Subproject D6: Interaction of Combined Module Variances and Influence on the Overall System Behaviour
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 6 | |
| Author | ||
| License | CC Attribution - NonCommercial - ShareAlike 3.0 Germany: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this | |
| Identifiers | 10.5446/68359 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
5
00:00
Capital shipPlayground slideGas compressorTurbineEngineBliskBladeFord FocusPhotographic processingFinger protocolRail transport operationsToolAircraft engineMarine propulsionModel buildingAmmunitionBlow torchTurbofanRadial engineMaterialThrustMode of transportSpaceflightTypesettingNaval architectureDifferential (mechanical device)BecherwerkCartridge (firearms)Casting defectJet engineSpare partVolumetric flow rateWearShip of the lineAbgasturboaufladungMeltingTurbineGas compressorTurboluftstrahltriebwerkAngle of attackModel buildingRemotely operated underwater vehicleRail transport operationsFlightEngineSeparation processAir compressorPlatingRoll formingNaval architectureThrottlePlane (tool)SizingMechanical fanThrustSchiffsdampfturbineMarine propulsionRadial engineForgingBlow torchFinger protocolPiston ringWireKlemmverbindungToolStrömungsmaschineHybrid vehicleAircraft engineTurbofanFord FocusFlight simulatorComputer animation
Transcript: English(auto-generated)
00:00
Thank you for the short introduction and welcome to today's final conference. My name is Jan Goeing and together with Professor Jens Bittres, we have researched the interaction of combined model variances and their influence on the overall system performance in the subproject D6. Subproject D6 entered CRC 871 with a focus on the overall system in the third funding
00:29
period. We are part of the project area D for holistic control of the regeneration process and this is also reflected in the system demonstrator where we are responsible for analyzing and
00:46
evaluating the influence of geometry changes of the high pressure turbine on the overall system and the overall system is represented by the EFAS research jet engine, the B25A1, which we were also able to integrate into the CRC.
01:04
Furthermore, we are using our institute's own simulation tool ASTOR to represent the B25 engine on the virtual layer and ASTOR stands for aircraft engine simulation for transient operation research.
01:23
Let's come to the motivation slide. The performance of the aircraft is directly and indivisibly linked to the performance of the propulsion system. In our case, the propulsion system is a turbofan jet engine which will continue playing a role in the future aviation, the increasing demand for more power, a lower
01:46
fuel consumption and better economic economy is leading to work at the aerodynamic and thermodynamic limits. One of these limits is the temperature downstream the combustion chamber and in the
02:00
turbines in order to increase the thermodynamic efficiency increasingly higher temperatures are reached that are to the load limits of the turbine plane. These temperatures lead to increased wear faster and heavier as, for example, in the case of the high pressure turbine in form of hot gas corrosion.
02:25
Other examples are erosion, which you can see here, or fouling by aerosols or super particles which change the surface structure on the plate. Fouling and erosion occurs everywhere on the jet engine.
02:43
So here is a high pressure compressor of the V25 represented. When new from factory, the surface is very smooth, as you can see here in the green box. And after several flight cycles and hours, it becomes noticeably rougher, as you can see
03:04
here in the yellow, red and dark red box. The wear has also a direct influence on the aerodynamic of the compressor. On the dioprop below, you can see the pressure ratio and worth as a mass flow. The green lines are representing the characteristic of the compressor.
03:24
Furthermore, you can see the transient operating line of the compressor doing an acceleration between a low and a high thrust level. And as you can see, during the acceleration, the distance to the stability limit becomes smaller. In principle, we want to work as close as possible to the stability limit,
03:45
because there the efficiency of the compressor is very high. If you walk, the map shifts downward to the left and the operating curve shifts upward. The results in an even smaller gap between both lines or limits.
04:04
As you can see, wear has a strong influence of the performance, especially in a transient and dynamic load case. Based on that, we want to develop a model that can simulate this load case as accurately as possible.
04:23
Furthermore, wear occurs in all areas of the engine. We have falling erosion and tip gap in each turbo machinery, corrosion and melting in the hot gas part. And therefore, we want to investigate these interactions of combined model variances
04:41
with the respect of the overall performance, especially in transient load cases. To investigate these, there are some goals in the sub-project, which can be generally divided into three areas. And I would like to give a short overview.
