 Well, hello everyone. How are you this afternoon, morning or evening, depending on where you're joining us? Welcome to Engineering for Change or E4C for Short. Today we're very pleased to bring you this month's installment of E4C's awkward energy webinar series, focusing on the design of awkward systems, in particular on load and resourcing. My name is Yana Aranda and I'm the president here at Engineering for Change and I'll be your moderator for today's webinar. The webinar you're participating in today will be archived on our webinar's page and E4C's YouTube channel. Both URLs for those channels are listed on this slide. Information on upcoming webinars is available on our webinar's page. E4C members will receive invitations to upcoming webinars directly. If you have any questions, comments and recommendations for future topics and speakers, please feel free to contact the E4C webinar series team at webinars at engineeringforchange.org. If you're following us on Twitter today, I'd like to invite you to join us in conversation with our dedicated hashtag, hashtag E4C webinars. Now, before we move on to our presenter, I'd like to tell you a bit about Engineering for Change. E4C is a knowledge organization in a global community of more than one million engineers, designers, development practitioners and social scientists who are leveraging technology to solve quality of life challenges faced by underserved communities. Some of those challenges include access to clean water and sanitation, sustainable energy solutions, improved agriculture and more. We invite you to become a member. E4C membership is free and provides access to news and solid leadership insights on hundreds of essential technologies in our solutions library, professional development resources and current opportunities such as jobs funding calls fellowships and more. The E4C members also enjoy a unique user experience based on their site behavior and engagement. Essentially, the more you interact with the E4C site, the better we will be able to serve you resources aligned to your interests. For more, please see our website and sign up. Today's webinar is the home looks the last, nearly last one in the Off-Grid Energy webinar series. Additional topics covered in the series are drawn from the book titled Battery Fundamentals for Off-Grid Electrification authored by our presenter, Dr. Henry Louie. The final webinar in the series is actually coming up next month and the previous webinars are all listed on this slide. We encourage you to check out those recordings if you didn't have a chance to participate in previous webinars and for those of you who are looking forward to part two of the design of off-grid systems that you'll receive information directly in your mailbox if you're an E4C member or you can go ahead and click on the link to register on the site. For reference, you can find examples of off-grid energy products alike of the multi-cell solar home system in the E4C solutions library. There you can learn about technical performance, compliance with standards, academic research and use of provision models of these systems. All the information is sourced by E4C's research fellows and reviewed by our community of experts and the solutions library is available to E4C members free of charge, so definitely check it out. Now a few housekeeping items before we get started. We'd love to practice using WebEx with you and the first way you do this is to tell us where you are in the world. So in the chat window, which is located at the bottom right of your screen, please type your location. If the chat is not open on your screen, try clicking the chat icon at the bottom of the screen in the middle of the slide. You can use this window to share your marks during the webinar and if you have a technical question just send a private chat to the Engineering for Change admin and I'll also add my own answer here. I see some of you are answering in the Q&A so I do encourage you to please use the chat window as we are dedicating the Q&A window for other things but I do welcome you. It's great to see folks from Nashville, from Milwaukee, from Ohio, from Germany, from the UAE, from Morocco. We are really thrilled to have you here. Now during the webinar we'd like you to use the Q&A window which is located below the chat to type in your questions for the presenter. Again if you don't see it click the Q&A icon at the bottom of the screen in the middle of the slides. So please note we will aim to stop today's webinar 15 minutes before the hour or at 1145 Eastern Standard. I'll make sure to set aside enough time for Q&A. If you are listening to the audio broadcast and you encounter any trouble try hitting stop and then start. You may also want to try opening up WebEx in a different browser. All right so I see more folks have answered here. We have a participant from Denver, from Savannah, from the Philippines. Welcome one and all. We are really thrilled to have you and I'm glad everybody is getting a hang of the difference here between the Q&A and the chat. There we go and you should be also seeing those answers for those of you who are now accessing the chat from all of our wonderful participants around the world. So E4C webinars qualify engineers for one professional development hour. To request your pdh please follow the instructions on the top of the E4C professional development page after the presentation or you can also go to your member dashboard to get that directly. You will also have a record there of all the webinars that you've attended to date. Well fantastic. Now I'd like to take a moment to introduce our presenter. Dr. Henry Louis is an associate professor and Fort Francis would undoubtedly research chair in the department of electrical and computer engineering at Seattle University. His research areas include electricity access and developing communities, renewable energy and appropriate technology. He's the president and co-founder of Kilowatts for Humanity, a non-profit organization providing electricity access and business opportunities in Sub-Saharan Africa. Dr. Louis served as a Fulbright scholar to Copper Belt University in Kittway Zambia. He is recognized as a distinguished lecturer of the IEEE and as an associate editor of the Journal of Energy for Sustainable Development. He's the author of the book that he'll be speaking about today which is published by Springer Nature and we are very thrilled to have him meeting our webinar series. So with that I'm going to turn it over to Dr. Louis. Take us away. All right thank you and let me just make sure that I can advance the slides on my end. Did you see that advance? Not yet, not yet. So let me see if I can pass me the ball. Give me one second. Yep, I'm doing it right now. There you go. Let's try it again. All right, how