Rewiring Urban Heat with Data: Digital Twins, Heat Networks and Glasgow’s Decarbonised Future

Tunde Olaoye (00:02)
gonna be investing £300 million in the city over probably the next five years. The whole idea is to create like the backbone of a decarbonised heat and power network, which serves a lot of key institutions, NHS, public sector, housing associations, industrial clients. And we’re trying to really think about it like infrastructure rather than just straight up heat network.

Adam Sroka (00:46)
Hello and welcome back to the Hypercube podcast. I’m delighted today to be joined by Tunde Olaoye Director of Product and Innovation at Gren. Thank you for joining us here in Glasgow. Would you like to just introduce yourself a bit and Gren for the audience?

Tunde Olaoye (01:01)
I’m Tunde I’m Director of Product Innovation at Gren. So I’ve been working in energy infrastructure for about 12 years now and started at Gren in 2023. Gren’s a UK startup, we’re focused on heat networks, power networks, and at the moment, one of our biggest strategic cities is Glasgow and we’re doing a large project, which we’ll talk about later.

Adam Sroka (01:21)
Yeah, I’m excited to get into that because people in the office were quite stunned when they found out what it was and where it was and what was going on. But it’s amazing having you here in Glasgow with us today to do this live. You’ve had a really interesting journey through energy, like straight out of Man Uni, fellow Man Uni alums, so I’m really happy, into Babcock, WSP, SSE. Do want to just take us on that journey a little bit and sort of help us chart a course through energy to where you are?

Tunde Olaoye (01:44)
Yeah. So it was actually, you know, as well as I do Manchester Uni was a lot of fun. And as I was getting to the fourth year in my degree, I was applying to quite a few different places. And for me, I was doing mechanical engineering. I did business management and I just loved the idea of when you fuse them together. There was like that kind of catalyst of like, I’ve got this design idea. How do I commercialise it? How do I build it? And towards the end of my fourth year and our final project, I was like, I’m too busy to do so many aptitude tests for like Rolls Royce.

And BAE and I was like, I just can’t, I just need to go traveling. And Babcock was the first grad scheme, was like CV, two interviews, and they offered me a job in April. So it was wonderful. Cause I was like, I can go traveling and get a loan and start my job. But the grad scheme was in building service engineering. And I was a bit like, I’m not quite sure what that is. Cause it’s not one of my course titles, but yeah.

Let’s just do it because I can go travelling and it’s four placements across the UK and Babcock’s a large company. So whatever, worst comes to worst. I’ll just reset in a year. And it turned out it was a very, very useful place to start, but it didn’t seem that way. So Babcock at that time had the infrastructure projects for the Ministry of Defence. So what they were doing is looking after all their infrastructure in terms of buildings, facilities, RFM, Soft FM.

And one of their biggest pushes were trying to help decarbonise. So there was one of them in the Southwest and one in the East. And they had a lot of old building stock and obviously a lot of fossil fuel boilers and a lot of electricity that’s being consumed in inefficient ways across the whole fleet from some of the infrastructure in terms of how they manufacture tanks to just the lights in people’s bedrooms. So I went into that company, worked there for two years, did multiple placements and it was quite an interesting place because they kind of

taught me about first principles of engineering. It wasn’t glamorous because I was on like a seventies building site and everybody was really old and I’d just come from Manchester. I was like to my dad, I’m quitting. This is different. Yeah. I’ve been using laser machines and 3D printers and I’m out here and somebody’s got like a normal printer. You know, the ones with the wheels on it.

So that was really cool, interesting. And it was at the time when a lot of the EU regulations were coming in around 2010 about EPCs, display energy certificates, energy performance certificates, and really thinking about how do you regulate the built environment. So they were using that to kind of push through a lot of energy efficiency, energy upgrades, CHPs, solar panels. So it just gave me a big overview of actually things aren’t really made as well as you would think they are. They’re not designed that well and the energy systems that were put in.

From like 1970s are very hard to decarbonise and reduce energy. So it was a very tough place to start. So that was kind of where I began my journey and did a few interesting projects. I was doing a solar cooling project, which taught me a lot about rather than just go straight to the answer, cause they’d have like a planned maintenance thing where I say, right, this building’s overheating. We need massive heat pump system in there. And then I looked at it, was like, it’s overheating because it’s got loads of big windows and it’s south facing. Maybe we just put some solar film on it. Cause count. was like solar film.

cuts your heat loss by 75% and it’s about 90% cheaper than putting an AC system in. So it was one of those first times I was just like, get back to first principles. Don’t always just have a tick box because a heat pump’s an easy thing to do at that time. So then from there, I had a fun in Babcock, learned quite a lot, but then I was like, okay, I’m moving to London now and I need to get into probably a bigger consultancy with a bit more reach because I’ve probably

learned as much as I can in Babcock, which really good formative years and a lot of solid engineers I work with and really great scheme. Got chartered and then went to WSP. So joined the energy team. And then that was really cool because obviously everyone was young and I was like, Hey, I’m in London. I work for a consultancy. I’ve made it now. Drinks after work. Yeah. That’s the way forward. And that was really cool because they had a smaller energy team and that was really about

Adam Sroka (05:37)
Yeah, there we go.

Tunde Olaoye (05:52)
That’s when I got to understand digital systems in the built environment. So I did a lot of work on automatic meter reading and measurement and verification systems and BMS optimisation, as well as using software like IES to kind of thermally model buildings for the energy performance certificates. So in there, I did quite a few projects looking at metering systems, AMR systems, BMSs, pulling the data out.

