Film cooling is a crucial technique for protecting critical components of gas turbines from excessive temperatures. Multiparameter film cooling optimization is still relatively time-consuming owing to the substantial computational demands of computational fluid dynamics (CFD) methods. To reduce the computational cost, the present study develops a data-driven framework for predicting and optimizing the film cooling effectiveness of high-pressure turbines based on deep learning. Multiple rows of cooling holes located on the pressure surface of the turbine blade are optimized, with the coolant hole diameter, the incline angle, and the compound angle as design parameters. A conditional generative adversarial network model combining a gated recurrent unit and a convolutional neural network is designed to establish the complex nonlinear regression between the design parameters and the film cooling effectiveness. The surrogate model is trained and tested using independent CFD results. A sparrow search algorithm and the well-trained surrogate model are combined to acquire the optimal film cooling parameters. The proposed framework is found to improve multi-row film cooling effectiveness by 21.2% at an acceptable computational cost.

The impact of inflow flow state on the flow-field and far-field noise of supersonic jet flow has been studied. A new turbulent inflow boundary condition based on the synthetic eddy method is implemented in an in-house LES code. The implementation was validated using a turbulent channel flow. Simulation results of an under-expanded supersonic jet with a converging nozzle have shown that turbulent inflow leads to thicker and turbulent initial shear layer with rich vortical structures. The thick initial shear layer hinders the growth of flow instability and break the acoustic feedback loop, which is vital to the generation of screech tone. The alternation of the flow-field by the turbulent inflow also leads to changes in the far-field noise radiation. The screech tone is eliminated, the noise in the downstream direction is considerably suppressed, and the noise radiates in the upstream direction is slightly amplified.

Nowadays, the film cooling strategy is widely adopted to cool down turbine blades and significantly extend their service life. To protect the turbine's structural integrity while improving its efficiency, the effectiveness of air film cooling needs accurate evaluation in engineering. This study investigated a classical flat-plate model with a trench hole under multi-row superposition conditions. However, conventional semi-empirical correlations are hard to predict well. Thus, a Conditional Generative Adversarial Network (CGAN) model was developed to predict the two-dimensional film cooling effectiveness based on the film cooling parameters. The CGAN model consists of a Gated Recurrent Units (GRU) network generator and a Convolutional Neural Network (CNN) discriminator. The depth and width of the trench, compound angle, hole location, and blowing ratio information are mapped to the flow history information. Based on the sequential information, the GRU generator predicted the wall film cooling effectiveness, while the CNN discriminator focused on identifying generated image being real or fake. The adversarial training process significantly improved the accuracy of the generator, and the generator is capable of predicting the wall film cooling effectiveness well.

A state-of-the-art large eddy simulation code has been developed to solve compressible flows in turbomachinery. The code has been engineered with a high degree of scalability, enabling it to effectively leverage the many-core architecture of the new Sunway system. A consistent performance of 115.8 DP-PFLOPs has been achieved on a high-pressure turbine cascade consisting of over 1.69 billion mesh elements and 865 billion Degree of Freedoms (DOFs). By leveraging a high-order unstructured solver and its portability to large heterogeneous parallel systems, we have progressed towards solving the grand challenge problem outlined by NASA, which involves a time-dependent simulation of a complete engine, incorporating all the aerodynamic and heat transfer components.

