the converging beam acoustic Doppler profiler. It's a little bit of a mouthful, hence the abbreviation. And this is something that sits on the seabed for a month or more at a time. And it measures those flows beneath the surface. So, I mentioned earlier how you couldn't get that with the drone, and that's surface data. This is one of the ways we can get that information. And I'll talk about that in a moment. So, to start with the turbine, we started with a simple mechanical device. So, we weren't generating electricity with it. Uh it actually had a water pump off the back of it. We were using that as our kind of way of measuring how effective it had been. And it's an odd thing to say, but as a university, we're not trying to make money directly from this device. So, it was never designed to be the best device we could build, but something that was practical to put together with local suppliers and local businesses as well. It's a floating platform, with the turbine behind it on an arm that can rotate and extend into the water. Which is very nice cuz you tow it out, and you only actually need that deep water once you get onto site. And you'll see a picture in a minute where you can see the blades on it. It's got a shipping container, so all your electronics are mostly in the dry. Still got some things beneath the surface that you have to think about. And the design we went with was something that was Okay. Low cost for something that is effectively a boat. Um but then we put some expensive sensors on it for the science that we can then take use make use of. And so that gives us this device here. So, it you see it's a floating platform made out of a couple of repurposed sections of bridge. Uh which is a nice little tidbit that says fair amounts of recycling before we even started. You see the shipping container in the middle to give us that working environment. And the two turbine blades on the back. Now, why two rather than three from the pictures earlier? Well, if you've got two turbine blades, it's a lot easier to just have some box section metal for the hub. Makes it a lot easier to manufacture. And you've only got to pay for two blades. You know, there's that. For scale, that's 3 m from tip to tip on those blades. So, not huge. Now, trying to build this, obviously trying to put electronics subsurface to measure things like the loads on those blades. Um And that's quite tricky. Got things that are rotating, things that are obviously going to get wet. So, you know, try to do things properly. Spend a bit of money on the connectors. So, you might see this black electrical cable running across there. So, we've got everything through one nice, fairly expensive electrical connector that's designed to be, I think, good for 100 m, which we're definitely not putting it that deep. There you go. And work from there. Excellent. Should be fine. We've got our mechanical seals. Plenty of gaskets in there along with some grease to seal it, all done up nicely. The blades could be put on without disturbing it, so we could test that it was all waterproof before we left the university. And then we got it back. And for those of you who can see the images more clearly, yes, that's water inside what should have been our nice dry electrical space. So, it's safe to say that the sea is nothing if not determined, and it doesn't matter how much you want to keep something dry, it will find a way to where it takes you. So, So, the sort of thing we looked at. In terms of the technology, this was all feeding back to a Raspberry Pi and laptops in that uh control cabin. So, fairly simple, but gave us good data for what we were looking at. So, to move on to the Doppler profiler, this was based off existing technology. So, you can buy these devices off the shelf that sit typically at the bottom of the sea or on the surface, and they use the Doppler shift. So, I think we're probably all familiar with the sound of a siren on an emergency vehicle going past. You get a higher pitch as it approaches, lower pitch as it goes further away. You can use sound under water to do the same thing. So, beams of ultrasound send a ping from the device. It scatters off in uh particles in the water. The Doppler shift lets you work out the speed and the time it took from sending that pulse to getting your echo. Lets you know how far away from the sensor it was. So, with one sensor, you're getting velocities of the water at multiple depths as you go through. So, if you're trying to work out how deep your turbine goes, it's quite a lot of useful information. Downside is these beams spread apart. So, typically if you're in 20 m of water, your beams would be 10 or 20 m apart at the surface. And that's fine if you could assume that the water is all doing the same thing across that space, but if you look at the photo I'll put up at the end, we know that that's not true. So, how can we work around that and get better information, particularly around things like turbulence that are quite small variations? Now, unfortunately, that means you need to build a bigger device. The standard devices are quite small. We're looking at about less size, and and they're sitting at the seabed, and you get that divergence problem. If you want the beams to converge instead, they've got to start further apart. So, that means more sensors, power supply, and something to coordinate them together. Trying to build something that was robust, fairly simple to build, and try to avoid needing a big enough boat with a crane to put it in the water. And we ended up with this. So, again for scale. That's a diggy trailer in the middle. Um corner to corner on this device is 10 m. So, from your little sort of cake tin size device, that's quite an increase. But, it means we could get that better quality data. The electronics went into a titanium canister that sat in the center. And again, Raspberry Pi with a clock, which is quite important cuz 20 m under the sea, you've not got the internet to set the time on that one. Okay. And again, some nice waterproof connectors to connect the data. And this is where you start running into some of the issues of dealing with the sea. If you see the picture up at the top right, those connectors are no longer the bright shiny brass color they went in. And unfortunately, you put two different metals in a conductive liquid like salt water, and you get a battery, and one of them gets eaten. And unfortunately for us, that looked like those connectors. Which we didn't think was going to happen, but it's something we learned the hard way. Now, in terms of the device, we had some mechanical issues. The Raspberry Pi actually did its job. That worked really well. It wobbled. And this is one of those wonderful things where you get a nice bit of positive feedback, and you don't know any of this is happening until you've recovered it, cuz it's 20 m below the sea, and you can't talk to it. So, one of the arms sat there with a little bit of a wake behind it, and it wobbled. So, then the wake wobbled, which made it wobble more. And yeah, not ideal. So, we did have to do a little bit of a redesign. And then we have the bit where we get to the grappling hook. It was quite a big device. We didn't want to use a crane if we could help it. So, some of my colleagues had designed a system where those blue barrels we saw in the middle were filled with water when we put it in the the sea, and a compressed air tank to empty them and allow it to float to the surface when we were finished with it. With an acoustic trigger. Send a little coded message by the old sound from the surface, release the valve on that tank, and up it comes 10 minutes later. Lovely. Tested several times, worked every time. Until we needed it. We even had a backup option. Send a different code, it releases a buoy that will float to the surface of the rope, so we can tow it from there and lift it. And that didn't work either. Cuz the battery in the box at the top on the boat with us had been damaged. So, imagine we've then got equipment 20 m beneath us, it's got to be recovered that day. And that's where the grappling hook comes out. And I invite you to imagine having to make that decision of, "Yeah, stick a great big lump of iron down vaguely near it and hope that you grab a bit of the structure and not any of the bits of the electronics or the batteries." But we recovered it successfully and we've got some lessons for next time for building another one. And in the meantime, we've got more information about what's going on at sea, so we can build better marine energy devices and work from there. So, with that, I'll leave you with a lovely picturesque view of part of Southwest Wales where we were doing that testing. And you can see just on the surface some of that variation. Thank you all very much for tolerating that talk in the heat. If you want to know any more, I will be in the Q&A tent shortly. >> [applause] >> Well done. Great work.