普惠公司

This work overviews a physics-informed implementation of filtered Rayleigh scattering (FRS) to reconstruct the mean field of turbulent jets, despite the presence of noise in the FRS data. The physics-informed FRS methodology is applied to three different jet flows: a noise-free simulated test case, a Mach 0.8 jet flow with low levels of noise, and a Mach 1.25 jet flow with high levels of spectral noise. Using sparse observations of FRS signals, the physics-informed FRS is able to predict the mean axial velocity fields within 3% accuracy and can reconstruct continuous mean field profiles in high noise scenarios where traditional FRS processing introduces discontinuities. To the best of the authors' knowledge, this is the first implementation of a physics-informed neural network that relies exclusively on measured FRS data and is trained using two key components: (i) the compressible Reynolds-averaged Navier-Stokes residuals, and (ii) the residual between the measured FRS signals and a modeled FRS signal derived from the physics-informed FRS model predictions, where the FRS measurement model is incorporated directly into the training process.

Single crystal (SX) nickel-based superalloys represent a critical advancement in the aerospace industry, enabling the manufacture of turbine airfoils capable of withstanding the extreme conditions of modern high-pressure turbine engines. Due to the elimination of grain boundaries, SX airfoils exhibit superior high-temperature properties compared to their polycrystalline counterparts. Consequently, the onset of recrystallization (RX) constitutes the introduction of high angle grain boundaries (HAGBs) into the single crystal lattice, compromising the fatigue and creep performance of the airfoil under operational conditions. RX in SX nickel-based superalloys is primarily driven by the stresses introduced during the investment casting manufacturing process. These stresses create localized regions of strain with associated stored energy within the material, which can trigger the nucleation of RX grains during subsequent solution heat treatment. However, the appearance of RX displays stochastic characteristics, the underlying causes of which are not fully understood. This uncertainty, coupled with the connection between geometry and residual strain evolution, constrains the complexity possible in the geometric features of the airfoil and internal cooling channel, and thus imposes an upper limit on the performance and efficiency gains possible in turbine blade design. Herein, the present review systematically examines the physical fundamentals of RX in SX nickel-based superalloys and provides a critical synthesis of the state of the art in nickel superalloy RX research. Focus is placed on the evolution of microstructure and defects during solidification and subsequent heat treatments, and the interplay between these factors and the spatial distribution of stored energy. From this investigation, the dominant factors affecting the driving force for RX are identified, and potential strategies to mitigate the presence of HAGBs in SX nickel-based superalloy turbine airfoils are proposed

Evaluation of the wear behavior of composite coatings based on Stellite 6 with additions of WC-Ni and Cr 3 C 2 , produced via high-velocity oxygen fuel (HVOF) spray.

Ingestion of particulate within a gas turbine engine leads to undesirable consequences such as erosion, hot spots, and cooling hole blockages. These concerns are prevalent in the combustor wall region that commonly employs a double wall design made up of impingement and effusion liners. Numerous studies on double wall liners have demonstrated the negative impacts dirt deposition has on cooling performance; however, to this point, little has been reported to have found an effective means to mitigate dirt deposition. In the current study, a novel design was investigated that added an upstream impingement plate to the common double wall design, resulting in a triple wall liner. The purpose of the additional impingement plate is that it allows for a sacrificial middle impingement plate to capture the dirt prior to the flow impacting the effusion plate where heat transfer is most critical. Several parameters such as impingement hole diameter, dirt injection mass, and plate-to-plate spacing were evaluated for different configurations. Using the triple wall design, the data indicates that dirt deposition on the effusion plate is reduced by as much as 87% compared to the effusion plate of a double wall design. Performance comparisons were made whereby the triple wall design maintained the same pressure ratio across all three layers as that of the double wall. Flow blockages of the cooling holes were shown to diminish for a triple wall compared to a double wall design. Additionally, cooling features were integrated on the surface of the middle impingement plate, which further reduced deposition on the effusion plate by 25% compared to a triple wall design without features. Overall, the results in this study show that a triple wall design significantly reduces dirt deposition on the effusion plate surface leading to less flow blockage as compared to a double wall, making them desirable for improved liner durability.

A new higher-order, free-wake method for rotor aerodynamic analysis is presented in this paper. The method uses elements of distributed vorticity to model the lifting surfaces and their wakes. The use of higher-order spanwise elements ensures a higher load resolution compared to traditional filament-based, free-wake methods and does not rely on explicit vortex core modeling that requires user-defined core parameters. Since the present method uses a full-span, singularity-free, relaxed wake, it can resolve the effect of on-blade, partial-span devices. Correlation studies using well-documented experimental data suggest that this free-wake method is capable of accurately predicting the challenging aerodynamics of a helicopter rotor in hover and forward flight. The method provides equal or better level of predictive capability compared to other potential-flow-based methods in its class. Sensitivity studies show that the modeling parameters such as blade panel density and azimuthal time-step size do not have a significant influence on the free-wake solution. Moreover, the method is observed to be numerically well-behaved. The paper concludes with a demonstration of the free-wake method in modeling on-blade, partial-span devices, a capability that will add to the toolset for the design and development of the next generation of rotary-wing aircraft.