Search

The inverse synthetic aperture radar (ISAR) images are often afflicted by boundary-blurring, discontinuity, sidelobe effects of strong scattering points, a large dynamic range of gray values, and azimuth defocus, which pose significant challenges to image segmentation. This letter proposes a novel semantic segmentation method for ISAR images of space targets. The method is based on contrastive learning (CL) and nonlocal Unet (NL-Unet). First, the method roughly segments the target contour using binary semantic tags to remove sidelobe interference and image noise. Then, the nonlocal (NL) self-attentive mechanism with a global perceptual field is used to exploit the structural symmetry of the ISAR image. Finally, to improve the segmentation ability of target small parts, the method adopts a training method based on CL to overcome the relatively weak problem of the supervised learning model. The proposed method outperforms existing methods on the simulated ISAR dataset. Moreover, it is practical and can be directly generalized to the real-measured dataset without retraining. [ABSTRACT FROM AUTHOR]

From technical and economical viewpoints, an all‐dc offshore wind farm (OWF) with a high‐voltage direct‐current (HVDC) connection, which is also known as HVDC‐connected all‐dc OWF, is the future trend for offshore wind energy applications. However, due to the decoupling effect of power converters, all‐dc OWF cannot directly provide short‐term frequency support for the onshore ac gird, that is, primary frequency response and inertia support. To address this issue, this paper presents a distributed model predictive control (DMPC) scheme for all‐dc OWF. By directly suppressing the voltage deviation and the rate of change of voltage of the offshore dc collection network, this scheme indirectly decreases the frequency deviation and the rate of change of frequency (RoCoF) of the onshore ac grid. Meanwhile, the scheme ensures the stability of wind turbines in all‐dc OWF during the frequency events. Considering a large number of wind turbines in the OWF, the corresponding optimization problem is solved in a distributed way using the alternating direction method of multipliers (ADMM), thus reducing the computational burden. Simulations including the performance comparison with droop control validate the effectiveness of this scheme. [ABSTRACT FROM AUTHOR]

Porous silicon multilayer structures have remarkable optical and morphological properties that can be exploited for biosensing. In particular, a high internal surface area (> 100 [m.sup.2]/[cm.sup.3]) and a linear response profile to changes in the dielectric environment enable fabrication of sensitive devices and a straightforward quantitation of the optical response. These essential operating characteristics are illustrated for p+ mesoporous silicon (pore diameter 15-20 nm) optical microcavities. A series of devices were prepared to permit the immobilization of glutathione-S-transferase (~50 kDa) within the porous matrix. Enzyme activity was exploited as an indirect means to quantitate the amount of protein immobilized. Activity was positively correlated with the optical sensor response. However, at high enzyme load the activity becomes nonlinear while the microcavity response remains linear. These data were used to determine the transduction limit (minimum amount of protein required to transduce an optical response), which is reported as areal mass sensitivity ranging between 50 and 250 pg/[mm.sup.2]. This value is considered in context with the dynamic range of the bulk sensitivity, defined as the magnitude of the wavelength shift per refractive index unit, which was measured as a function of microcavity design parameters. This work has uncovered key parameters that can be tuned to improve the detection limit of this sensor modality. Because of the ever increasing number of emerging new biosensor technologies, defining sensor detection limits has become an ambiguous topic and a need exists to standardize measurements and sensitivity units. For chip-based devices, it seems appropriate to report sensitivity in terms of the minimum number of grams of bound target per surface area.