Pong-Jeu Lu

Numerical Investigation of Inlet Buzz Flow

The present research aims at developing a time-accurate, high-resolution total variation diminishing scheme and a data-processing procedure that can analyze the inlet buzz e ow problem. Special care has been exercised on the numerical buzz e ow initiation procedure to minimize the generation of spurious numerical waves. A 10-ft ramjet engine was adopted as the simulation model. The simulated results show that the buzz cycle is attributed both to the local e ow instability around the entrance and to the acoustic resonance modes appearing inside the plenum chamber. A revised upstream feedback mechanism is proposed in the present work. It was found that the feedback loop for the subcritical operation is established locally around the inlet region, in which the ree ected acoustic waves were sent upstream as a result of the impingement of the shock-induced separation vorticities on the centerbody surface and/or the cowl lip. For the supercritical operation, however, the formation of the feedback mechanism is ascribed to the fundamental acoustic resonance mode generated in the plenum chamber.

Application of autoassociative neural network on gas-path sensor data validation

Gas-path analysis holds a central position in the engine condition monitoring and fault diagnostics technique. The success of gas-path analysis, as concluded from previous investigations, depends mainly on the quality of the measurements obtained. For approaches using either a classical Kalman filter method or a contemporary artificial neural network approach, a high success rate of diagnosis can only be guaranteed when a correct set of measurement deltas is available. The objective of the present work is to design a genetic autoassociative neural network algorithm that can perform offline sensor data validation simultaneously for noise filtering and bias detection and correction. The neural network-based sensor validation procedure usually suffers from the slow convergence in network training. In addition, the trained network often fails to provide an accurate accommodation when bias error is detected. To remedy these network training and bias accommodation problems, a two-step approach is proposed. The first step is the construction of a noise-filtering and a self-mapping autoassociative neural network based on the backpropagation algorithm. The second step uses an optimization procedure built on top of these noise-filtering and self-mapping nets to perform bias detection and correction. A nongradient genetic algorithm search is employed as the optimization method. It is shown in the present work that effective sensor data validation can be achieved for noise filtering, bias correction, and missing sensor data replacement incurred in the gas-path analysis. This newly developed algorithm can also serve as an intelligent trend detector. A true performance delta and trend change can be identified with no delay, to assure a timely and high-quality engine fault diagnosis.

On the shock enhancement of confined supersonic mixing flows

Direct numerical simulations using high‐resolution total variation diminishing (TVD) scheme are performed for studying the shock enhancement of two‐dimensional confined (spatially growing) supersonic mixing flows. Several specially designed mixing enhancement schemes are examined with emphasis placed on the study of the fundamental aspects involved in the shock‐induced mixing enhancement process. The merits associated with these mixing enhancement schemes are evaluated based on a cost/effectiveness criterion, in which the cost paid for the total pressure loss encountered and the improvement in mixing gained are considered together. The results suggest that mixing enhancement using shock waves can only be effective if the stimulation is spatially persistent, and begins from the very upstream. Being motivated by this observation, an idea of using wavy‐wall configuration to generate the desirable periodic shock stimulation is proposed and investigated. The computed results show that, by an appropriate manipu...

Numerical investigation on the structure of a confined supersonic mixing layer

Direct numerical simulations using the high‐resolution TVD scheme are performed for the study of two‐dimensional confined spatially growing supersonic mixing layers. Numerical simulations, being forced by the linear instability wall modes, are justified by comparing the calculated results with both theoretical and experimental data. The numerically visualized downstream compression/expansion wave system explains the so‐called ‘‘strange waves’’ observed in experiments and are found to be responsible for the rapid growth of the mixing layer in that region. The existence of shock waves would lead to mixing enhancement by accelerating the process of extracting turbulent kinetic energy from the mean stream. However, the phenomenon of shock‐induced mixing improvement is found quite local around the shock impingement area, and often is followed by a saturation in turbulent energy absorption which impedes the further growth of the shear layer in the immediate downstream region. For the present cases considered, the subharmonic instability modes are not dominant, and the baroclinic‐torque effect is comparatively less important for the vorticity development. These observations suggest that compressibility effects may behave in a considerably different manner as the underlying instability mechanism changes from the subsonic Kelvin–Helmholtz type to the present supersonic wall instability type.

