 I'm going to talk a bit about heat pumps and heat pump monitoring. I recently got a heat pump installed at home. For anyone who doesn't know what heat pumps are, they're a heating system, bit like air conditioners, but running backwards, so you use electricity to basically move heat from outside into the into your house. So let's see if I can get this on full screen. Okay. I've worked on this with John Cantor, who is an expert on heat pumps. He's based in Mchentyrh, he's been working with this technology for probably several decades now, he's written books on it and is a consultant. bydd eich cyferwyr i'r dweud o'r cyflogu'r gael, ac mae wedi'i yn rhan o beth o mathを mwy nifer y bydd. Rydyn ni'n rhag o'r ddeunydd ond rydyn ni'n gydag o'r gysylltu eich meddwl o'r cyflogu'n rhan o fwy, a'r tyf yn оedd yn fwy o bobl sy'n cwyl o'r cyflogu cyflogu, followed byddur eich meddwl o'r cyflogu'n rhan o'r g Overview o'r ddysguion ynghyd. Rwy'n meddwl ynghylch gyda'r system a'r hyn o'r ddau am y cyd-dweithio gyda'r hynny'n cyd-dweithio'r hynny a'r cyd-dweithio'r cyd-dweithio... ...yna'r cyd-dweithio cyd-dweithio. Felly, rwy'n cael ei ddweithio'r hynny yw hynny, rwy'n cael ei ddweithio'r hynny. Rwy'n cilydd i hyfforddiadol yn ei ddweithio'r cyd-dweithio'r hynny yw David McKay. yn cael ei hyffin o gwahanol gyda'r cyfle i ddestunutu cyffinol i gyr polishing a'i'r hyffindwch i gydiannol ffordd i gyda ffosil. Mae Ysgwrdd Cymru yn gwneud y flying нихu yng Nghymru ac yw'n ddigon na ddym yn gwahanol. Mae'r ardalol yn meddylion iawn yn ymddangos â gweithio cyflogol, ac mae'n meddylion iawn arall o'r gweithio cyflogol. Rwy'n meddwl bod yn yn yn oed yn ymddangos, mae'n rhai gwneud yn ymddangos cyflogol yma. Rwy'n meddwl bod yn ymddangos i'r gweithio ar gweithio. Mae'r gwneud yn ymddangos, mae'n meddwl yn gweithio ar gweithio arno ymddangos am ymddangos, 10-year hourly energy model that's based on their original model, but it's not quite the same. And you can explore it in your browser. But essentially, if you look at space heating use in the UK now, it's about 34 kilowatt-hours per day per household. Zero Carbon Britain reduces this to 4.2 kilowatt-hours per day electrical input. They do this by you insulate houses up to a very high standard, and then you have a heat pump which uses electricity to produce the remaining heat. And overall, it's an 83% energy saving. I'm interested in the integration of all these technologies. How do you then power those heat pumps from wind energy or solar? How do you ensure that your variable renewable energy supply matches your demands? So this is where I live at the moment in North Wales. It's a small cottage. It's by no means a low energy building. But I had been heating it with a wood stove and electric heating, direct electric heating, and it was a pretty uncomfortable place to be in the winter. So in the long term, I'd like to live in a highly insulated building, and I'd love to build something. But at this point, I needed just to get a heating system so I could get through the next winter. So I put in this heat pump with John Cantor's help and set about monitoring it and really getting a good understanding for how it works. So I did some modelling beforehand to work out how much heating energy I'd need. I used a tool I'd been building with CarbonCorp in Manchester to model how much heat the building would need based on SAP. So I predicted I'd need about 6.7 kWh per day to keep it at the SAP design temperature of 21 degrees. I mean, I don't keep it at 21 degrees for the same amount of time, so I don't actually use this. This is a picture of the heat pump at the back. So it's a Mitsubishi air source eco-dan, 5 kW output. The system I've installed is quite simple. Essentially, because it's this old cottage, it's kind of like one main room. There's two radiators inside. One is a standard convector. The other is a smart-rad fan-assisted radiator. You've got the pipes coming in from the heat pump through the corner. There's a pump expansion vessel, a heat meter. That's pretty much the extent of the system, so it's quite a simple setup. Then I've added monitoring to this to be able to see what it's doing. I guess the headline figures, I've had it installed now from October 17th last year. So it's 113 days up to the point I finished this test. I've used 413 kWh of electricity input, and it's provided 1,405 kWh of heat output. So that's a COP of 3.4, which I'm pretty happy about, because there's lots of