 I'm delighted to stand here to have the opportunity to share some thoughts on the subject of wind energy. This is going to be a very brief tutorial, as it were. It's not going to get in any technical depth in any of the aspects that I will talk about. I understand that there is an audience here with very diverse backgrounds. And so hopefully I will say things that will make sense to you. If you have questions about anything particular and so on, please come talk to me later on and we'll be happy to engage with you. All right, so I sit in these two departments, aeronautics and astronautics and mechanical engineering. And the topic of wind sort of fits nicely across these departments. Wind energy is about extracting energy from the wind. That involves both mechanical and aerodynamic aspects to the problem. All right, okay, let's move on. So what I will try to do today is have some very basic sort of concepts associated with wind energy. That's how we will start. And then we'll jump into some current issues and opportunities that's going to be the focus. And hopefully if you're interested in some of these, you will find people who are working on these topics and so on and engage with them. And so for that, I will try to give you a glimpse of some wind energy research as Stanford. I'll talk more about work that I'm involved in, but I'll also try to provide glimpses of work going on in other labs. And a look towards the future, all right? Okay, well, wind is a renewable energy source. It's driven by the power of the sun. The sun shines on the land, on the ocean and so on. Heats up air and so on. And there are slopes, a terrain and so on. Sort of different processes set up wind in the atmosphere. And wind is being driven, the energy for the wind is coming from the sun. And that's sort of a basic thing we have to keep in mind. And it's a clean source of energy. We don't need to use any fossil fuels or water or and so on in terms of environmental impact, the operation of wind power plants has very little environmental impact in direct fashion. Wind energy resources are distributed around the world. And if we focus on the United States, here the map at the bottom is indicating where the current, this is a map from 2015, where the various wind farms are. And the color coding is based on the capacity, the total capacity in the state. State of Texas has the largest developed wind power. California comes next. Iowa and others in other colors you can see. So you can see that it's spread out across the United States. And in terms of cost, the economics, costs have come down over the decades and so on. And now, depending on exactly how you do the calculation and those who are in business and economics majors and so on can do this better than I can. So it of course matters as to what the policies are in terms of subsidies, in terms of assistance and so on. But so the numbers might come out a little different depending on how you do them. But one can argue that the cost is now comparable to a cost of other traditional energy sources, which is really dramatic. This is not where things were 20 years ago. Install capacity today, well it's 2015 data, is 73,000 megawatts, this is significant and we'll understand these numbers more as we progress. Another key thing to keep in mind is that wind is intermittent, it's not blowing at some fixed velocity all the time, it's not. And so if we're gonna use, make significant use of wind energy in the overall energy that we are going to provide to people, then we have to cap, we have to factor this in because we need to be able to take peak loads but also balance them out and so on. And so this becomes challenging as the when the fraction of energy that is produced by wind becomes significant. And then of course where the high winds are might not be where the highest demand for it might be. And so you need efficient transmission of energy from where it's being generated to where it's gonna be consumed as something which is true of many sources. Okay, let's discuss some very basic operation. So there are two types of wind turbines depending on the orientation of the axis around which the blades of the turbine rotate. So if the axis is vertical as is indicated here then it's called vertical axis wind turbine and if the axis is horizontal and these blades only part of the blades you can see in this picture and note the scale of human standing there on top of this tower, these blades are gigantic. If we look at the size of these blades, they are larger, the tip to tip distance from one tip to the other tip, that diameter is bigger than the largest airplane. A380, which is the largest airplane you may have flown in A380 or in 747. These machines are huge now and they produce large amount of power. Seven megawatts is a machine from Siemens. They range from a few kilowatts that you can install in your backyard if you have some open space, two large capacity machines. So of course the crucial thing is the scale and how efficient is that installation. For an isolated wind turbine efficiency is very straightforward to define. You can take the ratio of the energy that the machine is extracting from the wind to that which is available. And for conventional wind turbines, there is a well-known way to think about this and it comes up with a limit which is called the Betz limit. It's sort of like 57% or something. Why can't it be more is something that we can easily appreciate by imagining what is happening. The wind that is going to be intercepted by these blades that is going to influence these blades, that wind which is coming in has to also leave. We cannot just capture all of the wind, all of the power associated with the wind. If we were to do that, if we were to capture that airflow, that amount of air somehow we'd have to store and we'll have to grow in the region where it's going and we just can't sustain that. So wind has to come in and has to leave. Because it has to leave, it has to carry a certain amount of energy with it. And therefore there is a maximum, there is a maximum efficiency of a wind turbine which can be predicted and for isolated turbine. And an interesting question is, how should one think about this when one doesn't have an isolated turbine? When you have a large-scale wind farm, you have turbines here and there, spread over tens of kilometers of distance, how should one think about the turbine of the efficiency of the collective farm rather than efficiency of an individual device? That's still an open question and it's one that we actively work on. Each turbine has a power curve here in the schematic power produced by the turbine as a function of the wind speed. There is a cut-in wind speed below which it's not economical to operate the device. So it's stationary, it's not rotating. Then starts producing power. Power rises rapidly as the wind speed rises essentially like cube of the wind speed. And