Search

This work shows the application of Multi-fidelity Uncertainty Quantification to Wind Engineering problems. As test case a rectangular shape is used, with a fillet radius, in order to represent the geometrical variations that can affect buildings or other bluff bodies. The rectangular cylinder used has a chord-to-thickness ratio 5:1. This rectangular shape is an important basic shape for wind engineering tasks, e.g. in case of buildings or other bluff bodies exposed to the flow. Moreover it is well investigated and documented. Coarse and fine meshes are used as low and high fidelity models respectively. To perform CFD simulations, the stabilized finite element methods are used in both the high and low fidelity model with a CFD code developed by TUM and the International Center for Numerical Methods in Engineering. The underlying UQ framework is based on a Sparse Arbitrary Moment Based Algorithm (SAMBA) developed at ICL. In the formulation the number of simulations is reduced using a Smolyak sparsity model. The multi-fidelity extension, with application to wind engineering problems is discussed and presented in this work. The overall goal of such formulation is to gain an accuracy of mixed lowhigh fidelity simulations comparable to the ones obtained with only high fidelity simulations, at a fraction of the computational cost.

The knowledge of the water vapor distribution in the Earth's atmosphere is of great importance for weather prediction. Meteorological models, including the so called limited area models, can assimilate humidity measurements, increasing the reliability of the simulated atmospheric dynamics. Existing atmospheric remote sensing instruments are still insufficient to cover global scales with systematic WV measurements in the lowest part of the troposphere (5-6 km): this would indeed help to improve both climate modeling and numerical weather prediction capabilities at short time scale. In order to solve this problem, a new approach (the Normalized Differential Spectral Attenuation-NDSA) capable of estimating the integrated water vapor (IWV) through attenuation measurements in the Ku/K band along microwave is assessed. Nevertheless, the developed technique needs to be demonstrated with real measurements before moving toward space applications. A recent project (SWAMM-Sounding Water Vapor by Attenuation Microwave Measurements) had as main goal to develop low-cost instrumentation to perform the first NDSA measurements. The instrument was designed to operate with the transmitter located on the ground and the receiver placed on a moving platform (e.g. airborne, stratospheric balloon, HAPS), even though, as a preliminary test, the instrumentation was tested in a ground- to-ground link configuration. An instrument prototype, which includes a synthesized microwave transmitter and a Software Defined Radio (SDR) microwave receiver has been designed, assembled and tested in laboratory. Laboratory tests confirm that the transmitter generates two tones, respectively at 18.8 GHz and 19.2 GHz, with very high stability in both amplitude and frequency, and the receiver measures the amplitude (and its fluctuations) of the received signals with enough accuracy (< 0.1 dB) over a wide dynamic range. The differential attenuation is ±0.02dB then far below the theoretical value of the minimum sensitivity required to measure IWV. The instrument was then tested on a ground-to-ground link during an experimental campaign, which was carried out in the period June-October 2018. In order to obtain a transect in the troposphere of tens of km, (thus providing a value of differential attenuation high enough to be measurable by the instrument), the transmitter was installed at sea level and the receiver on the top of a mountain (around 2165m a.s.l.). In detail, the receiver and the transmitter were installed at the CNR laboratories, respectively in Bologna (central Italy) and Cimone mount in the Apennines; as a result, the length of the transect is 62 km. Data are acquired continuously for four months well covering different atmospheric conditions occurred in the summer/autumn seasons. Differential attenuation exhibits temporal variations, at both weekly and monthly scales, which can be correlated to the IWV. As first validation we compared the IWV calculated by NDSA measurements with the same quantity calculated, under the assumption of plane parallel atmosphere conditions, by using different hygrometers placed along the transect and daily radiosoundings of a nearby WMO station. The comparisons show that data are generally in good agreement as both absolute values and temporal variations. However, further investigations are necessary to better interpret the results also in terms of a few discrepancies found. Finally, Sentinel- 3 OLCI L2- IWV product will be used as additional term of comparison for the retrieved IWV over the same transect. A detailed analysis will be presented at the conference.

This paper presents an investigation of the aerothermal performance of a modern unshrouded high-pressure (HP) aero-engine turbine subject to nonuniform inlet temperature profile. The turbine used for this study was the MT1 turbine installed in the QinetiQ turbine test facility based in Farnborough (UK). The MT1 turbine is a full scale transonic HP turbine, and is operated in the test facility at the correct nondimensional conditions for aerodynamics and heat transfer. Datum experiments of aerothermal performance were conducted with uniform inlet conditions. Experiments with nonuniform inlet temperature were conducted with a temperature profile that had a nonuniformity in the radial direction defined by (T max-T min)/T-=0.355, and a nonuniformity in the circumferential direction defined by (T max-T min)/T-=0.14. This corresponds to an extreme point in the engine cycle, in an engine where the nonuniformity is dominated by the radial distribution. Accurate experimental area surveys of the turbine inlet and exit flows were conducted, and detailed heat transfer measurements were obtained on the blade surfaces and end-walls. These results are analyzed with the unsteady numerical data obtained using the in-house HybFlow code developed at the University of Firenze. Two particular aspects are highlighted in the discussion: prediction confidence for state of the art computational fluid dynamics (CFD) and impact of real conditions on stator-rotor thermal loading. The efficiency value obtained with the numerical analysis is compared with the experimental data and a 0.8% difference is found and discussed. A study of the flow field influence on the blade thermal load has also been detailed. It is shown that the hot streak migration mainly affects the rotor pressure side from 20% to 70% of the span, where the Nusselt number increases by a factor of 60% with respect to the uniform case. Furthermore, in this work, it has been found that a nonuniform temperature distribution is beneficial for the rotor tip, contrary to the results found in open literature. Although the hot streak is affected by the pressure gradient across the tip gap, the radial profile (which dominates the temperature profile being considered) is not fully mixed out in passing through the HP stage, and contributes significantly to cooling the turbine casing. A design approach not taking into account these effects will underestimate the rotor life near the tip and the thermal load at midspan. The temperature profile that has been used in both experiments and CFD is the first simulation of an extreme cycle point (more than twice the magnitude of distortion of all previous experimental studies): It represents an engine-take-off condition combined with the full combustor cooling. This research was part of the EU funded Turbine AeroThermal External Flows 2 program. © 2012 American Society of Mechanical Engineers.