stories of heat pumps getting much worse results than this. This kind of validates that you can get it to work, and it's a relatively simple system, not perfect by any means, and it's still getting good results. At a COP of 3.4, if you use electricity from the grid with standard grid intensity, it is equivalent of 151 g of CO2 per kWh that he delivered into the room, which compares quite well with gas heating, which is around 230. So you're making a saving already just with standard grid electric. We've been monitoring the grid CO2 intensity over the last few months, and it's actually lower than this figure that's often quoted. At that, it's even lower, it's at 94. So it's potentially a large saving, even with the existing grid, which is not... I mean, there's an increase... We've got a lot more renewable energy than we did have a few years ago, but it's still not the larger share. So, to do this monitoring, I've been developing an open hardware heatpump monitor, which is this board in the corner here, and it has... It's an Arduino or an ATMEGA 328 in the centre. It's got an embus reader. It's got CT-based power monitoring, electricity monitoring, pulse counting, temperature sensing, and then it's got the ESP2866 Wi-Fi chip on there. So it's quite a versatile board. You could use it for other things. You could use it for monitoring the heat output from a gas boiler, but I guess we've developed it with heatpump monitoring in mind here. This is all open hardware, open source. It's all on GitHub on that link there. With this, I've been monitoring the electricity input in watts, total electricity input in kilowatt hours over the test period, heat output power in watts, total heat output outside temperature, room temperature, flow, return. Those are the main ones. This is an example of a kind of graffing we're able to do with that data. This is like one heating run. The blue here is the electrical input and the yellow is the heat output. These are the flow temperatures up here, flow and return. You can zoom in on a particular heating period and you can look at the COP that was measured. The COP is the coefficient of performance, which is the crucial thing. It's how many units of heat did you get out for every unit of electricity put in? In this particular example, I had 3.54 units of heat output for every unit of electricity I put in. You can compare it to what it should do theoretically with this Carnot efficiency equation. It was good to see that it matched up to what was expected. I guess that's when you do this kind of monitoring, that's always the thing that you're asking, is you have models or you have something that you've read and you want us to see that when you've actually put this technology and is it working as my understanding of it says it should work, especially if I spend a lot of time doing modelling and then I'm telling people about models, I want to know that if you do put the kit in, it works as I expected too. A couple of minutes. As far as I can see, only Ken and I are going to speak up. I don't think there's not much to it. There's not much more. I'll have five minutes. These are just some graphs of system properties, flow temperatures. These are all useful to understand how well the system is working. This is perhaps one of the more interesting aspects, which is the matching between real-time data from grid zero two intensity and real-time data from wind output and then look at what is the degree of matching between the supply and the heat pump demand over that period. I've recorded with this heat pump 35% supply demand matching, which is not particularly high. The modelling suggests that you should be able to achieve about 57%, but there's a lot of reasons for that. I won't go into it now, but this is a good base point for me to then improve on it over time and hopefully be able to demonstrate higher matching in the future. The next step is we make this like an open initiative for anyone who has heat pumps and they want to see this kind of data for their own houses, for their own systems. We want to try and explore a way of maybe having a league table where you can compare these kind of headline figures and see if there are certain configurations that will achieve higher performance, higher matching. And explore different control strategies or turning the heat pump on the heating hot water when the grid CO2 is low or when there's high wind output. That's the next step. Thanks for listening.