05:03
First, we have developed a virtual twin of our research engine. For this proposed, we have extended the classical 0D technique to a quasi-one dynamical model in the gas part and 2D method in the heat transport.
05:23
With this, we are able to consider the dynamics of the system and can even simulate highly dynamic cases. On the diagram on the right side, you can see the pressure over the mass flow again. And here we get over the stability curve, which is the red line.
05:43
Then we have compressor stall until the flame in the combustor goes out. And after that, the compressor recovers or the complete engine recovers. Furthermore, we are interested in validating our model. Here we do this first with the models from the literature.
06:03
On the diagram here, you can see again pressure ratio over mass flow. And here you can see a conversation between ASTOR and the classical global engine method technique. We can use the literature to validate our model
06:20
and to evaluate the difference between the state-of-the-art techniques and our model, which considers the geometry as well as the interactions in the gas part. Furthermore, we used our small turbojet engine to validate ASTOR. And finally, we carried out a complex measurement campaign
06:50
with our research engine, the V2.5. Finally, we transferred the validated model into a meta model, which is used for the third area.
07:01
In this area, we want to evaluate or we want to investigate the causalities and detections of the model variances. For this purpose, we started a numerical experiment in which we found out the power-specific sensibilities. This study is based on the scaling factors, which describe the impact of geometry changes
07:23
on the aerodynamic of the different turbo machineries. In the following, I would like to present some of these results in more detail. As I mentioned, we developed a virtual twin of the V2.5 jet engine in higher order.
07:43
Therefore, we used the unsteady Euler equation, which you can see here, to describe the gas part. Different from the state-of-the-art techniques, we take the volume information to describe the capacity and the inertia of the system in the gas part. Quasi-1D means that we use the length of the control volume as well as the inlet and outlet area,
08:03
which can be different for the convection part of the Euler equation. This is the second part. We also use, for example, surface forces to describe the impact of the compressor and so on.
08:22
Furthermore, we developed a reduced order model to consider the heat flow. These are especially important in the gas area, which you can see here. Here is the combustion chamber as well as the turbine plate. The heat flows goes into the turbine, for example,
08:41
into the turbine plates, into the housing, and also into the disc, which expand differently due to the terminal and mechanical stress. We also consider the radial expansion in order to map the influence of the efficiency, which is strongly related to the radial gap in the transient load case.
09:03
In particular, looking into the future aviation where different physical domains are gaining in importance, for example, the hybrid electric propulsion systems, there's a strong motivation for an overall system-unified modeling,
09:23
which is generally understandable. Therefore, the complex systems, as I described above, has been transformed in the Bézoidé-Bondcraft notation in order to improve the fundamental research in this area. The Bézoidé-Bondcraft theory
09:41
divided all quantities into flows and efforts. Flows are something like mass flow, rotation, speeds, while efforts are pressure or temperature, for example. Here you can see the same control volume as described above,
10:02
and there's also used as energy and momentum conservation. The flows and efforts are connected by 0 and 1 junctions where analytical rules prevail. This allows us to describe the interactions of the differential equation system,
10:22
and also integrate the capacity and inertia, which are shown here with the IC or TF symbol. Furthermore, other systems can be connected, such as heat conversion in a very high system or tough power.
10:41
The diagram now describes the complete differential equation system and can be extended by deserved so that further systems are integrated analytically correctly. After this was done for the complete engine, we have validated this model. So the most elaborated test happened at the beginning of this year.