And then try to use that to create strategic investment strategies for building portfolios. Yeah. So you’re looking at it just be like, if you look at your cooling systems, they’re running 24/7. Your cooling load is X. Your cooling system is sized to be Y. There’s like a 20% Delta in that and it’s not controlled properly because these are the set points. So there’s a lot of work with asset managers and landlords. So that was cool. That was just stuff like a meter. It’s a really simple thing.

Adam Sroka (06:24)
Yes ⁓

Tunde Olaoye (06:47)
And I kind of feel sorry for the smart metering rollout because

Adam Sroka (06:51)
This is a topic on this podcast

Tunde Olaoye (06:52)
Yeah,

it gets a bashing. it’s a weird thing because a meter is such a simple thing, such a simple instrument. But it’s really quite hard to configure in multiple ways and make sure the reads are coming through and all the data systems that underpin it. It’s like that simple complexity. So I learned a lot about sub-metering, AMR systems and just how something so simple and so powerful to a certain extent can actually be quite difficult to implement and OT systems.

Adam Sroka (07:00)
Stupid.

Tunde Olaoye (07:21)
or first before you think about the IT systems and the data that comes out.

Adam Sroka (07:25)
Must be the first person that stood up for the smart meter rollout on the show.

Tunde Olaoye (07:30)
Yeah,

well, it’s not great, but I feel their pain. Like I’ve done loads of meter and rollouts and I’m like, if I was doing that across the country and trying to get into everyone’s house, that’s painful. That’s painful. But we’ll get there eventually.

Adam Sroka (07:43)
Yeah.

Tunde Olaoye (07:45)
Let’s not dig into that one then, yeah, meters are difficult. I accept that. Then after that I joined SSE and I was just lucky enough to join SSE at a time when they were starting to really, let’s call it flirt with multiple energy technologies. At that time, obviously it was big thermal, it was trading and it was their business energy and they had residential at that time. So that was their kind of four pillars. And they started to look at EVs, heat networks.

energy performance contracts, which is really focusing on like campus estates and local authorities and saying, you need to decarbonise your estate by 20%. What do we do? And how do I get like a 10 year payback or I want to reduce my energy consumption by 30%. How do I get a 10 year payback? You just redesign my building systems to achieve those goals. And then they were also looking at behind the meters solar. So it’s quite a cool little place. We have a small little office in like Bramah house in London.

And was like, we were in SSE, but it was kind of a, again, a quasi start up because we were so small, like not even a rounding error on the SSE. Like, so we could just kind of get on what we needed to get on. We had the SSE brand behind us, but we weren’t necessarily having that oversight of SSE looking deep into it yet. And then all the different businesses start to scale independently and get slightly above the parapet. And over like seven years, we went from like me and my boss,

Kostas Papadopoulos, big shout out for he listens to this. He was a great boss and leader to get this started. And we basically started to build into frameworks and then turn like zero revenue into like a team of 16 and eventually got to about 20 million pounds of revenue in terms of EPC contracts.

Adam Sroka (09:25)
So was that almost like an internal consultancy?

Tunde Olaoye (09:28)
Yeah, sorry. Yeah, it was like an internal consultancy in a way, but we bid directly. Yeah. We would go to like local authorities would have a certain framework called Refit and then universities and hospitals also put tenders out on that. And then Refit had basically set up a structure where they’d have like measurement and verification plan contracts. They’d have like a structure of how you set your bid up, how you put your energy data in. So at least when everyone

went to bid for those projects. They had like a contracting form that was accepted by both parties because obviously I’m guaranteeing carbon or energy saving. So that’s where normally the complexity came when people try to underwrite that kind of investment. Yeah, you’d have to spend loads on an independent and a bespoke contract, whereas our framework allowed it. So we did some internal consultancy work for different measures like behind the meter solar and some of the heat networks team, but we were quite outward focused.

Adam Sroka (10:22)
Yeah, you’re building services kind of armed to SSE powered by the SSE brand and the behemoth machinery that they’ve got. That’s really cool.

Tunde Olaoye (10:31)
Yeah. And it was cool because we just got to touch any technology because our basically output was decarbonise or reduce the energy. So we could look at solar panels, heat pumps, control systems, LED lighting, HVAC. was basically you look at a building, just strip it back to its kind of individual energy streams. What are you actually trying to achieve? How does the people use the building? And then you redesigned it for that and thinking, all right, let’s go smart lighting. Let’s get PV on the roof.

Let’s change your control system. So it integrates with your lighting system. So when someone comes into a room, your HVAC comes on, your lighting comes on. And then let’s think about in the future, if you want to decarbonise, what’s the best route to do it. And not always heat pumps is going to be the best route because it’s going to be very expensive.

Adam Sroka (11:14)
So like, it was like a thread of decarbonisation and of big building estates and big old kind of semi-public sector defence has been a thread through your career. Is that fair to say? And was that what sort of led you naturally towards Gren because you were at SSE for a while, right?

Tunde Olaoye (11:30)
Yeah, I think that’s fair to say, I think it was trying to decarbonise and reduce energy and expose the harder buildings. You’re not starting from scratch. You’re looking at the existing stock and thinking about those challenges and how do you really think hard and deep and make this affordable? yeah, exactly what led me to Gren because my CEO now, used to work at SSE. So he used to be part of the heat network team. So he’s in that small Bramah house office and a few of us have stayed in contact over the years.

Adam Sroka (11:58)
Cool. Because like I naturally was going to say, but I’m aware of the questions I’m going to ask you. Naturally, the next thing for me was like, oh I just wish we did more heat network stuff in the UK. Yeah. What a segue, right? So yeah. Do want to talk about some of the work you’re doing and like the big projects up here in Glasgow?