about that? Thank you. Thank you so much. It's great to be back here for my fifth webinar with Engineering for Change and today we're going to talk about off-grid system design and this is going to be the first part in a two-part webinar series. The next one is going to be on March 6th. So today we're going to talk about load and resource estimation and the reason why we care about estimating the load and energy resources of a potential site for an off-grid system is that these two pieces of information are critical to the input to our design process. So if we want to get our system appropriately designed we really need to know the load as well as the energy resource potential. So like the other webinars, today's webinar is based off the book that I wrote called Off-grid Electrical Systems in Developing Countries. It's a textbook designed for engineering students and practitioners and you can access it through a number of ways through my website. It's available on Amazon and also through the publisher Springer and today's webinar is going to follow chapter 11 which is load and resource estimation and if you want additional details or problem sets, examples, and so forth you should check out the book and follow along in chapter 11. So today what we're going to cover is how we characterize the consumption, the electrical consumption of off-grid users and we're going to focus primarily on the mini-grid solutions to rural electricity access. We're also going to describe approaches to estimating load and the energy potential of different types of power sources like wind, solar, hydro, and so on. So let me just provide a brief overview of the design process for off-grid systems. We begin with a few inputs and most important are the inputs are the estimations of the load and the energy resource. These then feed into one of several design approaches that we might take which we'll discuss in next month's webinar and the output then is going to be the ratings and specifications of the major components like batteries inverters and other components depending on the type of off-grid system that we choose. So today we're really going to focus then on the inputs to the design process. Now when we design an off-grid system what we're trying to do is we're going to try to manage the trade-off between reliability and cost. So we can design a system that is really really reliable but unfortunately it's going to be really expensive and I think the engineer and many of us like to think well that's not bad to have a reliable system but we maybe spent more money on that system than we needed to and instead of impacting one community maybe we could have impacted two or three for the same cost. On the other hand if we if we drive towards extreme affordability then perhaps the system doesn't meet the needs of our users and it's not reliable. The key to striking this balance is to manage the input energy and the output energy and trying to make sure that we have enough input energy to to match the load and of course whatever losses might have. So today we're going to focus on on the output energy to start with and then we'll get into the input energy at the end of the webinar. So let's imagine that we are trying to serve a small community with an off-grid system so a mini-grid for example. We have to ask ourselves what characteristics of that load do we need to know in order to design a system and I think first we need to know how much load and we conceptualize the load in two ways. One is the energy required and we usually think of that as the daily energy as well as the peak because several of our components in our system need to be designed around that peak power consumption. So let's first talk about the average daily load. So the average daily load is important because this is typically what we base our financial models on in off-grid systems. So average daily load we're talking about energy and energy is what we bill our customers on. So it's important to find customers then that have a lot of consumption otherwise our grid won't be economically viable. Now we found in research that the average daily load really depends on the customer type. So an industrial user is going to use more than a commercial user and a commercial user is going to use more than a household but there's a lot of variation within those those classes and overall what we found is that there's this long tail this distribution of average daily consumption has a long tail meaning that some users are going to use a lot but most users are going to use very little. Most users might use somewhere between 20 or maybe 30 watt hours a day and it's really really hard to recover your costs at that low consumption. So the table on the right is just an example of how this might play out for a system connected to a mixture of households and and businesses. So you can see that a small fraction less than 1%, 0.5% of the users are going to consume about as much energy as 75% of the users those low consuming households and so it's critical to find the users that are going to have high consumption. These are the users that are really going to make your grid economically viable. In a way they're subsidizing all those other users with low consumption. So if you're designing a grid you should be prepared for that. Most practitioners now are adopting the approach where they're looking for for what they call anchor loads. So usually these are businesses or industrial facilities that they know are going to use a lot of energy and once they've identified and agreed to connect these users they then will branch out and connect other houses who may or may not consume a lot of energy. Now we talked about the peak load being important and so the peak is just that maximum power that we expect any given user to use and this is again important because when we design our inverter and generators and so forth we need to know the peak value to plan around. Now we also might we also will need to know when that that energy is consumed and when the peak is in order to serve our our customers. And so one way of expressing the timing of the consumption is to through something called the load profile. So the load profile is nothing more than the average consumption across across a day and from this you can get an idea of the trends in consumption and the timing is actually really important especially for systems that might rely on on solar or wind because we really want the consumption to to match or be coincident with that energy production. So unfortunately most households that connect to mini-grids their consumption peaks at night. They use very little during the day and it's only after the sunset that they start turning on lights and turning on their television. So obviously this doesn't correlate with solar production