11:02
Here you can see our research engine and the MT-WANOVA testbed. Together with several projects from the CRC, we did a very successful test. Here in the background you can see the boss ring from the TFD, as well as the wires for special instrumentation
11:23
on our engine. This allows us to record additional data in the high pressure compressor behind high pressure turbine, low pressure turbine, and to be used individual thrust curves to validate astro. The right diagram,
11:41
you can see the rotational speed over the time during a slam acceleration and deceleration. The blue lines are representing the results from astro and the green lines from the real V25 engine. And as you can see in the cycle, we are able to calculate the agility
12:02
of the system reaction. And in the left diagram, the different temperatures over the time in the same acceleration and deceleration are plotted. Below is the temperature behind the low pressure turbine,
12:21
TT5, and above the temperature behind the high pressure turbine, TT45. And with our model, we are able to calculate the curves with an average deviation of 1.2 until 2.1 during the acceleration. And in the deceleration,
12:41
we are below 1%. Based on the data, we are also able to calculate realities that are difficult to measure. As an example, you can see above the temperature TT4 and the transient load peak. This is the highest load point that has been created
13:01
or effect on the load of the high pressure turbine, for example. Further examples are the transient operation curves of the compressors as well as the distance to the stability. Based on the validated model, a complex numerical experiment was carried out to investigate the interactions
13:21
of combined model variances. In this step, we analyzed the parameter space. Therefore, we have a great support from several other sub-projects in the CRC which carried out a lot of CFD studies to find out the impact of
13:42
geometry variances on the aerodynamic. Furthermore, we performed a gap path analysis between our research engine and a new V2.5 engine which allowed also to estimate the parameter space. By using this parameter
14:00
space and by defining some missions, we performed a thousand different performance simulation of different engine configuration using latent hypercube sampling and this results were then transferred into a metamodel
14:21
which in our case is our deep learning architecture. With the help of the AI more simulation could be generated in a short time and the sensitivity analysis was applied to find out the coastalities and in the following I will present some of the results
14:41
of this experiment. In the upper diagram you can see the influence of different engine configurations on the TT5 temperature and the vertical X shows the increasing temperature
15:01
compared to a new engine and on the XX you see different engine configuration first only the high pressure turbine deteriorated followed by the self deteriorated high pressure turbine at FM and so on until the complete engine is deteriorated
15:21
and in general you can see that the temperature T5 increase and the red bars represent the influence of the high pressure turbine and the blue bars the influence of the other components however the influence of the high pressure
15:41
decrease in this example at the first bar we are at 49 Kelvin and at the end it's only 41 Kelvin this effect is based on the system reactions by the high pressure and low pressure system further more it can be seen that the TT5 temperature
16:00
increase in particular due to the wear of the high pressure turbine this is the first bar and when the high pressure compressor joins in the deteriorated engine condition however the influence of wear on the system cannot be evaluated by
16:21
only one parameter in this diagram you can see the distance between the operating curve and the stability limit during the acceleration over the rotational speed the dashed lines are the steady the steady state
16:41
operating line and the solid line stands for the unsteady operation and green again the factory new engine and blue high pressure engine with the deteriorated high pressure turbine and blackened engine with the deteriorated high pressure compressor
17:01
and it is clearly visible that the high pressure compressor which still has a similar influence on the temperature TT5 as the high pressure turbine has reduced the distance here significantly more and therefore more important for safety
17:21
for the safety of the operation further more you can see in red the combination of high pressure compressor and turbine which is again not the sum of both because of the system reaction let us come now to the conclusion we developed a
17:40
model higher order to include volume dynamics as well as interaction effects in the gas parts and between different systems this enables us to more accurately simulations and load cases this model
18:00
is used for the system demonstrator on the virtual layer to evaluate the influence of geometry changes of the high pressure turbine on the overall system and in collaboration with other studies we found out the influence of the high pressure turbine geometry variances
18:20
on the summarized aerodynamics I see the pressure ratio the mass flow and the efficiency and we have variances in the tip gap in the stagger angle and the radius of the leading and trailing edge as well as the crowd length
18:42
and based on that we conducted a sensitivity study which showed us that the stagger angle had the greatest influence of the pressure ratio mass flow and efficiency and the next step we combined the results with the results of other studies to a global sensitivity analysis
19:01
in order to transfer the influence of the change aerodynamic size on the overall performance here you can see the deviation of the fan the high pressure compressor, turbine and low pressure turbine on different performance
19:21
values for example Tt5, the fuel flow, Tt3 and 1 and 2 and the pressure below the low pressure turbine and in red are the interaction effects as you can see
19:40
the high pressure turbine has a strong influence especially on the fuel on Tt3 and Tt5 so this parameter strikes strongly when the stagger angle could be the cows furthermore we see that the high pressure compressor also many peaks
20:01
so that the conclusion is that the model variance in the high pressure that the model variances in the high pressure system dominates the performance in this case, in the case of a conventional aircraft engine as a Tt5 engine however we also observe
20:21
that the low pressure system those deflection and interactions in these areas are strong and noticeable therefore we assume that in modern aircraft for example in a ultra high bypass engine the low pressure system will play an important role in making
20:41
such analysis more important than ever thank you very much for your attention and for any questions please go ahead