Flutter analysis of cantilever composite plates in subsonic flow

The high-precision 18-degree-of-freedom triangular plate-bending finite element is used to study the effects of composite filament angle, orthotropic modulus ratio, sweep angle, and aspect ratio on the vibration and flutter/divergence characteristics of cantilever plates in subsonic flow. The element stiffness and mass matrices are generated according to the classical lamination theory. Unsteady airload acting on the plate is evaluated by the use of lifting surface theory, which is solved numerically by the doublet-lattice method. To facilitate flutter analysis, interpolation using a surface spline is employed to interconnect the structural nodal and aerodynamic control points. The flutter/divergence tradeoff of aeroelastic tailoring is found, in a more involved manner, for the present plate-like low-aspect-ratio wings as compared to that associated with the beam model first discovered by Weisshaar. Effective enhancement on flutter/divergence performance can be attained by varying the orthotropic modulus ratio while as appropriate fiber orientation is selected. Also, as discovered previously, the present numerical results confirm that the structural tailoring can provide a harmonious balance to the sweep angle effect upon the aeroelastic stability characteristics of a wing. From the numerical examinations performed, it is concluded that substantial improvement upon flutter/divergence characteristics can be achieved by using composite materials. Nevertheless, the benefit can be gained only through a thorough parametric investigation because the aeroelastic stabilities are also complicated by the directional stiffness of composites, in particular for the cases of low-aspect-ratio wings.

Flutter suppression of thin airfoils using active acoustic excitations

A theoretical analysis of the flutter suppression of oscillating thin airfoils using active acoustic excitations in incompressible flow is presented. Closed-form unsteady aerodynamic loads induced by a simple harmonic acoustic excitation on a typical section model are derived. The acoustic wave generator used in the present flutter suppression analysis is activated by a state feedback control law that particularly takes into account the relative phases between the sensed states and the acoustic excitations. The flutter boundaries of the typical section, with and without the acoustic excitations, are evaluated using both the V-g and root-locus methods. The results show that, although the acoustic wave is a weak flow perturbation per se, the induced aerodynamic loads can be large enough to be employed as the flutter control forces. The circulatory part that makes the flow satisfy the Kutta condition at the trailing edge contributes the most to the magnitude and phase of the acoustically induced airloads, in particular when the acoustic excitation position is placed close to the trailing edge. Parametric study reveals that both the phase of the feedback gain constant and the acoustic excitation position are critical for the present new flutter suppression technique.

Hemodynamic and metabolic effects of para- versus intraaortic counterpulsatile circulation supports.

Despite the success of intraaortic balloon counterpulsation, data on physiologic indices and optimal inflation/deflation timing control of chronic counterpulsation devices are unclear. This study explored the acute hemodynamic and metabolic efficacy of a novel 40-ml stroke volume paraaortic blood pump (PABP) versus a standard intraaortic balloon pump (IABP). Acute porcine model was used with eight pigs randomly divided into PABP (n = 4) and IABP (n = 4) groups. Hemodynamic and metabolic measurements were obtained with and without mechanical assistance. In one pig, the inflation/deflation control was adjusted to different settings, with corresponding performance indices measured. The PABP significantly improved classical counterpulsation indices (p ≤ 0.05) and achieved an average beneficial effect on these indices 1.5–3.5 times greater than that of the IABP. Classical metabolic indices (tension time index and endocardial viability ratio [EVR]), and indices new to chronic counterpulsation research (coronary perfusion, left ventricular stroke work (SW), and a newly derived EVR) were also used in assessment. Both IABP assistance and PABP assistance improved these physiologic indices, with a trend toward PABP superiority in reducing left ventricular SW (p = 0.08). An optimal PABP deflation timing occurs during systole (25 milliseconds after the R-wave) and can minimize coronary regurgitation.