Tunde Olaoye (12:13)
Yeah. So energy on Clyde is really cool. So what we’re trying to be is we’re gonna be investing £300 million in the city over probably the next five years. The whole idea is to create like the backbone of a decarbonised heat and power network, which serves a lot of key institutions, NHS, public sector, housing associations, industrial clients. And we’re trying to really think about it like infrastructure rather than just straight up heat network. We’re trying to think about how do you

manage your energy flows and the source of the energy, as well as some of the key arteries of delivery. So we’re just trying to think about heat networks differently because often it’s a bit of a chicken egg because you’re thinking to yourself, got to get the contracts with the off-takers, but we’ve also got to capture the energy source and buy that at the right price, then sell at the right price. Whereas we’re trying to think about it slightly differently. And a lot of the assets that we own in Gren are energy from waste plants or

waste wood biomass plants. So we’ve captured kind of the energy source in the projects we have, and then thinking about how do we build the big bits of infrastructure that allow us to transport the heat and power to where it matters. So it’s quite an interesting business model and we’re trying to think about it differently because we really believe that, you know, clean energy should be affordable for everyone and it can definitely deliver a win-win. So that’s how we’ve been kind of building our business model and trying to kind of scale that out now to have that real focus on.

It’s infrastructure. It’s not just the heat network. It’s kind of as important as your water pipes are going to be. Also your fibre installation. I would say gas, but gas is going.

Adam Sroka (13:52)
Get rid of the gas. I suppose some of the people that listen to this podcast might not actually be too familiar with what a heat network is. Because it’s a lot more common in the continent than over here. So if I’m like a hospital or a social housing stock, like why do I care about heat networks? Why, what, and how do I go about picking that over alternative approaches?

Tunde Olaoye (14:14)
So the whole idea really is it’s similar to your power network. Like it wouldn’t be great for everyone to have a little generator in their back garden making their own power. So it’s better to centralise all of the generation of heat into one place because I can get economies of scale by having a large heat generator and that could be air source heat pump, could be water source heat pump.

could be waste heat from industrial, could be from an energy from waste plant, could be the spare thermal from gas power plant. But the whole concept is if I’ve got a hospital, for instance, a local authority, a bunch of housing associations, a big industrial user, they’ve all got their own individual needs. So the hospital might need a 20 megawatt boiler. Industrial users might need 30 megawatt boiler. All the houses might need 50 megawatts. And if I combine them all together, I can get away with just doing a 60 megawatt boiler,

because I know that all their peaks won’t be in the same time. I can use stuff like thermal storage, which is super interesting, know, segue you can get into, but the most classic way of thermal storage is a large filled with hot water. So I can balance the peaks and loads. And if you do it correctly and you can co-locate lots of energy sources and a very effective heat generator, which utilises the best energy source for that particular location.

Adam Sroka (15:18)
Yeah.

Tunde Olaoye (15:34)
Then you can effectively deliver low cost, low carbon heat to more users rather than one person having to front the bill for their own energy system and look at their fuel costs, look at their maintenance costs, optimise the system to make it work better for them. They can just connect up to what we would call like a basically heat substation, which is just a small heat exchanger effectively that have two pipes coming into their building. And then they would take heat off that and then they pay per kilowatt hour. Same as like the grid.

So we put big pipes into the ground and effectively transport hot water through the city, through two pipes flow and return. There’s literally like a little microcosm of the grid in heat networks. You’re right, our Gren is from the Baltics originally. And there’s a lot of systems in the Baltics, a lot in Germany, lot in the Nordics as well, because they learned from very early on, because it’s so cold, that it wouldn’t be, you know, as efficient to be able to sell or build your own individual heat pumps and boilers for every house. So

that’s the kind of idea of a heat network and they’re now putting regulations in the UK to make heat networks more controlled and managed in the same way the electricity network is.

Adam Sroka (16:43)
You said city there as well. were you doing, because in my head it was always like either new build housing estates or like small isolated pockets on industrial and commercial sites. But we’re getting these like city centre networks happening now. That common?

Tunde Olaoye (16:57)
That’s not common fully. There are a few. There are a few in the UK, but that’s what we’re trying to push now to go to the level of saying, look, the way that we look at it is we won’t do it all because you can’t build a six scale heat network on your own. But what we want to become is like the transmission line, like the one, three, two KV cables. Like we want to be like, here’s a big generation source. We can generate the big bits of infrastructure. We can finance the big bits of pipe. We can take it to the key locations. We’ll have a connection to few big

customers that we can contract with. And then we’ll say, listen, new developer, new heat network operator. We’ve got heat substation there. You can take a bulk supply of that and build your own mini network or housing association here. We’re dropping heat to this location so you can connect up your houses and then we’ll give you kind of a bulk rate to go and look at that. So our kind of model is trying to work with the whole industry and not think that we can do it all, but think about the most difficult part is getting the big bits of infrastructure of the generation

and the big bits of pipe to where the clusters of use are. Once you’ve done that, then work with partners and different people across the industry to deliver.

Adam Sroka (18:04)
That’s

really cool. And the reason I’m interested, I guess, is like being based in Glasgow, live in Edinburgh. I hear a lot about actually heat networks might be the only solution, decarbonise some of the lovely tenement flats we have. We’ve got these incredibly beautiful flats across the cities that you wouldn’t want to touch or like stick air source heat pumps on the side of and things. People have said, heat networks will be the solution, but it’s a bit scratch head, not sure how to achieve it. And is that

a vision for the future you see that’s achievable and is there anything more on Energy for Clyde and what you’re doing in Glasgow that you’d like to share some details on?