which means we have to use larger batteries. Now on the screen you see load profiles from several different types of customers and you can see that they can be quite different. So like I said before households tend to peak at night, some entertainment centers like a video hall, they might also peak after the sun sets. But restaurants and hair salons they might peak at the different types or excuse me different times of the day. Now that peak is actually quite important and in power engineering communities we describe the peak often as the load factor. So the load factor gives you an idea as to how peaked the load profile is. The load factor is simply the ratio of the average power to the peak load. And so up at the top you see an example of a user with a low load factor and you can see it's got a sharp peak. And we don't like customers with low load factors because what it means is we have to build our system to supply that peak and for example the inverter will need to be sized to provide that peak but overall that inverter's capacity is really underused for most hours of the day. So maybe 20 or more hours a day it's really really not being used too much and it's only an hour or two a day where it's anywhere close to its maximum. At the bottom we have a profile with a high load factor and so here it's much less peaked. We still have a peak but it's much much lower and so we can provide this customer a lower power service and still provide them the same energy over the course of the day. Now we're also interested not only in what the load is on on any given day but how it might vary over time on both short scales, short time scales, and longer time scales. And I'll tell you why this is important. Here are two users user A and user B with the same average consumption. User A is actually going to be easier to supply in terms of reliability than user B and the reason why is that user user A's daily usage is much more consistent than that of user B and really reliability isn't so much dictated by the average. It's dictated by the extreme events. Perhaps the user has a few days a year where they consume two or three times the amount of their average. Those are the days where you're worried about from a reliability standpoint. So we need to understand not only the average but the distribution, the probability that they're going to be using two or three times or even more than their average in order to have a good understanding of what the reliability requirements of our system of our system is. In addition to the day-to-day variation we need to know what the long-term growth might be. Now it's commonly reported for many off-grid systems that we'll see a growth of maybe between five or ten percent a year but this is definitely not guaranteed and it's not consistent amongst customers and it's not consistent across all mini-grids but there usually is some growth that will occur and in your design you should anticipate it to the best of your ability. Now typically growth is a good thing and we want to encourage demand for our electricity but a barrier that exists is access to higher power appliances. It might just be untenable for our users to purchase a refrigerator or or another television set because they don't have access to credit or financing. So many off-grid system implementers what they do is they have to have some sort of program that makes access to these appliances easier. They might offer subsidies, they might offer financing, they might lease appliances but you really do need to stimulate that growth after a certain level in order to make it happen. And then I think the last characteristic that's important to understand to design our system is how the load is related between one user and the next, how coincidence they are or how correlated they are and let me show you why this is important. Let's say that we have an estimate of each of these users peak load so the maximum power that they're going to do and it's shown on the screen now. When we design our system we need to know the aggregate peak and so one approach to this would be to simply sum the individual peaks and if we did that in this scenario we'd end up with a peak of a thousand watts. So do we need to design our system to supply a peak of a thousand watts? Well naively you'd say yes we do because that's that's sort of the worst case scenario but in practice what are the chances that all those users are going to be consuming their peak load at the same moment and most likely they won't be and so there's a way of we do this in power engineering all the time in the United States and Europe and other places we don't design around that worst case scenario. If everybody in your neighborhood turned on every appliance all at the same time the power grid become crashing to its knees it's not designed to do that. In other words it expects there to be some diversity in consumption and the way we express this is something called a coincidence factor or a diversity factor and it's basically a number that shows how coincident the load is and we generally want lower values of coincidence because it means that the peaks aren't occurring at the same time. So to give you a graphical example of what this looks like here are the same five homes with their peak load shown in the load profile now and if we add them all up we get the load profile shown on the right now it still has a peak but the peak is 780 watts it's not 1,000 so if we were to design around the thousand watt requirement we will have resulted in an overbuilt system and perhaps we wasted money so of course every scenario is going to be different but the goal here is to connect diverse customers customers whose peaks are going to be different whose usage patterns patterns are different and in doing so the aggregate peak is going to decrease the load factor is going to increase it's going to be less peak than if they were all correlated and then the day-to-day variation is also going to be lower because what we're going to find is that when one user maybe they have some family over and their their television and lights are on more often there'll be another user connected to our grid and maybe they've gone out of town and so their consumption drops to zero so the more users we connect really the better the more diverse users we connect the better so to summarize this what we're looking for then is users with high average daily consumption after all that's what's going to pay our bills as a mini-grid operator we want their load profile to be coincident with production this is especially important in systems like wind or solar where we might have a a peak of production during the day or some other time we want our load to really match up with that we want a load factor near one which means it's not peak that lets us have smaller