Influence of film cooling hole angles and geometries on aerodynamic loss and net heat flux reduction

Turbine design engineers have to ensure that film cooling can provide sufficient protection to turbine blades from the hot mainstream gas, while keeping the losses low. Film cooling hole design parameters include inclination angle (a), compound angle (b), hole inlet geometry, and hole exit geometry. The influence of these parameters on aerodynamic loss and net heat flux reduction is investigated, with loss being the primary focus. Low-speed flat plate experiments have been conducted at momentum flux ratios of IR=0.16, 0.64, and 1.44. The film cooling aerodynamic mixing loss, generated by the mixing of mainstream and coolant, can be quantified using a three-dimensional analytical model that has been previously reported by the authors. The model suggests that for the same flow conditions, the aerodynamic mixing loss is the same for holes with different a and b but with the same angle between the mainstream and coolant flow directions (angle k). This relationship is assessed through experiments by testing two sets of cylindrical holes with different a and b: one set with k=35 deg, and another set with k=60 deg. The data confirm the stated relationship between α, β, k and the aerodynamic mixing loss. The results show that the designer should minimize k to obtain the lowest loss, but maximize b to achieve the best heat transfer performance. A suggestion on improving the loss model is also given. Five different hole geometries (α=35.0 deg, β=0 deg) were also tested: cylindrical hole, trenched hole, fan-shaped hole, D-Fan, and SD-Fan. The D-Fan and the SD-Fan have similar hole exits to the fan-shaped hole but their hole inlets are laterally expanded. The external mixing loss and the loss generated inside the hole are compared. It was found that the D-Fan and the SD-Fan have the lowest loss. This is attributed to their laterally expanded hole inlets, which lead to significant reduction in the loss generated inside the holes. As a result, the loss of these geometries is≈50% of the loss of the fan-shaped hole at IR=0.64 and 1.44. © 2013 by ASME.

Origin of transient crosstalk and its reduction in phase-only LCOS wavelength selective switches

The transient crosstalk in a phase-only liquid crystal on silicon (LCOS) based wavelength selective switch using a Fourier transform setup was investigated. Its origin was identified using an in situ test procedure and found to be related to the transient phase patterns displayed by the LCOS device during the switching. Two different methods were proposed to reduce the transient crosstalk without the need to modify the optics or electronics in use. Experimental results show both methods are able to reduce the worst-case transient crosstalk by at least 5 dB. © 1983-2013 IEEE.

Crosstalk reduction in holographic wavelength selective switches based on phase-only LCOS devices

A method to reduce crosstalk is proposed for holographic wavelength selective switches (WSSs) using a customized merit function. A reduction in crosstalk >8 dB is measured when multicasting with a phase-only LCOS device. © OSA 2014.

Linear dimensionality reduction: Survey, insights, and generalizations

© 2015 John P. Cunningham and Zoubin Ghahramani. Linear dimensionality reduction methods are a cornerstone of analyzing high dimensional data, due to their simple geometric interpretations and typically attractive computational properties. These methods capture many data features of interest, such as covariance, dynamical structure, correlation between data sets, input-output relationships, and margin between data classes. Methods have been developed with a variety of names and motivations in many fields, and perhaps as a result the connections between all these methods have not been highlighted. Here we survey methods from this disparate literature as optimization programs over matrix manifolds. We discuss principal component analysis, factor analysis, linear multidimensional scaling, Fisher's linear discriminant analysis, canonical correlations analysis, maximum autocorrelation factors, slow feature analysis, sufficient dimensionality reduction, undercomplete independent component analysis, linear regression, distance metric learning, and more. This optimization framework gives insight to some rarely discussed shortcomings of well-known methods, such as the suboptimality of certain eigenvector solutions. Modern techniques for optimization over matrix manifolds enable a generic linear dimensionality reduction solver, which accepts as input data and an objective to be optimized, and returns, as output, an optimal low-dimensional projection of the data. This simple optimization framework further allows straightforward generalizations and novel variants of classical methods, which we demonstrate here by creating an orthogonal-projection canonical correlations analysis. More broadly, this survey and generic solver suggest that linear dimensionality reduction can move toward becoming a blackbox, objective-agnostic numerical technology.