Tunde Olaoye (18:38)
Yeah, no, that’s definitely it. So I think for us, the whole idea would be if we can build a really interesting kind of infra model, then it’s much easier for a developer or the housing owners to think, oh, you know what, I’m going to wait till 2029 because I know that the Energy on Clyde pipe is going to be coming past our building at that point. And at that point I can engage my kind of building service contractor and say, look, we want to do a kind of building level, heat network

And then we can connect into it. So their model, and that’s where we’ll get interested in the future. And we’re actually got a lot more digitisation on our network. We’ll be able to say, yeah, we’ve actually got quite a bit of spare capacity now because we’ve connected to that hospital, but they’ve done some energy efficiency work. So their peak is lower. So you can actually connect. If you connect in this day, kind of like an Octopus model, we can give you this price. This is a connection charge. And then you can go manage your internal buildings. And then it’s easy for them to kind of scale up and work out, okay, how much it would cost me to fit out my building.

And then what tariff can I sell them? Cause I know what tariff I’m gonna get it from. So that’s the kind of model and that’s what we’re trying to in the future build out to. But the first bit starts at like getting that core bit of infrastructure in the ground and kind of, we’re kind of looking at like, we will put it in the ground and we will invest kind of Kevin Costner. You build it, they will come. Field of dreams, heat network style.

Adam Sroka (20:01)
I like it. And that touches on a point then, like ⁓ product innovation, it’s not just new and clever ways to make stuff heat up, right? You’re actually financial innovation. How do you bill for this? How do you evenly like, yeah, a block of eight flats. How do you divvy that up? And you’re talking about digitisation tariffs. Is that part of big part of your role then, looking at the different ways actually this is going to be achievable for the customer?

Tunde Olaoye (20:23)
Yeah, I think it is. Think you’ve got like, as we’re in an investment world, you’ve got your base case, you know, that you can always just say that we’re going to do this. And effectively the idea would be we can sell it for X amount of per kilowatt hour and X amount of capacity charges and the investment will stack up and it will pay for the pipe and all the connection infrastructure. But the more interesting thing is really, you know, we’ll talk about a bit later is like, how do you put like a digital twin on your heat generation source?

And then how do you really optimise your heat generation so you can make it lower or you can capture more heat from the same piece of pipe? How do you use digital twins to kind of model the flows in the pipe and to make sure that you’ve got it at the perfect, I suppose, velocity in the pipe versus the capacity and then stuff like Delta-T and get really deep into the engineering really managing the Delta-T across the network so you can get again, more heat out of the same bit of infrastructure.

Let’s just say we can build this and we can deliver really affordable energy to the people who need it the most. And then once you’ve got that in, you’ve got so many layers of digitisation and optimisation of the system and the network that you can start to think about more innovative products and kind of got one eye on the future. But some of the stuff is just as well as, you know, getting off the plans that we’ve got today is sometimes quite difficult. Just getting out of the

site and getting into the energy center and working out the space constraints where you don’t necessarily know what your future loads are going to be. So there’s a lot of kind of second guessing and thinking about optionality you can get from digitalisation, but trying to find out what’s the one number you can build today that kind of works for today, but also gives you a bit of spare capacity. So that’s kind of the place we’re playing in, but Gren’s quite flat at the moment. So there’s lots of little things I’m doing across the kind of Gren portfolio, just because

where there’s a need, that’s a must.

Adam Sroka (22:11)
Yeah. Change all the jobs specs to just be, just get stuck in and yeah. No, and that was going to be where I went next was digital twins. Like you must have a load of, cause I’m just like trying to figure out how would I do this if I was doing it myself? And you obviously start small, learn some stuff, get better. So you must be accelerating like your pace of delivery and learning. And you must have a load of like optimisation and engineering boffins that are like bringing these insights together to

Tunde Olaoye (22:14)
Smuck around.

Adam Sroka (22:39)
to develop these new products. Can we get into the weeds of the digital twins and what you’re doing?

Tunde Olaoye (22:45)
Yeah, yes, it’s a thing. Like initially I was the buffing. So Sheffield, when we bought that asset, it had a control system that wasn’t working as effectively as we would like it to be. And the site team effectively could not manage the availability and they couldn’t control the plant the way we wanted to. And obviously when it’s a bonfire in a box, controlling fire with waste wood burning, you need to have quite a good handle on your control system.

So the initial thought process was it’s a million pounds to replace this thing. So we’re going in, changing the controllers, going into the full digital control system automation and looking at what’s the control logic, is it allowed the logic? Are we going to rip out all the DLCs? Do we have to put new safety systems in there? And we had a bit of a plan, had some engineering teams looking at this, writing a new description of operation, trying to work backwards and to code the system. And we found a company in Latvia

who had this digital twin system, which effectively is a way to read and write a control system. So OPC servers, ways that are built into a lot of SCADA systems and it allows you, it’s effectively an interconnecting protocol, which means that if you’ve got an OPC server installed on most SCADA, it’s kind of like an open source protocol that if you understand the parameters and tags,

Then you can rip them out and put them into another system. So they developed their own kind of digital twin system where you could put an OPC server on, take all the SCADA tags, bring that into their own server. You’d have a small server on your site, like a Dell PC with most of the processing capability, but talk to their offsite servers. Then they would map the process, replicate the SCADA system HMI on a web UI. So you can see the SCADA system live from any computer.