rated inverters and generators and other equipment we want the variation from one day to the next to be as low as possible because it's the extreme events that that threaten our reliability we want there to be long-term growth but we want it to be predictable and we want a low coincidence factor so those are the characteristics that we're looking for ah but how do we know if a customer is going to have these characteristics well it's tricky and as an industry we don't have a great way of doing it but there are some approaches that are better than others and so i'm going to go over the state of the art here there's really four approaches that are being used which i'll talk about next the first approach is what we call a bottom-up approach and this is only really applicable if you have good knowledge of the equipment that is going to be connected in in a a given house or a school or something so you have an idea of the appliances and you know exactly how they're going to be used and in which case you just do the simple math of taking the power rating of each appliance multiplying it by the time it's going to be used and multiplying it by what's called a loading percentage which accounts for the fact that not all appliances consume the rated power when plugged in and you end up with the value of the the energy use per day so this is just an example of that calculation so if you knew that there were going to be five lights each rated at 11 watts used for four hours a day and because lights use the same amount of power when they're turned on consistently the loading percentage is just a hundred you get an average daily use of energy of 220 watt hours if you knew that it was going to be a refrigerator connected you would repeat the process the loading percentage is probably going to be less than 100 it's going to be maybe closer to 10 because the compressor is likely not running 24 hours a day and you'd get a 480 watt hour total so this user then would would have a total consumption of about 700 watt hours so this is the what's called the bottom-up approach and the big drawback to that is in most cases you're not going to know exactly what appliances are going to be used and how long they're going to be used for each day it just won't happen unless it's a special scenario like a like a school where you only know that they're going to be using lighting and and you're not providing other outlets or things to be plugged in so how do we improve upon or how do we extend the bottom-up estimation to the more general case well most of the time we would use surveys then so these are usually door-to-door surveys and they'll ask a number of questions and but really it boils down to two pieces of information you know when the grid gets here when we build our mini grid what appliances do you expect to own and how long will you use them for and based upon that information then we can develop an estimate of the average daily energy if we word our word our surveys cleverly then we can come up with an idea of the peak consumption and the load profile and some of the variations that we might need to better design our system now is bullet proof as this might seem it's actually very error prone research has shown that you know we can expect errors in excess of 300% most of the time people will overestimate their consumption but there are certainly times where people will underestimate it and it's easy to see why I mean most of these people that you're going to connect maybe they've never owned an appliance in their life and and so they're speculating on what they're going to use and when they're going to acquire those appliances they don't know how they're going to use it and in particular if they're paying for the energy they might not have a good idea of of what they can afford and then there's also on it from a technology standpoint you know if you look at the power rating of an appliance it's not constant and the actual consumption is going to vary so it's hard to do and then finally survey or bias it's incredibly difficult to design a survey and implement a survey that doesn't introduce some bias even if you you know it's done in the local language and you follow the best practices you're going to have some bias in there and some people might might bias it towards overconsumption because they think if they tell you that they're going to use a lot of energy that you're more likely to install the mini grid and more likely to connect them on the other hand if they some some people might try to underestimate their their consumption thinking that somehow this is going to affect the rate that they have to pay so if they feel that if they convince you that their consumption is low that they're going to pay very little so there's a lot of bias all over the place in doing surveys so if you want to get away from the bias there's a few other approaches that you can take one is to try to do a regression analysis where you look at maybe census data or demographic data and you try to correlate it to the consumption so there's been a number of variables that people have tried to relate to consumption and and this is a method that works in some some areas but I don't think I've seen a one that's been universally applied with high success so I'll direct you to references four and five for more information on those now the approach that the industry is moving towards is this data driven approach so as we put more and more mini grids out there and we monitor the consumption of the users we create an archive this historical database of consumption so now when we want to do a new mini grid we can just look for similar mini grids that have been implemented similar in the sense of the location the the type of customer and so forth and then we look at that data to inform our our our load estimates and this is a way of determining the peak the average daily use the variability it's really all there the problem of course is getting access to that data there's there's not a lot of data sharing that goes on in the industry and I think this is a real barrier to to you know greater rollouts of mini grids and so basically this is just a picture of how that works you look at existing mini grids you come up with an average per person consumption and you apply that to the future mini grids so that's an overview of the the output energy how we estimate it and the characteristics that we're most interested in and now we're going to go into the input energy so let's look at that same scenario and now we have a couple of options of how we might supply that that that load we might use wind or solar or biomass or something so to know which one to choose at least from an energy standpoint we need to know how strong the source is for example the wind speed or solar we also want to know how