 So thank you all for staying here after coffee. I know it's difficult. But I'm grateful for, for you to be here. So I wanted to present some of the work I did at the Seia Cadaraj, the South of France, just besides the ether side. This was a work I did for the internship. A two month internship, very short one. Where we profited with the researcher Remy no it has. And this is specifically the work was to improve the injection, the gas injection system for the west of my. Well, what's the motivation of the project. Well, we wanted to achieve what were the objectives, a bit of the main piece I have to model which was the kids electric actuator. And at the end of results, which are well the case electric model, we obtained the application of the real kids electric model. So, translate them, the computational model to the real one, and then coupling this gets elected to the real injection system, the west of my. And then some conclusion. So the motivation is the importance of keeping a control density profile in a to come out. So as you know, we want to construct a real time reconstruction plasma the city for the west of my for the system that you've already built it, but we want to construct a real time reconstruction plasma density for the west of my for the system that you've already built it. But the piece that I've worked with was still not developed. So you can see here we'll have a match of the west of my. The inside, before it was called the store super. So this is the main injection system. You can see we have the balloon which contains the gas, we want to put inside the machine. We have a flow meter for measures. And then we have the flow, the flow of the different species that one that should be injected into the vacuum vessel. Then we have the best the pizza ball is in charge of allowing the gas to enter the vacuum vessel. And this is my, this was my, my work to model the speed so it gets wild. And then talking to the computational model. So for you have a gas going into the vacuum vessel and you want to maintain a certain profile and an optimal profile. And then the data that's coming from the vacuum vessel is then send it to the, to the piezoelectric that allows you to control the amount of gas that is injected. And then keeping and steady. So, some of the objectives. It was to open to the model for the piezoelectric. They humiliated that injection line with this get electric model for the West parameters. And then the most difficult part I found for this was to obtain the control values to to maintain a steady pressure and then the profile. So, for example, some specific objectives like developing the mathematical equations that model is this gets electric. The, the services that I will talk about this. And then validate the simulation and contrast with the actual campaign running of the West of America. So, for the piezoelectric actuator, as I said, the value of the ball that allows the, the past gas to the, to the vacuum chamber. So, a material that when there is applied an electric field, it suffers some bending, and this can be measured and often is really, really accurate. And it has, it possess a nonlinear behavior. So, I, there is a neighbor actually. The favor for the electric, I use what's called a movement system of equations that model and hysteresis. And since here, the model is more than like a harmonica dump and harmonica simulator, where M is the mass of the piezoelectric be the dump the natural dumping of the material. And today is also a parameter of the, of the piezo and is just the hysteresis value that models the hysteresis behavior alphabet and gamma are internal parameters that shape and how long the, the, the voltage applied to the, to the and also how far is the, the torque and for the 20 minutes later. So, the first part was to match the model, the mapping model to the paper I use to change the model. With the values that we're already on paper, I just made a simulation in Matlab. And with the ones of the paper. So, I can be assured that the model model actually will be sold. And what these are the values that were on the paper, then. So, this is a simulation in teaming for which are not aware of what seemingly is just another way to write differential equations, a system of equations in a very way, it was my first time using it so I spent a lot of time, a lot of time learning how to use this program. So as you can see, these are just a reproduction of the same result of the paper. And in the same values days. So we have an effective displacement produced by the hysteresis behavior in the, in the voltage and in the, in the displacement suffered by the get selected by the hysteresis behavior. Now, the next step. Yes, let's use the real values, the get selected, they use in the real with injection system. As you can see, we have the massive 28 kilograms, the dumping from sand, the extra constant. And the alphabet gamma, which were more difficult to obtain things where we have to apply numerical methods to it. But it's still the result of things were were quite good. So that is the fabric, the fabric data, they sent from the, for the real. So that was my second step, much the model, the real model with the real electric using the West. Once this was done. The next step was to couple with the electric ball, which was not in the model. The actual compute computational model of the whole injection system. And you can see here, well, there's a lot to see. Here we have like the pressure of the balloon. Then all the system here is the one that the information you get from the vacuum chamber that speeds the information in the piezoelectric that allows the electric current applied to the pixel. Measure how much gas you have to inject to the vacuum chamber. It's quite quite interesting for me, since I work most in theory for me this was like an open window or a new world. So this is the gas injection line. This is the piezo, this is the piezo. Main system interpretation of this opportunity. And then, well, I have to use the actual values of uses in West Pokemon. Then much of the time, it was consumed by obtaining the PID control values, which they say they control. So the course that we'll see later, the course that actually matches the profile you want to have inside the vacuum chamber. So, a good way to do this is to do some numerical methods, the place in the medical methods to obtain the PID control values. But the thing is, it was, I only had two months for this, for this project. There was not much time to obtain the recommended so I had to start trying different values and that's much of it. So which what, which were the results based. So here we have the pressure vacuum chamber. So we wanted to achieve a 0.1 Pascal inside the vacuum chamber. So as you can see, at certain time, the, the piezo electric valve received the order to open up, we apply some, some voltage as you can see in the, the right, the right grass, we apply a very high voltage, then the piezo opens, and then the, the, the flow starts to come inside the vacuum chamber. Then as you can see, the voltage starts, starts to regulate itself, it starts to decrease, and then the piezo starts to close, until they reach the steady pressure you want to have inside the vacuum chamber. We have also the injected gas plugs. I think for this we use hydrogen and also the pressure difference we want to achieve. So the pressure difference is the pressure between the value you want to have and the value that is being injected into the vacuum chamber. So yeah, now the idea was to match all these results with the actual results taken from the, from the West of America. Last week, there was a water leak inside the chamber, from the vacuum chamber, so they can find the blade and it wasn't possible to get the real, real data. But some final conclusion, well, the piezo electric was a satisfactory model of them reproduced. The external behavior also matches with the real values and the values from the, from the paper we take the model and also from the real piezo electric use. And the coupling shows very, very promising results, but as I said, they need a final corroboration and until now, I don't know if the campaign from the West already started, I'm not sure. But it's necessary to make corroboration to the, to the model. And that's it. That's it. Let me see. No, we didn't do any like error analysis, because there was not much time to do the, I don't recall I have to check the report. Maybe some error propagation and it's necessary when when it's concerning PID control systems. Thank you very much. Thank you.