Then build some analytics on the data points that are inside the SCADA system. So you can look at like the temperatures in the boiler or the fans or the DP set points. Basically every IO you want to look into, you can look into and it took a while to, but when you get to 5,000, you’re like, how many do you want? I was like, 5,000. But it takes a long time to get 5,000. It’s like, let’s go Pareto. Give me 20% that cause 80% of those problems. So then you rip those in and then you can start to do analytics from that.

Because obviously if we know the temperature of the boiler, we know the fan flow rates, we know mass fuel balances, then we can start to calculate how much heat is coming out of the boiler. We know the movement of, for instance, the pusher, actually the ram feeder, which pushes wood into the back of the boiler. Every time it pushes it in, maybe it pushes like three, four kilograms. So if you can take your historical data from how much wood got delivered over the last month.

And then you look how many times a ram feeder moved, you basically work out how much is pushing in and then work out, okay, this amount of fuel has gone in. This is the heat mass balance. This is how much energy came out. And then you can do analytics. And then the final point is the machine is meant to control the machine. So every minute, once we’ve done the analysis on the boiler, understand the heat mass balance on the boiler and how it can work at its most efficient point with primary air, secondary air.

Should be controlling like some of the recirculation flans dampers. You can then say, optimise it. So we use the lowest amount of fuel, but achieve the temperatures really required in this particular boiler. And every minute the digital twin would then look at every setting, look at every important parameter, and then slightly adjust for instance, the ram feeder or move up the primary fan hertz in zone one up to

two hertz higher because you’re not getting enough burning in this zone because the temperature gradient is not right. So every minute it goes around and makes slight adjustments to rather than normally when you have a SCADA kind of operator on these boilers. Yeah. And it’s also when it’s fuel, it’s kind of like, just had a big bit of dry fuel got in. Temperatures just spiked. I know after the fact, my fourth temperature sensor has already gone up and it’s breaching. So now I’ve got to kind of, whereas the digital twin will start to see those spikes incrementally and start

change in secondary air to kind of cool the zone. So it’s been an interesting way to put technology in. And it’s been a big learning curve for us because we’re at probably like zone two where at the moment we’re pulling out all the analytics. We’ve got all of our boilers and all of our systems that generate power and heat now fully configured so we can go anywhere. We call it Netflix. Cause on a Saturday night, I’m like, how’s the boiler doing? How much? what? 5.2 megawatts love. Give me some wine. It’s a big night.

Adam Sroka (27:28)
That’s really interesting. You’re like, I have a few insights, like a million pounds to do that doesn’t seem like a lot of money actually. Like thinking about consultancy and doing projects and picking a part. Cause you’re effectively reverse engineering like a piece of software that was written however long ago, like secret squirrel IP. Wouldn’t know how that was done or why. And you’re turning that into, yeah, like an actively managed sort of continually optimised control system that you can scale digitally and learn from.

Build up this wider portfolio of analytics. Cause we do bits of optimisation, but lots of our customers. And I often like to talk about like the one in a million event, like how many one in a million events have you had? Probably not many, but you scale it across a portfolio of assets and you become really smart and like you just kind of the power of your analytics grows over time. It’s a really cool, really cool approach to doing it. Yeah. I’d love to like sort of see more under the covers of what you’re doing and have a look at some of that stuff.

Tunde Olaoye (28:27)
Yeah, yeah. No, it was actually, cost us a million. That was the, that was a price tag to replace the DCS. So doing this digital twin was significantly, significantly like in the orders of magnitudes cheaper to do that part. We had to spend a lot of money using very smart consultants to help us with the physics and working with a guy called Darren. And he’s really brilliant. And he was just like one of the areas that, which I didn’t quite anticipate being one of the biggest value adds is actually just the surfacing of the data.

Cause my thing was more like, right, we’re to get to the machine, running the machine and we can optimise this thing. So I was like, let’s sprint at this and let’s go. And if we get that, then we don’t need the people on site, but there’s so many little things wrong with that boiler that weren’t the control system. And when Darren came in, he was like, you know, able to crunch so much data so quickly that he was being able to like, yeah, let’s look at this, let’s look at this. He talks about mega, mega joules per second rather than

megawatts because he’s like in thermal world I’m going old school and I was like wow this guy’s serious.

Adam Sroka (29:31)
That’s very cool. What other ways? So like there’s quite a whirlwind tour of like a really complicated system. What other ways are you using technology to your advantage?

Tunde Olaoye (29:40)
Yeah, kind of building on that is I think this is, as you talked about, optimisation. And I think this is sometimes where people get too stuck into the business case at the first stage, because it’s kind of this thing of like, it’s corny, but once you surface this data, you can effectively build so much on it. there’s so many things you don’t necessarily know that you can build into use case. So one example is at the moment, we’ve got these kind of operational management

I can’t remember what the A stands for. I’m going to get told off by my data analysis, but we call it OMA tool. And because we’ve got the digital twin, one of our biggest challenges was our accounting team after basically do accruals. They’re like, how much wood got delivered? How much light how many consumables were delivered? And all they can see is the end invoice. And when they know how much power is generated, have to wait till end of the month and go into like the platform and then download that. And then there’s so much data they have to do. So when it gets to the end of the month,

think about portfolio of lots of different biomass and power generation assets, they were literally just flat out. And then by the time they finished, it’s the end of next month. So now with the digital twin, one of my data analysts has now built like a financial tool that sits on top. So we take all the digital twin data, we basically convert that into our P&L data. We know the cost per tonne of fuel, we know the cost per tonne of lime know how much urea, how much gas we’ve burnt for like secondary firing when we’ve gone down.

