 the presentation of this case that I'm again with Pax. But I think we control here a clear, yeah, to show the principle, a clear example of the principles that we are working with. This goes back to the joint publication between myself, Michael Fauer, and Dimitri Waal. It has been published in the 2017 ICOSAR conference. So, yeah, I'm following the scheme of the case that we present and I will try to make it a little faster. So that we can go for the lunch break. In relation to the last case study presentation, we are considering the operation phase. So the design has already been done. The wood park is ending and it's basically the commissioning phase. This is the situation we are finding now as the wood parks have been put in operation. Of course, the more substantial way to consider SHM and inspection and basically the territory management of the structure is to do the design phase. Like this is the approach of the previous presentation. So, and here we have some, yeah, how does structure information contribute to the surface life of extension of wind parks. So this is also what we find. The wind park has been designed. It has been put into operation. And now they also think about or already think about, yeah, but we should extend the surface life because it's very obvious there will be a higher return over investment. So it's very beneficial to extend the surface life of a wind park but also in general of the structure. So, and then I will go through these decision analysis. And we are after the SHM characteristics, so the management characteristics which are important to help the surface life extension. And this is done by the value of information. And it is the perspective that we usually, our decision analysis is very complex. And this is one of the last presentations. We need to find the right level of detail here. And if we cannot incorporate very sophisticated models in this decision analysis, then we need an interface. And here the result is basically an interface to the basic characteristics of SHM. We need for the surface life extension of wind parks. So, very important, second or another very important point here. I've been talking about before, it's the decision scenario. We are thinking in decision scenarios. Who is the decision maker? That's a wind park operator who has just started the operation of a new wind park. The decision point in time is the commissioning phase. The objective is the maximization of the expected benefits. So this is basically the energy production, but not just the energy production and the money, but a full performance modeling of the wind park, of the component of the wind turbine, and of the wind park, a full performance modeling and accounting for the structural risks in a more detailed way. So, well, there's a lot more to say. Wind parks are maybe a little different from conventional infrastructure. And of course it's the machinery which needs to be maintained and which requires regular maintenance. This is all accounted for in our cost model. But the structural risks or the structures do not need such maintenance efforts. But the important thing here is, and this also goes to our modeling and the level of detail. Again, it's the failure of the wind park, which is important because it is about consequences and it's about production loss and loss of the value of the wind park. And this is very important in relation also to our models. We have very extensive models and we can predict something extremely accurate. But for decision analysis, it's the decision scenario which is important. And so the level of detail is here and in the context of structures, it's the structural failure and the structural damage which is important here in this decision analysis. So the life-cycle phase we are looking at is the operation, so it's a complete operation of 20 or 25 years. And we have a performance model of the structure, reliability under deterioration and extreme events, so windwaves. And then the performance of the components of the wind turbine and of the wind park. I think I will skip this. This is our decision scenario. Our general formulation of the decision scenario. We have been talking about this, maybe also not the specifics of our way of information analysis, but maybe let's have a look here. So what system states do we have? We have the wind park operating. We have a component damage. We have a component failure which may lead the component damage here to wind turbine damage and to wind park damage. And the damage may also lead from wind park damage to wind park failure or from wind turbine damage to wind turbine failure. So this is all interconnected. We have such a model and again models. We can have an extremely sophisticated model here of the wind turbine to predict fatigue stresses. I've been doing volume elements on tripod foundations for instance. But this may not be so decisive here and it only gives a prediction, a very limited prediction. But this model is having all the system states of component, wind turbine level, wind park level 1st and 2nd over the complete operation phase. This is the models we need or this is the models we use in the decision analysis which should have a clear connection and an interface to the refined engineering models. The important information here, this diagram is maybe not the details but we have here the service life over 25 years and we can predict the structural reliability or here's the probability of failure or of damage. The light gray lines is the damage and the dark gray lines are the failures on the different limits. In this case it also includes the system functionality where we have the main parts of a cost-benefit analysis. So capital expenditure, operationary expenditures and Apex has something to do with the decommissioning. Then