Wave intensity analysis of para-aortic counterpulsation

Wave intensity analysis (WIA) was used to delineate and maximize the efficacy of a newly developed para-aortic blood pump (PABP). The intra-aortic balloon pump (IABP) was employed as the comparison benchmark. Acute porcine experiments using eight pigs, randomly divided into IABP (n = 4) and PABP (n = 4) groups, were conducted to compare the characteristics of intra- and para-aortic counterpulsation. We measured pressure and velocity with probes installed in the left anterior descending coronary artery and aorta, during and without PABP assistance. Wave intensity for aortic and left coronary waves were derived from pressure and flow measurements with synchronization correction applied. To achieve maximized support efficacy, deflation timings ranging from 25 ms ahead of to 35 ms after the R-wave were tested. Similar to those associated with IABP counterpulsation, the PABP-generated backward-traveling waves predominantly drove aortic and coronary blood flows. However, in contrast with IABP counterpulsation, the nonocclusive nature of the PABP allowed systolic unloading to be delayed into early systole, which resulted in near elimination of coronary blood steal without diminution of systolic left ventricular ejection wave intensities. WIA can elucidate subtleties among different counterpulsatile support means with high sensitivity. Total accelerating wave intensity (TAWI), which was defined as the sum of the time integration of accelerated parts of the positive and negative wave intensities, was used to quantify counterpulsation efficacy. In general, the larger the TAWI gain, the better the counter-pulsatile support efficacy. However, when PABP deflation timings were delayed to after the R-wave, the TAWI was found to be inversely correlated with coronary perfusion. In this delayed deflation timing setting, greater wave cancellation occurred, which led to decreased TAWI but increased coronary perfusion attributed to blood regurgitation reduction.

Numerical simulation of trailing-edge acoustic/vortical interaction

The objective or the present research is twofold. The first part concerns the construction or a high-resolution aeroacoustic flow solver, and the remaining emphasizes the wave/vortex interaction around a sharp trailing edge. Euler equations and the Osher-Chakravarthy MUSCL-type high-resolution upwind TVD scheme were used, respectively, as the flow model and the numerical algorithm to analyze the acoustically excited flow. Modification on the reconstruction of the cell interface values was first made to improve the scheme fidelity so that a wave propagationproblem can be solved on nonuniform mesh systems. An acoustic source modeling was also devised to simulate the generation of sound emitted from a monopole located on the solid boundary. The numerical algorithm was first evaluated by checking the computed results with several test problems that have analytic solutions. Results show that the currently proposed computational aeroacoustic scheme is accurate and reliable. An acoustically excited incompressible and low-Mach-number flow over a finite plate was then simulated. The results show that the unsteady airloads induced by acoustic/vortical interaction around a sharp trailing edge can be satisfactorily resolved by the currently developed inviscid Euler flow solver

An Evaluation of Engine Faults Diagnostics Using Artificial Neural Networks

Application of artificial neural network (ANN)-based method to perform engine condition monitoring and fault diagnosis is evaluated. Back-propagation, feedforward neural nets are employed for constructing engine diagnostic networks. Noise-contained training and testing data are generated using an influence coefficient matrix and the data scatters. The results indicate that under high-level noise conditions ANN fault diagnosis can only achieve a 50–60% success rate. For situations where sensor scatters are comparable to those of the normal engine operation, the success rates for both 4-input and 8-input ANN diagnoses achieve high scores which satisfy the minimum 90% requirement. It is surprising to find that the success rate of the 4-input diagnosis is almost as good as that of the 8-input. Although the ANN-based method possesses certain capability in resisting the influence of input noise, it is found that a preprocessor that can perform sensor data validation is of paramount importance. Auto-associative neural network (AANN) is introduced to reduce the noise level contained. It is shown that the noise can be greatly filtered to result in a higher success rate of diagnosis. This AANN data validation preprocessor can also serve as an instant trend detector which greatly improves the current smoothing methods in trend detection. It is concluded that ANN-based fault diagnostic method is of great potential for future use. However, further investigations using actual engine data have to be done to validate the present findings.Copyright