And we’ve got this great Power BI dashboard now, which get updated every week. So every week we have a call with finance. They can look into it and they’re like, okay, so what’s the P&L looking like? Where are we? why have we got this big shoot up in gas? ⁓ we a failure here or this thing happened or the lime cost looks much higher. We’ve got invoice for this. So how much we got in stock. And the whole thing is trying to compress down like something that takes, you know, 10 days for them to do accrual down to effectively.

It takes you basically half a day to do it accrual because every week basically got all the numbers and the only game is to keep on refining the OMA tool because when there’s a discrepancy, we just need to look at do we need to recalibrate the model?

Adam Sroka (31:45)
In your edge case. Yeah. That’s really cool. And I think like, want to ask the AI question, but I’ll get ahead of myself a bit. Like people always ask, like, what should we be doing? What should technology do for me? And I genuinely believe like the energy sector is complicated and messy enough. You’ve got like financial world, physical world, you’ve got all this stuff trying to come together that you don’t want the AI and the machine doing the thinking bit. You want it to do all the cruft.

Like get all the numbers and stick them into one table. So I just look at them and it’s easy and read all these excels and so on and so forth. So I totally see that, kind of tooling that OMA tool, sounds like the kind of thing that we always think is one of the best first passes to go after. If you’re in the BESS world, like settlement of mixed portfolios, like, how much did we say we were going to do? What did we do? Who gets paid? What is really, really tricky. So are you investigating AI? that like on, on

Tunde Olaoye (32:37)
Yeah.

Adam Sroka (32:41)
Everyone’s kind of lips or are you skeptical? What do you think impact is going to make at Gren.

Tunde Olaoye (32:47)
I’m a user. I’ve made friends, you know, with multiple GPT Claude, all of them little friends. I think it’s a really interesting space. I always think of it like I was chatting to my wife the other day, cause she just went back to work after a year of maternity. Like often, because we actually met at WSP, we were basically, shout out to Erieta by the way, I’ll get told off. I didn’t say that. Love of my life. So she was going back to work and I was like,

listen, we always have long conversations. You come in because we work at home and she’s like, what do you think about this? And asked me about this. And I’ll explain something to her. And we’d be there for ages. Like you’ve got a new me now. You got Chatgpt and like, don’t look at it just a glorified search function. It’s like a calculator for like ideas and words and you can bounce stuff off it. And like what would take you ages before to go read lots of guides or lots of like industry standards. You can go like.

These are the core industry standards. I know that the information’s in here, but rather than sifting through like 300 pages, put that all in, ask it the right questions and just get started. Don’t think it’s like, I can ask it to like design me a heat network for a city scale, but it can tell me like, right, thinking about the new regulations that are coming in, what are the areas I need to really consider when designing a heat substation? And then you get these ideas. It gives you a place to really focus and then you can kind of jump into that.

And for us at the moment, we’re kind of looking at it and thinking, okay, where are the best use cases for it? And how do we as a startup as well, think about using it most effectively, but also take into account the risks and making sure that you’ve got a place where if you’re putting your information.

It’s protected. You know, you’re not necessarily training the model and stuff you don’t want to use and having a good policy. So put an AI policy out there. We’ve got our own kind of LinkedIn to Chatgpt open AI Gren version. It’s quite nascent with us. And one of the things that we’d love to do in the future is now we’ve got this digital twin. We got all this data. One of the operational team challenges is this pumps about to break. Yeah.

Like what was the spec originally and what do I need? Whereas if you had all this data and you knew the characteristics from the ET, can say, what’s the best size pump to achieve the flow rate that I require? And then it’d be like, yeah, these are the four pumps. And this is a good supplier that’s on your supply chain system. So order from them. That’s kind of what we’re thinking about how you can use it. But yeah, early days.

Adam Sroka (35:15)
I’m in agreement. It’s an interesting space and you’ve got to, I think every industry has got to tread carefully. We’re all still kind of learning. People ask me like, what should I do in AI first? I was like, I don’t know, just get going. Like no one knew it two years ago. It’s all new, this flavour of the technology. But that like looking at specs and parts, we did a piece that built like a recommender system. You know like Netflix is, ⁓ you might also like this or Amazon’s, like people also buy for…

Service call-outs. yeah, your boiler and your heat network off in the middle of nowhere goes down. And instead of sending someone to go to site, having a look and go, ah, that’s the thing. Can I look at all the log data and recommend the top three failure causes? So your engineer turns up with those three parts in case it is one of the top three and then it is massively reducing your call-out times, down times and stuff like that. And then seeing some really interesting stuff with LLMs interacting with

Tunde Olaoye (35:53)
system.

Adam Sroka (36:13)
optimisation engines and stuff like that, we, ran a community meetup up in Edinburgh recently and we had Lars from Edinburgh Uni who’s the reader for optimisation, like math super boffins, shout out to Lars. And he was, we asked the question like, is AI coming for your job to everyone? And he was like, and for his whole bit was like, you would have to either use the LLM to translate the universe into the math that the, the

Tunde Olaoye (36:31)
Try.

Adam Sroka (36:41)
optimisers need or the other way around and both really, really hard. So I think I’m safe for now, but there is some success in this space. It’s an interesting place. And I think like it is only two and a bit years in three years in. So we’ve got more use cases to come. No one’s sort of shown a game changing use case to the industry yet. I don’t think that stuff’s coming.

Tunde Olaoye (37:01)
No.