maybe this slide is also important. What is the interface to our structural health monitoring information at all levels from what we can gain from the structure? So it's the information, this is the interface, it's characterized by the type of the information, the precision, this goes to the uncertainties and the costs we need for each SHL strategy, basically these characteristics to be fit in our decision analysis. And then we have been looking at three different strategies. One is fatigue loading on a winter mine level, on fatigue loading, load monitoring on component level. So this is a hotspot measuring basically and here we slowly look on extreme low monitoring, extreme wind extreme waves and this is fatigue waves and fatigue wind or let's say better to say fatigue induced by wind and waves and here the same. And what we find the value of information is that the component load monitoring where we measure the fatigue straits here or the hotspot monitoring where we are measuring directly at the hotspot over the complete science life, this was the assumption here that this leads to a value of information of 27 or 30 truly percent. So that means 27 or 33 percent of the expected total life cycle costs or total operational costs we are in the operational phase and risks can be saved by SHL and we find also that the dust monitoring the loading on the winter mine this does not provide a value of information so it's negative here and of course we have accounted for the SHL system investment installation operation and replacement interval of 10 years. So the conclusions of this case study is that the wind park surface life extension is not optimal without SHL and the reason is because the structure and the risks are simply too high and we need structure life monitoring but with a rather high precision so this goes to the component hotspot fatigue monitoring and this is the third conclusion so we need to know something about this goes to the type of the information and we need to know something about fatigue and this will then be efficient and I think these results can rather directly be applied but of course the strategies which are modeled here they need further substantiation what this specifically means so we have been providing here what our finding is this is what I said in the beginning it is about the type of the SHL information so we have clarified the type it's the fatigue we need to monitor and then we have clarified that we rather need relatively high precision this is the second point here low precision may not be a value of information optimal so you may spend more than you are getting back and the third point goes to the costs so we have documented the cost model we have been assuming and of course cost modeling is also subjected to uncertainties thank you this was my presentation thanks just a few comments, questions before we go to lunch yeah looking at your slide there I can't understand the type of reasons for that for maintenance purposes and under the light of an existing structure you do not need to monitor the condition of the wind turbine modding specifically but if you look at the biggest uncertainties concerning the design of the wind turbine it involves the wind turbine modding so and there is a lot of research for instance your colleagues at BTU there is a big operation there that lives on wind action on wind turbines so it would be very valuable for them to get information from a big part like this the wind turbine modding part even though it does not directly affect the life of the existing structure yes yes I think this is a very good point that we need to feedback from our SHM measurements to the design I think this is a continuous effort when we look at the development phase when the first offshore wind turbines were developed then there were studies on how to calculate it and they had measurements then they decided for which which modding type and the design phases are appropriate and that is a time domain due to the dominance of uncertainties regarding the aerodynamics and also the structural part the soil structure and the action so this is only economically captured with time series analysis not like we have for conventional structures where it can be analyzed in the frequency domain this is mainly linear so it is a continuous effort it has been there from the very beginning but of course this is an important point thanks this is touching upon value of information in the context of prototyping so in terms of synthesization prototyping it is a feedback a lot of design data is actually gathered from just second turbine which you are monitoring in a very detailed way you have this part where you have the interaction between one turbine to the next turbine and that can create an entirely different logic pattern where it is not easily accessible unless you have the capabilities or possibilities to actually realize or see how it is in this type of situation very short time one thing is maybe we have to clarify a little bit more what is what we do because they are all interrelated I guess there are more slides I think to present but the second most important comment is there are a lot of slides to many slides if you plan to put a system in each one or put some representative so you can say it is a moment and correlate results to others because they are all in the same field as we do for all the slides for example the idea is to explicitly account for the dependencies and then it becomes possible to monitor this turbine and to infer the condition on the other one we are actually working on it that we optimize the inspections and monitoring by accounting for the dependencies because this allows us also to infer through the condition of the others I am closing the session thank you very much thank you