it varies or how consistent it is and then finally we want to know how how its availability changes on the short and the long term so one way of doing this is to use the capacity factor and the capacity factor is a commonly used metric that lets us in a single number express how a potential energy resource is utilized so it's just the energy that that's produced in this case estimated because we haven't installed our grid yet divided by the energy that would be produced if that that generator or that wind turbine were producing rated power continuously so as a quick example here let's say we have a one KW solar array and we monitor its production over the course of a day and we get 4.8 kilowatt hours so we see the the power increase in the morning and then decrease in the evening as shown there well the capacity factor then is simply the energy produced divided by the the rating times 24 hours because we're looking at a 24 hour period and we get a capacity factor then of about 0.20 or 20% so you can imagine if we put the same PV array in an area that has more sun that our capacity factor would be a greater value it might be 0.25 for example so if that's the case then we can easily compare those two locations the energy or the solar resource of those two locations based simply on the the capacity factor another way of thinking of that is if we have a capacity factor 0.2 as in the previous example we can replace that the PV array with a hypothetical generator rated at 200 watts that operates continuously at least in theory I mean this is just a theoretical explanation of the capacity factor now not surprisingly different resources are going to have different capacity factors so gen sets they're going to have a high capacity factor because you can more or less run those at rated power continuously you know only having to shut them down every so often for maintenance a PV arrays it's going to be lower because the sun is only shining for at you know really at most like half a day on average and and then even then it's not shining uh fully during the morning in the evening and so the right column then is just a way of of comparing those in terms of capacity so this is the size of the the capacity of the different source to provide one kilowatt uh one kilowatt hour per hour for the course of the day so that you see that you know a one kW gen set is not equal to a one kW PV array you need a much larger PV array to supply the same amount of energy as a gen set in fact you might need a a PV array that is between four and seven kilowatts to equal a gen set that's that's equal to one kilowatt so the capacity factor then is a useful way of expressing the utilization of a source now we're also interested in this variability that our different energy sources are going to have and the different sources are going to vary across different timescales and so when we think about designing our system we really need to know that variability if there's a certain time of the year where the wind doesn't blow we absolutely need to know that if we're going to consider a wind energy conversion system so we have a we have to have a sense of the variability and the variability shows up in many different ways for different types of resources so the point here is that we really can't rely on one capacity factor calculation for each resource we need to rely on several and maybe we look at it across a month or a season or even multiple years to get a sense of the true production of that so then we come up with a capacity factor table we calculate the capacity factor for each month or each season and and that gives us a better picture of the production so this is an example for a wind turbine and it's going to be in this example it's more windy in the spring and the autumn than it is in the summer and winter and so what we're curious or what we're most interested in is the capacity factor of the lowest month and so in this case it's January it's a capacity factor point one four so here if we were going to design our system we would have to consider January and say well if we want to reliably serve our load in January we need to look at the capacity factor point one four and that'll tell us how many wind turbines we do need to to install and I'll talk more about that calculation in the next seminar so I'll briefly cover the the different resources and how we come up with the data needed to design for solar most of the time you can consult a solar map and an online database and you can get an idea of of the irradiance and insulation and the irradiance is going to change throughout the year it's going to depend upon several factors like your latitude and your tilt but the capacity factor is is nothing more than the the insulation divided by 24 and the insulation itself is just the area under the irradiance curve and so typically what we're going to find is we're going to find that the insulation is going to vary between maybe four to seven kilowatt hours per meter squared for sub-Saharan Africa and so we're going to get capacity factors that are somewhere around you know 0.15 to you know maybe 0.25 depending on the time of year there's lots of factors that affect the capacity factor here here's the effect of latitude on a horizontal surface the closer to the equator you are the more consistent the generation the further away the more variability that that you see so Cape Town is going to have very very low production during their winter and high production during the summer we can balance this out to some extent by tilting our PV array which is which is commonly done as shown here this is from Nigeria. Getting into wind there are wind resource maps available but most wind resource maps that you'll find are at a hub height for utility scale wind which is not the hub height for for small scale so they might do 50 meters where for small scale you might be looking at the wind speed just 10 meters above the ground and in addition you're not going to find a wind map that has the resolution needed for for small scale wind so then you have to rely on measurements and so you have to put up a small MET tower with an anemometer as shown there and then you measure the speed for one to even two years to really get an idea of the distribution of wind speeds like that shown on the right. There are a few ways of filling in the gaps in the data and I'll refer you to reference six for more information about how you do that but once you have an idea of the measured wind speed you can convert it to the power output by looking at the power curve of the wind turbine that you're considering in the middle there they're all going to look a little different but you just map each wind speed and it's a probability of occurring through the power curve to come up and you get a distribution then of