then beyond a certain wind speed, the power is regulated to the maximum power and this is for purposes of both aerodynamic efficiency and for safety. You don't want the speed of the blades, the speed at which they are rotating to keep increasing and the loads on the blades to keep on increasing. Those loads are gonna go up as the speed of the wind is going up. And so since mechanical loads have to be well within the structure limits of the device, one cuts off the operation at some rated power and so although wind can increase, the power is then not increasing. This is essentially the situation which is called stall control where the orientation of the blade is changed so that the amount of power that it produces remains constant. All right, okay, let's continue on. So at present around the world, one finds many installations. I hope that the image is coming across here. It's not just one row of turbines, but there's another row of turbines, there's a third row of turbines, there's a fourth row of turbines. And on this particular day, a nature cooperated in terms of producing clouds that sort of gave rise to these regions of condensation of water vapor. And so we can, in this image, we can see the wake or the region of the wind which has been affected by this turbine. That region is flowing into another turbine. And so if we say just qualitatively, if we think about how much energy the wind has and if we extract some of that energy by a turbine, then downstream of that turbine, there'll be less wind, there'll be a smaller wind velocity, there'll be less energy in the wake. And if that wake is to impinge on another turbine, then this turbine is not gonna see as much wind and therefore the amount of power it can produce is gonna be significantly less. So if you have a situation where the wake of one row of turbines hits another row of turbines head on, then you have a very significant loss in the power that the collection can produce. So this is what is called a wake loss and this wake loss can be substantial at the farm scale. It can be as much as 30 to 40%. And so imagine that your $1 is now only going to buy you 60 cents worth of power and so on. So it's a very significant number and of course it depends on many factors. It depends on the spacing, it depends on the wind orientation and so on. And so this was not very well understood when some devices like this one farms, this Horns Rev farm off the coast of Denmark was built. The story is actually that it was designed with wind turbines of a smaller size and by the time the regulation, the process for permitting went through and they started building. At that time, the turbine technology had improved. A larger turbines became available and they were more efficient. So they thought, why not deploy these larger turbines? They are more efficient, let's put them everywhere. Well, their calculations, which were for smaller turbines probably would have been fine for the spacing that they have, but not for these larger. So there is a very significant power loss that is occurring. You can also imagine that in the wake, the wind that hits this turbine is more gusty, is not as uniform as it might be for the first row of turbines. And therefore, fatigue and maintenance costs associated with failure of components of the turbine also become more severe in a farm situation. And of course, this is a picture of an offshore wind farm. There can be significant waves and wave loads on the structure supporting these turbines also become a critical topic. All right, of course, one can contrast large scale wind farms to distributed winds, wind farm, this term is used by Department of Energy to describe isolated wind installations in remote areas and so on, where there is no other source of energy and so on, maybe solar and distributed wind can nicely combine for remote locations. Okay, all right, let's move on. So current issues and opportunities are reducing the cost of wind as a critical topic. And there are many ways in which this can be achieved. And we can talk about this later on, developing new energy resources. And this offshore wind is being very strongly emphasized. Integrating horizontal axis turbines and vertical axis turbines is something we are pursuing. We're also pursuing strategies for reducing wake losses and for reducing fatigue. And of course, improving turbine aerodynamics is a topic that is always of interest. There are some other issues that I would like to bring your attention to. Addressing the intermittency of the wind that I stressed is a critical issue. And of course, sort of the big issue with respect to the energy infrastructure itself is an important issue that has to be addressed as well. All right, okay. So let's now switch gears and talk a little bit about different research threads that are being pursued in different groups. Professor Alonso, Crow, and myself, we all three are in aeronautics department. I'm also in mechanical engineering as is John DeBiri. And we are interested in wind turbine aerodynamics, different problems being looked at, performance prediction and optimization for a wind farm is a topic of interest to at least these individuals, perhaps more, full scale field testing of turbines is something that John DeBiri does. And distributed energy systems is also a topic of interest to him. And of course, there are others who are interested in wind energy policies. So that's not a very fair way of describing everybody's research, but we need to move on since we are going to be out of time very soon. So we developed computational models that can do a flow around a particular wind turbine or an assembly of wind turbines. And we can develop models that can then describe the performance of a farm. And we can use that for optimization. And John does full scale field testing and also is interested in these distributed energy systems. So a big challenge that we face is that wind is highly variable. Wind has variations of the order of days. You have weekly variations, seasonal variations. You have daily variations as the sun rises, many places it's calm and then it becomes turbulent as convection starts and so on. And of course, there are variations that are associated with turbulence in the wind. And so there is a plethora of time scales. And so of course, the issue is we have to model this appropriately. We can't capture everything. And so there is many different methods that are available. There is a high fidelity simulation method where you try to capture all the details and so on. And as I summarize here at the bottom, this is very expensive. So if you were to imagine doing this for a thousand turbine array, you're gonna be sitting for quite some time on a high performance computer system. This will take about two million CPU hours. All right. It's something that we do do for research problems and so on these days. So I want to give you a sense that high