The film cooling of cylindrical holes embedded in transverse trenches under superposition has shown promise for protecting the critical components of a high-pressure turbine from thermal damage. To optimize the relevant parameters and provide a suitable film cooling strategy, it is important to predict the effectiveness of lateral-averaged adiabatic film cooling with the trench effect on the surface of a blade. However, high-fidelity semi-empirical correlations for film cooling under superposition conditions with a trench have rarely been examined. This study establishes a gated recurrent unit (GRU) neural network model to predict the effectiveness of lateral-averaged film cooling under multiple-row superposition conditions with a trench. In general, a GRU neural network model is built with a large sequence of one-dimensional parameters, including the depth and width of the trench, compound angle, location of the hole, and blowing ratio. The computational fluid dynamics (CFD) method is used to provide a training dataset for the model. After careful testing and validation, the results predicted by the GRU agreed well with the CFD results. Moreover, the performance and robustness of the GRU were better than those of other recurrent neural network models, such as the long short-term memory model. Integrated with the GRU model, the sparrow search algorithm was adopted to optimize the parameters of the trench. The film cooling effectiveness of the optimized case improved by 1.6% compared with the best case, 28.5% compared with the worst case in dataset, and 23.5% compared with the no-trench case.

Broadband shock-associated noise (BBSAN) is one dominant component from off-design supersonic jets. A time-domain BBSAN model is proposed and evaluated, which is based on the vector Green’s function solution of the decomposed Navier–Stokes equations. The source term is the scalar product of the anisotropic velocities with the mean pressure gradient. The approach uses large-eddy simulation (LES) databases. Proper orthogonal decomposition (POD) is applied to the scalar source field to extract the large-scale coherent structures associated with BBSAN. The key features of the BBSAN spectra are preserved when only a small fraction of POD modes are used to reconstruct the source field. The leading POD modes of the source term show strong amplification in the observer direction. Noise source contour plots calculated with these POD modes reveal that most noise sources are distributed at the shock-turbulence interactions. Predictions are validated with the Ffowcs Williams and Hawkings method and experimental data.

Blade vortex interaction noise is a problematic and dominant component of rotor noise. Plasma actuators strategically placed at the tip of the rotor blades can reduce the strength of the tip vortices. This reduction has the potential to significantly reduce blade vortex interaction noise. A combined experimental, numerical, and theoretical program shows supporting evidence that low power plasma actuators can effectively lower coherence of the blade tip vortex and reduce blade vortex interaction noise over-pressure by up to 80%. For a nominal small five-bladed unmanned aerial vehicle, we predict an approximate 8.88 maximum ΔdB reduction for a 150 m/s tip speed. Experimental, computational, and acoustic modeling support these predictions. This study represents a fundamental investigation in the fixed-frame, which provides evidence for higher level research and testing in a rotating framework.

The noise from large-scale coherent turbulent structures within jets remains the dominant source. For the purpose of developing future control systems for the large-scale noise source, we investigate the statistics between upstream and downstream radiating waves. We investigate two off-design supersonic jet flows with instability theory and associated noise radiation, large-eddy simulation (LES), and experiments. We compare the auto-correlation, cross-correlation, coherence, and other statistics predicted by aeroacoustic instability theory. As instability waves are closely connected with the formation of large-scale turbulent structures, they yield insight into large-scale noise statistics. We investigate two nozzles at two supersonic off-design conditions. The first is a biconic nozzle operating at an unheated condition, and the second is a NASA nozzle operating at a heated condition. We find that for these jets, the noise from instability waves is coherent between 0.40 to 0.70 at large-scale radiation frequencies between the downstream and upstream radiation directions.

Supersonic jet noise reduction has become one of the major challenges for the design of high performance aircraft. The broadband shock-associated noise (BBSAN) is analyzed using the simulation data of a heated under-expanded supersonic jet from a high-order compressible large eddy simulation solver. The flow-field results are compared with a Reynolds-averaged Navier-Stokes simulation at the same jet condition. The far-field acoustic results are validated with experimental measurements. An acoustic analogy based on the decomposition of the Navier-Stokes equations is used to analyze the noise source of BBSAN. The source term of the BBSAN is identified, which is the product of anisotropic velocities andmean pressure gradients. The source term is validated by comparing the BBSAN spectra with the measured total spectra. The BBSAN spectra in the far-field are preserved when the flow-field is reconstructed using different numbers of proper orthogonal decomposition modes at different observation angles. The majority of source intensity is distributed on the oblique shock waves and the region near the end of the potential core. Source statistics show large positive correlations paired with negative correlations at the local and neighboring shock waves.