the power production and from that then we can calculate the capacity factor because we have our estimated production now and you're going to see that's going to vary of course across the year depending on depending on the wind speed so if you look at the right there you get a sense of the capacity factor and how it how it varies. In terms of hydro we're interested in two pieces of information ahead and the flow rate and these are you're going to get wet when you try to measure the flow rate there's several different ways of doing it and I'll refer you to seven references seven and eight for for more details on how you how you do these these measurements. When we think about the production of of energy from a hydro resource what we do is we look at that flow rate usually in liters per second and then we will reserve some of that for the stream so there's some minimum flow rate in our stream and and the table shown on the screen there we've picked a minimum flow rate. So that leaves you then with the value which is your maximum micro hydropower flow rate and from that you will look at different turbines and their rated flow rate and you can calculate a capacity factor of each. So in this example if we decide that to use a very very large turbine with a flow rate of 185 liters per second you know its capacity factor is going to be really low during those dry months and and so it's not going to be utilized very well there. If you pick a smaller turbine it's going to be operated much more consistently but of course its power output is going to be much lower. So these are that's really important information then when you think about your hydra system do you do you want to have it oversized but underutilized or do you want to have it smaller but utilized to its capacity more frequently. And then finally we'll wrap up here with with biomass so if you're planning a biomass system what you really need to know is the energy available in your your feedstock because that will tell you the energy input to your your reactor or your digester and based upon that efficiency you'll get an idea of the energy that you can put into your gen set. Knowing the gen set's efficiency you can calculate output energy. So there are some databases that are available that that describe a particular region's crops and crop yields or you can rely on local local surveying but basically you look at the crop yield and if you're for example doing a gasification project it's really the the residue of the crop that you want. So you can't just use a crop yield you have to multiply it by a factor which is the residue to crop ratio. So how much how much residue do you get for each kilogram of crop and then to figure out how much of that is available in terms of mass you look at the the residue you multiply it by your estimate of your collection efficiency if the community is using some of that residue for other sources you need to subtract that off and then that gives you the mass of residue that you have so you multiply it by specific energy to convert it to megajoules and then you multiply it by the two efficiencies of the process that to get the energy output and then you can calculate the capacity factor as shown there. So to wrap it up here you know depending on the source that you are considering and most often you're going to consider several resources the data requirements are somewhat different and you really need to sample more and for a longer time if you expect the underlying resource to change so it's not good enough to just take an anemometer to measure wind speed and put it up for a day a week or a month because you might have picked a month where it's the most least windy and and that bias will drastically affect your your design. Similarly with hydro you don't want to measure the flow rate when it's in the rainy season and you don't want to measure it alone when it's in the dry season you really need to get a sense of it throughout the course of the year. So I've described today then you know the characteristics of load that we're looking for and attempt to measure it as well as the energy resources. Now both of these are really nasty and there's a lot of error that can be introduced but if you're not diligent if you just make up numbers or use you know averages you're probably going to end up with a system that that doesn't doesn't perform the way you expect it to. So with that I'll plug for next webinar where we're going to take these input pieces of information and we're going to come up with how to design off-grid systems. So thank you so much for for listening in here is the the list of references and I'm happy to take questions. All right thank you so much Henry for giving us a sense of what it means to write size system. So with that I'd like to open up the floor to questions. We did have one come in already so I'm just going to throw that out here and that has to do with estimating load percentages of appliances and how you go about that for greatest accuracy mainly for things like fridges, longrees, ovens and whatnot. Yeah this is challenging there are a few resources that are out there and for example the United States government they do I think I think through the EIA has information on the loading percent of different appliances but I'll warn you that that's that's the appliances we have in the United States are going to be different than you're probably going to encounter in the rural area. So unfortunately there is not a authoritative source for this particular application. If you're a researcher listening in this this is probably an area where we could we could all benefit from having a better understanding of that loading percentage. If you don't you know you can always test this you could you could purchase the the appliance yourself and you could do some measurements. Alternatively what you can do is some manufacturers will provide the yearly consumption of larger compliance larger appliances like refrigerators or televisions and given that information and the power rating you can calculate the loading percentage but those annual consumption values that are given are based on some set of assumptions like there's a the appliance is operating in the controlled temperature environment and that's not going to really be the case in most off-grid users home so the appliances the refrigerator might have to work much harder than in the the test case that the manufacturer used. So there's not a authoritative research or resource that I know that collects all this information there's bits and pieces gathered throughout research and and larger organizations might might have have lists but not necessarily tailored to the off-grid situation. Right and on the note of data resources you spoke about you know the use of data driven kind of approaches in