Array最新文献

At present, the planning of transmission lines mainly relies on human decision-making and lacks intelligence. This paper combines the advantages of GIS in processing spatial data with the advantages of genetic algorithm to explore the optimization method of transmission line planning. The combination of GIS and genetic algorithm can minimize the interference of human factors and quickly solve the path planning problem of transmission lines. According to the theoretical model of genetic algorithm, this study constructs the transmission line optimization model based on genetic algorithm, and realizes the Add-ins plug-in development of the transmission line planning model based on genetic algorithm with the help of C # language. Taking 500 kV overhead transmission line about 150 km from Jiantang Substation (starting point) in Shangri-La County to Tai’ an Substation (ending point) in Lijiang as an example, two groups of experiments are designed under the conditions of considering traffic single factor and comprehensive multi-factor respectively. It is obtained that the path optimization effect of genetic algorithm is the best under the condition of comprehensive multi-factor, which proves the rationality and superiority of the model constructed in this study.

Previous learning-based methods only employ clear images to train the dehazing network, but some useful information such as hazy images, media transmission maps and atmospheric light values in datasets were ignored. Here, we propose a local feature residual network (LFR-Net) for single image dehazing, which is aimed at improving the quality of dehazed images by fully utilizing the information in the training dataset. The backbone of LFR-Net is structured by feature residual block and adaptive feature fusion model. Furthermore, to preserve more details for the recovered clear images, we design an adaptive feature fusion model that adaptively fuses shallow and deep features at each scale of the encoder and decoder. Extended experiments show that the performance of our LFR-Net outperforms the state-of-the-art methods.

Quantum computers have provided a promising tool for tackling NP hard problems. However, most of the existing work on quantum annealers assumes exclusive access to all resources available in a quantum annealer. This is not resource efficient if a task consumes only a small part of an annealer and leaves the rest wasted. We ask if we can run multiple tasks in parallel or concurrently on an annealer, just like the multitasking capability of a classical general-purpose processor. By far, multitasking is not natively supported by any of the existing annealers. In this paper, we explore Multitasking in Quantum Annealer (QAMT) by identifying the parallelism in a quantum annealer from the aspect of space and time. Based on commercialised quantum annealers from D-Wave, we propose a realisation scheme for QAMT, which packs multiple tasks into a quantum machine instruction (QMI) and uses predefined sampling time to emulate task preemption. We enumerate a few scheduling algorithms that match well with QAMT and discuss the challenges in QAMT. To demonstrate the potential of QAMT, we simulate a quantum annealing system, implement a demo QAMT scheduling algorithm, and evaluate the algorithm. Experimental results suggest that there is great potential in multitasking in quantum annealing.

Tissues such as tendons, ligaments, articular cartilage, and menisci contain significant amounts of organised collagen which gives rise to the Magic Angle effect during magnetic resonance imaging (MRI). The MR intensity response of these tissues is dependent on the angle between the main field, $B_0$, and the direction of the collagen fibres. Our previous work showed that by acquiring scans at as few as 7–9 different field orientations, depending on signal to noise ratio (SNR), the tissue microstructure can be deduced from the intensity variations across the set of scans. Previously our Magic Angle Directional Imaging (MADI) technique used rigid registration and manual final alignment, and did not assume any knowledge of the target anatomy being scanned. In the present work, fully automatic soft registration is incorporated into the MADI workflow and a priori knowledge of the target anatomy is used to reduce the required number of scans. Simulation studies were performed to assess how many scans are theoretically necessary. These findings were then applied to MRI data from a caprine knee specimen. Simulations suggested that using 3 scans might be sufficient, but in practice 4 scans were necessary to achieve high accuracy. 5 scans only offered marginal gains over 4 scans. A 15 scan dataset was used as a gold standard for quantitative voxel-to-voxel comparison of computed fibre directions, qualitative comparison of collagen tractography plots are also presented. The results are also encouraging at low SNR values, showing robustness of the method and applicability at low field.

The ability to create low-cost, high-availability, moderate-performance, low-power, sustainable file storage clusters that may be organically distributed throughout an organization would allow organizations to bring data back from cloud-based providers, provide local backup solutions, create local distributed storage pods, and allow remote developing countries to have access to information and other compute resources. The Internet of Things has driven much of the development in low-power ecological systems. The emergence of these devices allowed for the creation of this research project. This research utilized the design science method to create an instantiation of this concept as a demonstrative artifact that could be powered on USB power provided from almost any source. This includes the ability for small solar arrays to provide adequate power to charge the onboard power banks, allowing for continual use over periods of power loss or darkness. This artifact was evaluated using real-time direct download from up to twentyfour workstations. During the course of the research for a period of over approximately 400 days, the artifact performed without interruption. This could be an indication that it may be possible to replace cloud-based storage with organically-distributed sustainable systems for enterprise-level use.