Yeah. That’s really cool actually, the boiler thing as well, because, you know, these are the little things that kill projects, especially in heat networks. You know, the biggest thing that kind of kills a lot of projects is where you put that O&M line into your project finance spreadsheet. You’re like, what’s the O&M going to be on this heat network? And it’s like, well, what are the parts? I’m going have to go up to all these things. Then it just like, you pass that onto your consumer. And then it’s like, why is this twice as expensive as me getting a heat pump? It’s defeated the purpose of me doing this. And

when you break it down, you’re like, the energy charges, maybe 30% of what you’re getting charged. Big charge of that is like 15, 20% is now your O&M cost because somebody’s made a big assumption on what that’s going to be because it’s quite a hard thing to O&M, so with tools like that and actually being able to go backwards and look at it, actually, what did we spend on O&M or what does the industry spend on O&M and then compressing that and having a fair mechanism to charge, like if it goes above X, then you have to pay for this. But the natural bill will be charged

just what we think we need to do to keep you going.

Adam Sroka (38:05)
That’s why I think what you’re doing at Gren’s so interesting because as you scale and do more and learn more, that body of knowledge will only grow. it feels like you’re at the forefront of building a very digital first, data first organisation, which can’t be said of a lot of the energy sector. How do you think the pace of change is in the industry? What are we good at and what are we not so good at?

Tunde Olaoye (38:31)
Yeah, it’s this very interesting one. I think it’s like, it’s kind of like the last story in a sense where, because you’re in this technical field, like, and I’ll give you an example of these digital twins we’ve had to put into our power stations. Like we had a bit of an incident where we went, we ran before we could walk at an Icarus moment because we put the technology in, we started to utilise it a bit early and effectively a few decisions led to a failure on the plant.

Which was driven by the digital twin. I mean, everyone’s like, I see this what happens. We’re like, yeah. And then we look back at the data and we realise, yes, there was probably some process and we should have slowed it down to allow everybody to fully understand each parameter change that’s going to be made. But at the same time, there was some underlying issues with the boiler. And I think where this gets to is kind of like in the energy industry of a lot of very intelligent, technical people who are doing very difficult things like.

Running a power station, understanding the thermodynamics of a combustion system, the fluid dynamics of a pumping system, but I’m thinking chemistry of a battery system, like electrical engineering, like they’re like, I’m doing such a complicated thing. How could this machine or this data help me? You know, I’ve studied for X amount of years or I’ve been a technician for X amount of years. So there’s this slight level of hubris thinking that I’m doing this really complex thing here. This digital system could no way help me

to do my job and it’s not like one or the other. And I think that’s one of the kind of risk problems in our industry. They put too much risk and like, it’s going to be too dangerous to do this kind of thing. And you’re like, well, again, just get started. You don’t necessarily have to use all the functionality, like surface the data, look at the insights and then make decisions like, yeah, it can’t size my cable to go in here, but it can give me a bit of a fault study because it can see how much my

electricity consumption is and it knows what the generated characteristics are. So it can do one job for me, but the other nine I’ve got to do. So I think that’s a bit of the challenge of trying to get people to be open-minded and not too stuck in there. Like in a saying for legal and commercial risk, where they’re like, there’s no way it could help write our contract. There’s too many risks and regulatory things around it. It’s kind of like, we need to be a bit more open and say, it’s complicated, but it’s not necessarily like brain surgery.

At that point, it can help you do some stuff.

Adam Sroka (40:57)
Yeah. And I talked to the team a lot about the two, I think about the two flavours of hard work and you’ve got like quantum physics, complicated quantum physics is hard, right? And move a hundred thousand rocks is hard and hard in very different ways. And we should be pointing the AI at moving the rocks. Yeah. Like, because it can do that pretty quick. Whereas we’ll keep the quantum physics for us for now. Yeah. And like people are like, Oh, is it going to take my job? I was like, well, yes. If you can write down

everything you do in your job. Yeah. I’ll write you an AI that will take over and like people can’t even write down what they’ve done in a day. So I think we’re safe for now and that I totally believe in what you’re saying like little bits like pick off things, small wins, get the ball rolling.

Tunde Olaoye (41:34)
Yeah

But what’s also interesting is reading a book called Utopia for Realists. It’s quite an interesting book. I can’t remember what the author’s called it’s a super interesting book actually. And it was just like, I just wanted to read something different, like not just business or like managerial or just like fiction. So was like, this is a bit of a out there topic. And one of the things you talk about is if you look at the industrial revolution, there was like, I think it’s like a hundred fold new jobs.

Even though like you had the mills and the pump replacing lots of old jobs and like a lot of horses lost their jobs. But you had a lot more taxi drivers, you know, like there was only so many horse and carriage drivers, but there’s so many more taxi drivers. And even with like new technology now we’re saying self-driving or driverless cars, we’ll lose all the taxi drivers wherever they go. But it’s kind of like, well, there’ll be lot of maintenance on those cars. There’ll be a lot of things that the machines won’t do. So there is, there is a truth of like, there will be a lot of job

change over, but I think it’s, we’ve to get to the point of thinking about, all right, how do I leverage this to make me better and give me more chances to like study more, think more, be more creative. Like think about actually what are we doing? Cause we’ve got to a point where we’re just doing stuff because we have to, like you get to work, answer emails, do this contract bit, do that design, go to that board meeting, make that presentation, do that bit of research. And then you’re like, am I actually spending enough time? You’re like, what am I trying to achieve here?