order to understand the historical consumption of similar many grids is there are there some kind of definitive or trusted resources that you can point us to a database of that information that you'd like to share? Yeah so there are a few few efforts to do this provide that data if if there if you can do a search for the IEEE PES working group on sustainable energy solutions for developing communities there is a data that's really long I'm sorry there is a there there is a data archive there and and maybe if it's easier I can provide a link at the start of the next webinar. Yeah that's a good idea. So it does exist it's small but I'm actively involved in that group and we're trying to grow it so if you if you're listening and you do have data sets of load and and any other relevant data we'd be happy to include it and curate it. More and more you're seeing published research that they're they're providing data sets and and you know so you can consult that and look use that data but there's not there's not a definitive source and again there's probably lots of data that scattered around the internet I believe MIT had a repository that they were building so they do exist. Yeah and I think that's that's a really great call to action is to recognize the fact that there is a need for a central repository. So there's a number of questions that's coming we're going to try to tackle them all just as a reminder please enter your questions into the Q&A window so we can go ahead and keep track. So one question is regarding the complexity of applying these methods to a hybrid production system so for example something that marries solar with wind and hydro how would how would you approach that? Yeah so in terms of the the resource estimation side you treat them entirely independently so if you're going to do a hybrid wind solar and I've done hybrid wind solar and and in terms of estimating the resource you do them separately it's when you marry them into a design that that the correlations between the two of them come into play we didn't talk about that today we'll talk about that in the next webinar but when you have a hybrid hybrid solution or a hybrid system that you're going to design you really need to rely on on computer simulation to to really get an idea of how those two sources interact with each other but I will say that if you're going to do a hybrid system you know to the extent that you can you want to measure the the you want to take your measurements at at the same time you don't want to have you know hydro flow water flow measured in September and wind measured in in July you know you'd want to try to have simultaneous data sets yeah. And how about the frequency of power synchronization for power from different types of generation like DC with solar AC from solar hydro and so forth? Yeah okay great question um this is we didn't cover that in this it's in chapter 10 of the book but anytime you take different AC sources even of the same type so even two AC generators and you want to connect them to the same AC bus they need to be synchronized and and some generators will be able to do that automatically and some will require an external control system so using the example of gen sets which are the most common common used AC generator they will they they will often operate on what's called droop control so as the load increases the frequency that they produce power will decrease and they'll naturally synchronize the frequency. When you connect them in the first place you do need to make sure that the voltage waveforms are synchronized with each other and there's usually a piece of equipment that will let you call the synchroscope that will let you know if they are in sync with each other but this is something that you absolutely do not want to overlook if you have an inverter and you want to connect it to another inverter or a generator on the same AC bus that inverter better be capable of synchronizing with with other AC sources so this is a very critical control piece of control that you need to have in your system if you plan on doing that and there's a question came in I think that is actually going to be a good plug for previous webinars so could you say a bit more about the assessment required to determine cost effectiveness of an off-grid flash mini-grid development versus grid extension what's the time horizon for such a calculation and what data is required to incorporate the correct assumption yeah so this is the topic of the the third webinar that we did we fly everywhere yeah so the the short answer is that as you get further away from the existing grid off-grid systems become more attractive be it a solar home system be it a mini-grid whatever as you get closer to the grid then grid extension makes more sense to understand which which is the preferred method you need to have a lot of information about the cost of grid extension you need to know what the cost of the transmission lines are the distribution lines the the cost to connect each customer so you need a lot of economic information and then you need to make some assumptions about the how long that grid extension project is going to last and the energy loss is along the line and the cost of the energy then you do a calculation that that shows you the the cost per kilowatt hour of that grid extension and then you would compare that to the the cost per kilowatt hour of your mini-grid using something called a levelized cost of energy so that's the the very short answer to a very complicated question i get into you know more details including an example in the the third webinar that we we did in the series or maybe it was a second for every second for everybody's benefit i've gone ahead and added the url to that webinar in the chat window so it's appropriately titled why the power grid isn't everywhere and the whole little grid extension unless you see access so do take a listen to that recording and it will give you a more in-depth answer to your question and there's lots of papers out there that that analyze that yeah so you mentioned for solar energy the diversity factor or was it the coincidence factor is there a database in which the values for the factor is shown for different situations or cases well uh yes for uh for okay so yes and no so yes utility utilities in the united states and europe by asia and parts of africa they're going to have that information and you can find it in in books i i actually have that a plot and a reference in my book that you can look it up but again that's that's not for the off-grid context the plot that i've shown in this webinar is is an estimate based upon about 200 customers of what that coincidence factor is and i don't intend that to be authoritative the difference though that we see between the off-grid case and the on-grid