Search

Due to its advantages of small size and light weight, multi-pulse rectifier has a broad application prospect in the field of aviation. The 18-pulse rectifier is more popular in practical engineering application because of its better rectifier characteristics than 12-pulse, its simpler structure process and easier to realize than 24-pulse. The theoretical research and engineering application of the 18-pulse rectifier have achieved some success. However, whether the reliability of the system can be satisfied has always been one of the key problems to be solved in practical engineering. The aim of this work is to study the reliability modeling and prediction of the 18-pulse rectifier power supply. Its reliability block diagram and mathematical model are established. According to US Military handbook MIL-HDBK-217F, the most authoritative and widely used in the world, the two most commonly used methods for reliability prediction are parts count method and part stress analysis method. The reliability data are estimated by these two methods. It is concluded that the value of mean time between failures in airborne environment is 23326 h, which meets the requirement of 1.5 times of 15000 h. According to the reliability data, the failure mode and effect analysis of its weak links are carried out, and the improvement measures are proposed. This will promote and guide the reliability growth and engineering popularization of the 18-pulse rectifier power supply. At the same time, the reliability of the actual project can be extended to other related engineering applications to ensure the reliable operation of the system.

Developments in defect descriptors and computer vision-based algorithms for automatic optical inspection (AOI) allows for further development in image-based measurements. Defect classification is a vital part of an optical-imaging-based surface quality measuring instrument. The high-speed production rhythm of hot continuous rolling requires an ultra-rapid response to every component as well as algorithms in AOI instrument. In this paper, a simple, fast, yet robust texture descriptor, namely selectively dominant local binary patterns (SDLBPs), is proposed for defect classification. First, an intelligent searching algorithm with a quantitative thresholding mechanism is built to excavate the dominant non-uniform patterns (DNUPs). Second, two convertible schemes of pattern code mapping are developed for binary encoding of all uniform patterns and DNUPs. Third, feature extraction is carried out under SDLBP framework. Finally, an adaptive region weighting method is built for further strengthening the original nearest neighbor classifier in the feature matching stage. The extensive experiments carried out on an open texture database (Outex) and an actual surface defect database (Dragon) indicates that our proposed SDLBP yields promising performance on both classification accuracy and time efficiency.

Solid-state disks (SSDs) tend to replace traditional motor-driven hard disks in high-end storage devices in past few decades. However, various inherent features, such as out-of-place update [resorting to garbage collection (GC)] and limited endurance (resorting to wear leveling), need to be reduced to a large extent before that day comes. Both the GC and wear leveling fundamentally depend on hot data identification (HDI). In this paper, we propose a hot data-aware flash translation layer architecture based on a dynamic virtual page (DVPFTL) so as to improve the performance and lifetime of nand flash devices. First, we develop a generalized dual layer HDI (DL-HDI) framework, which is composed of a cold data preclassifier and a hot data post-identifier. Those can efficiently follow the frequency and recency of information access. Then, we design an adaptive parallelism manager (APM) to assign the clustered data chunks to distinct resident blocks in the SSD so as to prolong its endurance. Finally, the experimental results from our realized SSD prototype indicate that the DVPFTL scheme has reliably improved the parallelizability and endurance of nand flash devices with improved GC-costs, compared with related works.

Accurately evaluating channel state information (CSI) is an extremely important precondition for wireless communication to effectively obtain exact sending data. In order to fast obtain CSI, channel prediction tends to get its popularity in obtaining CSI as a result of the advantages of lightweight calculation and negligible feedback delay. In this paper, a jointly optimized extreme learning machine (JOELM) is proposed for the short-term prediction of fading channel. The JOELM scheme consists of two key steps: intelligent optimization and targeted repair. First, in order to gain high prediction accuracy, the firefly algorithm is imported to intelligently optimize the traditional extreme learning machine. Second, the Savitzky-Golay filter is innovatively adopted for reducing potential prediction errors. Extensive experiments about computational complexity, influences of repair coefficient and weight coefficient, contributing degree of two key steps, amplitude, transmission bit/symbol error rates, root-mean-square errors, and four typical statistical properties are given in final simulation section. The analyzed results indicate that the proposed JOELM can more accurately and efficiently deal with the short-term prediction of fading channel.