Adam Sroka (43:04)
Oh, a hundred percent. I talk like, what’s the value add that you as a human, with dreams and ambitions and likes and dislikes, what’s the value add that you are bringing to work? Yeah. It’s not copying numbers out of that spreadsheet and putting them in that one. That’s not, that’s the bit that we should be getting the machines to do all the boring stuff that no one, when they were six years old, said, oh, what do I do?

Tunde Olaoye (43:29)
Yeah, even email sometimes is just like when you’ve got really, you know, it’s hard sitting between a technical commercial and like sending emails to different audiences. Now I can write like my just core email and then say, okay, is this understandable for someone who’s not technical or is this understandable, it’s not commercial? Like, no, that’s not, use this phrase like thanks. Whereas before you sit there like,

why has it taken me 45 minutes? Like I’ve ruined my whole day trying to write this really good email to get it across. Cause I don’t want loads of questions. And now it’s like two minutes, done. Great email, which everybody understands in clarity. And it’s like, that’s great for me. Cause I’ve got like a little bit dyslexia. So writing emails and stuff like that is way easier. And it makes it way easier to communicate really complicated stuff without wasting 45 minutes of my day, which can now be used for making nice coffees.

Adam Sroka (44:20)
I’ve gone a step further in that I can’t stop talking. So I found there’s a tool, Whisper Flow, and you install it on your computer and you can just talk into any text box on the machine. I just talk to myself all the time, like, and it, it runs, it doesn’t just transcribe it, runs it through a large language model that learns to kind of speak like you. So that Slack message email, it’s like, I’ll just waffle for 10 minutes and then it will just write this like role formatted bullet points, all that. It’s really good. It’s a really good tool.

Tunde Olaoye (44:48)
wow.

Yeah, I like that.

Adam Sroka (44:51)
Okay, interested to think about outlook, right? If we were to do another episode in three years time, you come back to the Hypercube podcast when you’ve conquered Great Britain with heat networks what would success look like for Gren and for you?

Tunde Olaoye (45:06)
I think success for us would be energy on Clyde is like a mainstay in Glasgow. And like people talk about it, like the armadillo or some of the other buildings. Like when people talk about it, like, oh, you connected to energy on Clyde and whatever our nickname is going to be, Costa del Clyde. I snuck that from you earlier. It’s just so hot in my house. But that’s where we want to be where we’ve got like a big chunk of the infrastructure built. We’re connected to some of the serious players. So hopefully we

working with the local authority, working with the NHS, some of the significant industrial builders and delivering heat and power. Like beginning then to start to say that now we’ve got that core infrastructure in. Our two next steps are how do we get a partnership of different heat users and heat network developers connecting into our assets and building their own mini heat network? And then how do we really start to leverage our kind of data platform to think about, I’ve got all these thermal stores,

I’ve got this power market that’s moving up and down. How do I really leverage, you know, the flexibility in the system to try to drive down my heat price, keep on driving it down and really utilise heat when it’s the best time to utilise heat, put power when it’s the best time to utilise power. So I think we want to be in a space where we’ve got core infrastructure in the ground and we’re starting to really go into two levels of building partnerships with other heat network developers, power developers

and then really starting to really flex out the system so we can really deliver kind of innovative heat network products that actually provide affordable, low cost, low carbon heat so everybody can connect and working out how do we get like some of those most difficult like housing associations, how do we decarbonise them at a price that’s like, yeah, that works for you. And you can help a lot of people who are in fuel poverty, get out of fuel poverty at the same time as decarbonising. So that’s kind of the goal.

And I think that’s when we come back in three years.

Adam Sroka (47:04)
That’s not even that long a timeframe to do all that, to lay core infrastructure. Is that achievable? Do you think that’s how this will play out? You’re gonna have significant infrastructure in Glasgow in under three years.

Tunde Olaoye (47:16)
Yeah, that’s the plan. That’s the plan. we’ll definitely in two, three years, we’ll have our planning permission, have the core pipes and cables going in and have the big connections ready. And hopefully we’ll be ready to cross the river. Cause obviously we’re in the south side of the Clyde. we’ll be north of the river. I think I’ve got that right. And yeah, so we’re making good stead and we’re doing a bit of our first pipe digs in the next couple of weeks at Chief energy on Clyde. So we’ve kind of just said, look, we’re going to.

open up the road, put the pipes in and Kevin Costner. If you build it. They will come to muster in Clyde.

Adam Sroka (47:52)
I love it. Amazing. Well, look, this has been a pleasure. I’m really excited to know about heat networks and yeah, you’ve had a really interesting career and thanks for joining us here. Before we let anyone leave, we always like to just check if there was anything you would like to plug or promote or if people in the audience want to find out more about you or Gren where should we point them?

Tunde Olaoye (48:15)
So if you search for energy on Clyde, you’ll go to the Gren website. I think it’s a definitely good place to start there and then get into the Gren universe from that point. I think we’re doing some interesting stuff. We’re doing a couple of events in the next few weeks trying to get into Glasgow. So definitely point them to energy on Clyde and gren.com to find out more about us. And then if you’re in Glasgow and you want some heat, reach out to us and tell you when we’re going to be in your area.

Adam Sroka (48:39)
Well, look, it’s been an absolute pleasure. Thank you again for joining us today and yeah, all the best.

Tunde Olaoye (48:44)
Alright, thanks a lot for having me.

Adam Sroka (48:49)
As always, it’s been a pleasure and thank you for listening to the Hypercube podcast. If you want to find out more about what we’re up to or meet other people in the space, join the Beyond Energy Community where data and energy professionals from across the globe collaborate on what’s coming next across the industry