performance computing has come a long way is this is not something that you would just dismiss when you can justify the cost. One does do in fact, simulations of this kind but it's not something you can use routinely for purposes of design. One can also do low fidelity simulations where different models are put together and so on. And these are now relatively inexpensive. You can run this in a matter of few hours on a desktop. You don't need a high performance computer system but there are deficiencies. There are a certain important physics that are not captured. For example, dynamic loads associated with wakes impinging on another turbine are not captured by such a model. And so what we have been doing in our group is developing models which allow us to do both at relatively low cost. And so we call this multi-scale kinematic simulations. This is work that my graduate student Aditya Ghatte has developed. We have some papers and we can talk about details later on. I just have some words describing it. I'm gonna move on. We take lots of pain and care to make sure that these models are validated, that they produce realistic results and so on. And details, if they are of interest to you, come talk to me later on. And so we are now applying it to large scale wind farms. And so to give you a sense of this, this is motivated by this big fire wheel wind energy complex, which is, when it's fully built, is gonna be the largest wind complex in the world. And this is being done in collaboration with map reality. And so this is what we are able to do today. This prediction I'm showing you is for about 200 turbines. They are located in here. We synthesize the turbulent wind that is going in. The colors are showing the wind speed. Downstream of the turbine, you see the blues where the wind has slowed down. This is for a north-south flow. This is for east-west flow. And you can see that the structure of the wakes and the manner in which they interact is very different. And so the power loss for north-south orientation is 13%, whereas it is a large 40% for east-west orientation. So the idea is that with tools like this, we can look at the sighting. How do we place these so that the overall output is optimized not for one particular orientation, but for the distribution of wind and wind orientations that one expects for a particular location and so on. So these tools are gonna be very helpful in future. So this is my last slide. If you look at the Department of Energy website, you will find this very fancy glossy report. DOE is very good at conducting very comprehensive studies of particular problems and then sticking out their position. So in 2015, they published this report called Wind Vision. And the vision they have is that in 2020, 10% of electricity could be generated by wind. And this fraction increasing to, this is the goal of 35% in 2050. The amount of land area this might need is very small. It's 0.04%, it's tiny just to make sure that we don't get concerned about that. This growth in blue here is that targeted for offshore wind and that's sort of a big new technology push that Department of Energy is envisioning. These are maps where the offshore wind energy resource has been mapped. And if we compare it to what the install capacity is and you can plot that data is available over years over the last many decades and you can see a rapid growth. This is from 2010 in three years, there is a 20 gigawatt increase. And so one can look to new deployments at large scale of wind and the problems that offshore wind at large scale introduces are multidimensional and so on. There's the problem of structures, building these large structures, supporting them in water under the loads associated with waves and currents that may be present. And of course, installing the turbine itself and then questions of power that will need to be drawn from these installations back to where the high demand for them is. So cabling that is gonna carry the transmission line that is gonna carry this power back to where it's used. And then of course, other issues of fatigue and maintenance and so on, never mind those. They are also important, they have to be factored in and so on, but lots of exciting opportunities not only from technology side, but also from the side of policy and from the side of economics and where we as overall community move forward. Of course, it's a nice way of addressing challenges associated with sustainability are reducing our greenhouse gas impacts and so on. Those are something that very naturally and in a clean way happen. So without further ado, let me close and take some questions. Well, so John is very keenly interested in vertical axis wind turbines. And the vertical axis wind turbine, well actually I think if I go to the next slide here. Yeah, this is a picture of turbines that he's interested in. And of course, this is the simplest configuration that you can have and it's easy to install and easy to inspect and so on. One can have more complex aerodynamic shapes in here and get more power out. And so on. I think this is an area which is less developed. There are fewer designs that are being used of vertical axis wind turbines. So it's a naturally rich area for further exploration. I should say that overall efficiency wise, these guys are more efficient than these guys. But these guys can be small. They can sit close to the ground. And so there's no reason why we can combine these with an installation of the horizontal axis turbines. And that's an area of collaboration between my group and John DeBerry's group. Well, so it's a question that you have to ask, what is the penetration? How much wind resource are you deploying? How much of it is being modified by the turbines that you're operating? And if that is significant, then in the local region, there may be other effects because obviously you are modifying the atmospheric boundary layer. And so the manner in which it may be transporting, say moisture or it may be dispersing other things in the atmosphere is affected. And so these are topics that are of interest and are being studied and so on. Now, whether the impact is negative or positive depends on the quantity and also depends on how everything is oriented, what the details of each situation is. And so it's a topic of interest to many faculty here. So I can't say that there is a very simple short answer to that, yeah? One last question, yeah? So I would say all large wind turbine companies are interested in tools and models of this kind. They have in-house capacity to do a sort of the ranks level. I mean, I don't know whether technically I can speak about the details. But the kinds of models that we are developing, we're trying to collaborate with the companies, create these relationships and so on, and make these tools available to them and so on. Because the more they're used, the better it would be. Thank you for that question. OK.