or excuse me the mini-grid case and in developing communities versus you know my neighborhood here in seattle is that um the the coincidence factor is is much higher in the off-grid case because most people in that off-grid community you know they have the same appliances they have a couple of lights and the television and a radio and they're all going to use them at the same time you know when they get home in the sunset uh if you know in in large cities it's going to be completely different you know people there's much more diversity and when people come home from work or they might be going out and you know at a restaurant and and so forth so there's much more diversity in in my situation than when i lived in a rural community for example and again another another ripe area for research yeah another ripe area for a lot of exciting opportunities for those of you who are a graduate student so you didn't undergrads looking for a senior project i think henry you are just dishing them out today very exciting um so relate to that uh in and in terms of estimations and so forth in cases where the primary source of electricity is a diesel generator can diesel fuel consumption be one indicator to derive the kilowatt hours per day in certain cases perhaps when compared with load-based data yeah absolutely and i know of at least one operator that did that so um maybe they're the one that asking the question so in this in this strategy what you do is is and this would only really make sense for a much larger mini grid um you would say well i i don't know what the load is so i'm going to temporarily install some some generators and i'm just going to for you know a years worth of time power power my customers with that that generator and and during that one year i'm going to collect data on how they've actually consumed it and so once i have that data i'm going to devise a solar powered system or maybe a hybrid system that is based upon that known load profile and then you move the generators away or use them for a different purpose so that is in fact another very clever approach to doing it but you know it is another level of sophistication that's involved in in arranging that logistically so it probably only makes sense for a much larger mini-grid and i think this is going to be our final question and this question is almost seeking some case studies but in this instance this listener says first of all most amazing presentation i couldn't agree more does a typical system intersect between all resources you mentioned this listeners resident country of Nigeria what difficulties do you face implementing such systems in Nigeria so perhaps some examples relative to Nigeria specifically that you can share if they come to mind ah so i haven't personally done systems in Nigeria there are quite a few operators there that i know of that that have installed it i mean as much as we don't like to to lump the 50 some countries in Africa it's just one you know one region uh i think there are in fact a lot of commonalities and the challenges faced in terms of electricity access in rural areas of one country to the next i mean i you've run into problems like uh people's inability to pay so i talked about how a surprisingly large amount of people connected to mini-grids use so little i mean 20 or 30 maybe 40 watt hours a day which which is so so little it's far below what any estimate would would put their ability to pay it's it's far below what certainly what their need is for electricity and and it's it's confounding i mean it's really challenging that to figure out what we should do how should we approach those those users that's one big challenge i think as as we see more and more mini-grids that are installed will have a better sense of how to manage that challenge but then we we enter new challenges you know the regulations of mini-grids the subsidies for mini-grids these these are all things that uh really need to be resolved and i don't think any of any of these countries with with very low electricity access rates have haven't figured out yet yeah lots of work to do um so it's exciting times to be in this particular field so unless item is more of a comment but you can also certainly answer it as a question so uh one listener is curious if you're in touch with any african university universities to develop this into an organized university course of study um they think that this might be more relevant than say merely studying electrical engineering so any thoughts on that yeah for sure i mean this is this is a course i teach a course at my university in seattle on off-grid electrical systems particularly in the developing community context the book that this webinar and the other ones are based off has been adopted by a few universities i know for example that carney geen melin has a rewanda kigali campus and they're going to be using the book there for for one of their short courses so i'd love to partner with their universities there now i do have a lot of content that that related to the material and uh i agree this is you know african engineers but also engineers in europe asia united states this is all important information that i mean from a strictly business standpoint it's it's i mean that's where a lot of growth is going to be you know if we want to build power plants and transmission lines and heavy infrastructure you really need to understand that the context that you're getting into and i think in many universities we don't we don't give our students that knowledge and we are in violent agreement with you on that point henry and for those of you who are listening once the series is complete we do encourage you to use the series as a resource do share it with your professors see if they will be interested in integrating it into the curriculum because it is one way to get the information now ahead of it let's say becoming an official course or short course so with that i'd like to echo the sentiments of many of our listeners and say thank you so very much henry for taking the time as always we've gone over time even though we always intend to be short about this but i was nearly a hundred attendees today that that's bound to happen and i'd like to thank all of you for tuning in listening to us don't forget that we have one more installment in the series that's going to happen on March 6th for those of you who are interested in pdh's the link is on the screen to be able to get that pdh and if you have any more questions or anything that we didn't get to cover do send us an email at webinars at engineeringforchange.org so with that i'd like to wish you all a good morning good evening or a good afternoon wherever you may be i encourage you to become e-perceive members to get information on the upcoming webinars and thank you all again bye bye