Abstract:The installation of stationary super-capacitor energy storage system (ESS) in metro systems can recycle the vehicle braking energy and improve the pantograph voltage profile. This paper aims to optimize the energy management, location, and size of stationary super-capacitor ESSes simultaneously and obtain the best economic efficiency and voltage profile of metro systems. Firstly, the simulation platform of an urban rail power supply system, which includes trains and super-capacitor energy storage systems, is established. Then, two evaluation functions from the perspectives of economic efficiency and voltage drop compensation are put forward. Ultimately, a novel optimization method that combines genetic algorithms and a simulation platform of urban rail power supply system is proposed, which can obtain the best energy management strategy, location, and size for ESSes simultaneously. With actual parameters of a Chinese metro line applied in the simulation comparison, certain optimal scheme of ESSes' energy management strategy, location, and size obtained by a novel optimization method can achieve much better performance of metro systems from the perspectives of two evaluation functions. The simulation result shows that with the increase of weight coefficient, the optimal energy management strategy, locations and size of ESSes appear certain regularities, and the best compromise between economic efficiency and voltage drop compensation can be obtained by a novel optimization method, which can provide a valuable reference to subway company. OPEN ACCESSEnergies 2015, 8 11619

Although remarkable progress has been made in salient object detection (SOD) in natural scene images (NSI), the SOD of optical remote sensing images (RSI) still faces significant challenges due to various spatial resolutions, cluttered backgrounds, and complex imaging conditions, mainly for two reasons: (1) accurate location of salient objects; and (2) subtle boundaries of salient objects. This paper explores the inherent properties of multi-level features to develop a novel semantic-guided attention refinement network (SARNet) for SOD of NSI. Specifically, the proposed semantic guided decoder (SGD) roughly but accurately locates the multi-scale object by aggregating multiple high-level features, and then this global semantic information guides the integration of subsequent features in a step-by-step feedback manner to make full use of deep multi-level features. Simultaneously, the proposed parallel attention fusion (PAF) module combines cross-level features and semantic-guided information to refine the object’s boundary and highlight the entire object area gradually. Finally, the proposed network architecture is trained through an end-to-end fully supervised model. Quantitative and qualitative evaluations on two public RSI datasets and additional NSI datasets across five metrics show that our SARNet is superior to 14 state-of-the-art (SOTA) methods without any post-processing.

No abstract

No abstract

No abstract

Abstract-Naïve Bayes classifiers is widely used for text classification because of its simplicity and effectiveness. In this paper, an improved Naïve Bayes classifiers was proposed, using multinomial model to modify its rough parameter estimation and parallel competing with MapReduce to categories to text documents. The experimental results show that the proposed method is able to improve the accuracy of Naïve Bayes classifiers dramatically, and has good scalability and extensibility for large-scale text classification.

No abstract

A new iterative algorithm to be used to precisely reconstruct near-field distribution from an interferogram of a laser output generated by a cyclic radial-shearing interferometer is proposed. First, by use of a window function around the zero-frequency part of the Fourier transform of the interferogram and calculation of the inverse Fourier transform of the zero-frequency part, we obtain the background intensity distribution of the interferogram. Then, according to the iterative algorithm, the near-field distribution of the laser output is precisely reconstructed from the background intensity distribution obtained in the first step. A numerical simulation and an actual experiment of the near-field reconstruction of the laser output with arbitrary amplitude distribution are implemented successfully.