A high-order large eddy simulation (LES) code based on the flux reconstruction (FR) scheme is further developed for supersonic jet simulation. The FR scheme provides an efficient and easy-to-implement way to achieve high-order accuracy on an unstructured mesh. The order of accuracy and the shock capturing capability of the solver are validated with the isentropic Euler vortex and Sod's shock tube problem. A heated under-expanded supersonic jet case from NASA's Small Hot Jet Acoustic Rig (SHJAR) database is used for validation. The turbulence statistics along the nozzle centerline and lip-line are examined. We predict the acoustic radiation with the Ffowcs Williams and Hawkings method, which is integrated with our solver. The far-field acoustic predictions show reasonable agreement with the experimental measurement in the upstream and downstream directions, where the shock-associated noise and the large-scale turbulent mixing noise are dominant, respectively.

The impact of large-scale turbulent structures on the aerodynamic flow-field and far-field radiated noise is investigated through analysis of an over-expanded supersonic jet. The jet operating conditions are Mj = 1.3 and Reynolds number 1.6 × 106. The Kirchhoff surface (KS) method is used to predict radiated noise based on large-eddy simulation (LES) and applied to calibrate amplitudes of an instability wave model. The acoustic pressure time history in the near- and far-field are constructed with the instability wave model. Predictions are validated with experiment and compare favorably. Cross-correlation and cross-spectral analysis shows that the noise from instability waves is highly correlated the upstream near-field and downstream far-field radiation directions. However, the radiated instability noise in the upstream direction is dominated by fine-scale noise. Results show that it is possible to design a control system for large-scale structure noise based on upstream control if the instability noise can be extracted.

Jet noise remains a community annoyance and a major source of hearing damage for military personnel. The physical and mathematical understanding of the noise source are critical for the purpose of its reduction. We decompose the flow-field into a base flow, large-scale spatially coherent structures, small-scale relatively incoherent structures, and associated radiated noise components. The large-scale and fine-scale structures are ascertained via an additional numerical decomposition approach. An acoustic analogy based model is used to predict the noise from the relatively small-scale incoherent turbulence and the large-scale coherent turbulence interacting with the shock-cell structure. We examine these source models combined with the decomposition of large-eddy simulation (LES). We show validated noise predictions of our LES solver with both the Ffowcs Williams and Hawkings equation and our newly developed jet noise model. We evaluate the source models obtained from the acoustic analogy, that are the two-point cross-correlations of the Navier-Stokes equations, to quantify all noise sources. The major advantage of the developed approach is the ease of quantifying both the shock-associated noise and the fine-scale mixing noise sources. Finally, we discuss how our model can be used with existing LES of turbulent flows for noise reduction.

We examine the acoustic radiation from multiple high-speed subsonic and supersonic free shear layers. We decompose the flow field into a base component (an average), a component associated with the spatially and temporarily growing and decaying instability waves, and the acoustic radiation associated from the instability waves. We find an analytical solution for the acoustic radiation through the use of an acoustic analogy. The arguments of the acoustic analogy involve the two-point cross-correlation of quantities associated with the base flow and instability waves. The instability waves are modeled with a newly proposed basis function. A combination of large eddy simulation, steady Reynolds-averaged Navier–Stokes solutions, and turbulence modeling is used to close the acoustic model. We compare our predictions to those of previous investigators and our predictions match previous theory. We find that the dominant acoustic radiation is due to the large-scale highly spatially coherent turbulence. The interaction of the instability waves causes secondary broadband radiation at higher observer angles.

In this paper,emphasis was focused on the structure,function and pathogenesis of the fimbriae and SLT-IIe.A few of factors which induce edema disease of swine were also illustrated in order to make reference for the prevention and therapeutation of this disease.