ATSP:时态网络可视化的自适应时间片划分方法

现有的时态网络可视化方法大多采用等量时间片来可视化网络的演变，不利于时态模式的快速挖掘和发现。为此，根据时态网络固有的特征提出自适应时间片划分方法（Adaptive Time Slice Partition method，ATSP）。在时态网络的两种表示方式（基于事件的表示方式和基于快照的表示方式）的基础上，构建了ATSP的基础模型，同时提出了一种改进模型用来描述事件间隔时间服从长尾分布的时态网络。为了实现时间片的不等量划分，针对探索任务的不同提出了基于时态模式的ATSP规则和基于中心节点的ATSP规则，并提出了实现算法--层次划分算法（Hierarchical Partition algorithm，HP）和增量划分算法（Incremental Partition algorithm，IP）。实验结果表明，ATSP方法比传统的时间片划分方法更能准确地表示网络的时态特征，且该方法应用于可视化时，能有效归纳并展示网络的特征，明显提高了视觉分析的效率。     

基于神经网络算法的井下裂缝诊断与堵漏技术

在钻井过程中，常常钻遇不同宽度的井下地层裂缝。钻遇裂缝时容易发生钻井液漏失现象，甚至发生钻井液失返现象，严重影响了安全、高效钻井。目前裂缝封堵的方法常存在封堵成功率不高、堵漏承压能力低的问题，其中一个重要的原因是对井下地层的裂缝宽度等特征认识不清。基于地层裂缝产生的岩石力学机理，确定影响裂缝宽度关键的6个力学和工程因素，并利用神经网络计算的非线性、大数据特点建立了井下地层裂缝宽度的分析模型，模型包含输入层、输出层和3个隐藏层。通过该模型诊断井下裂缝宽度，提高了计算精度，平均误差仅为2.09%，最大误差为5.88%，解决钻井现场仅凭经验判断裂缝误差较大和依靠成像测井成本较高的问题。同时根据神经网络模型诊断得到的裂缝宽度优化堵漏材料的粒径配比，提高了裂缝内的架桥封堵强度和架桥的稳定性，封堵层的承压能力达到12.8 MPa，反向承压能力达到4.5 MPa。现场堵漏试验最高憋压10 MPa，经过封堵作业后大排量循环不漏，达到了裂缝性地层高效堵漏的目的，堵漏一次成功。      

Impact of the anode operating conditions on the liquid water distribution in the cathode gas diffusion layer

Previous experimental results indicate that the humidification conditions at the anode have an impact on the liquid water distribution in the cathode gas diffusion layer. Numerical simulations are developed to reproduce and analyze this effect. Results consistent with the experimental results are first obtained by playing with the partition coefficients of an advanced pore network model computing the liquid water formation and transfer in the cathode gas diffusion layer (GDL) for a large range of operating conditions. Then, a model for the full anode – cathode assembly is developed by combining the pore network model of the cathode GDL and a 1D model describing the heat and water transfer in the various components of the anode-cathode assembly. This enables one to generalize the dry – wet regime diagram introduced in a previous work by incorporating the effect of the humidity condition at the anode.     

Engineering grain boundary anisotropy to elucidate grain growth behavior in alumina

Current grain growth models have evolved to account for the relationship between grain boundary energy/mobility anisotropy and the five degrees of grain boundary character. However, the role of grain boundary networks on overall growth kinetics remains poorly understood. To experimentally investigate this problem, a highly textured Al₂O₃ was fabricated by colloidal casting in a strong magnetic field to engineer a unique spatial distribution of grain boundary character. Microstructural evolution was quantified and compared to an untextured sample. From this comparison, a prevalence of (0001)/(0001) terminated grain boundaries with anisotropic networks were identified in the textured sample. These boundaries and their networks were found to be driving grain growth at a faster rate than predicted by models. These findings will allow better modelling of grain growth in real systems by experimentally exploring the impact thereon of grain boundary plane anisotropy and relative energy/mobility differences between neighboring boundaries.     

Super-Resolution Reconstruction of Images Based on Microarray Camera

In the field of images and imaging, super-resolution (SR) reconstruction of images is a technique that converts one or more low-resolution (LR) images into a highresolution (HR) image. The classical two types of SR methods are mainly based on applying a single image or multiple images captured by a single camera. Microarray camera has the characteristics of small size, multi views, and the possibility of applying to portable devices. It has become a research hotspot in image processing. In this paper, we propose a SR reconstruction of images based on a microarray camera for sharpening and registration processing of array images. The array images are interpolated to obtain a HR image initially followed by a convolution neural network (CNN) procedure for enhancement. The convolution layers of our convolution neural network are 3×3 or 1×1 layers, of which the 1×1 layers are used to improve the network performance particularly. A bottleneck structure is applied to reduce the parameter numbers of the nonlinear mapping and to improve the nonlinear capability of the whole network. Finally, we use a 3×3 deconvolution layer to significantly reduce the number of parameters compared to the deconvolution layer of FSRCNN-s. The experiments show that the proposed method can not only ameliorate effectively the texture quality of the target image based on the array images information, but also further enhance the quality of the initial high resolution image by the improved CNN.     

面向电信承载网的监控系统设计与实现

面对电信承载网连接的日益增长的海量终端设备，运营商需要结合网络拓扑对终端设备产生的数据进行高效的汇聚统计、异常分析、故障定位处理等操作。针对已有系统存在的操作困难、分析效率低等问题，设计与实现了一个面向电信承载网的高效监控系统，提供实时与离线数据分析和多维可视化分析的能力。对网管、认证、终端等系统及设备采集的数据进行结构化存储，对采集的数据进行拓扑相关性和时间序列方法分析，根据分析结果实现基于动态阈值控制的异常实时告警、定位等操作，并提供多维度可视化分析对网络状态进行实时监控。实际应用结果表明，该系统性能优异，具有良好交互性，能较好地满足承载网运维人员业务分析需求。     

Design of disturbance observer based on adaptive-neural control for large-scale time-delay systems in the presence of actuator fault and unknown dead zone

This article presents an adaptive neural compensation scheme for a class of large-scale time delay nonlinear systems in the presence of unknown dead zone, external disturbances, and actuator faults. In this article, the quadratic Lyapunov–Krasovskii functionals are introduced to tackle the system delays. The unknown functions of the system are estimated by using radial basis function neural networks. Furthermore, a disturbance observer is developed to approximate the external disturbances. The proposed adaptive neural compensation control method is constructed by utilizing a backstepping technique. The boundedness of all the closed-loop signals is guaranteed via Lyapunov analysis and the tracking errors are proved to converge to a small neighborhood of the origin. Simulation results are provided to illustrate the effectiveness of the proposed control approach.     

Multi-resource scheduling for FPGA systems

In modern cloud data centers, reconfigurable devices (FPGAs) are used as an alternative to Graphics Processing Units to accelerate data-intensive computations (e.g., machine learning, image and signal processing). Currently, FPGAs are configured to execute fixed workloads, repeatedly over long periods of time. This conflicts with the needs, proper to cloud computing, to flexibly allocate different workloads and to offer the use of physical devices to multiple users. This raises the need for novel, efficient FPGA scheduling algorithms that can decide execution orders close to the optimum in a short time. In this context, we propose a novel scheduling heuristic where groups of tasks that execute together are interposed by hardware reconfigurations. Our contribution is based on gathering tasks around a high-latency task that hides the latency of tasks, within the same group, that run in parallel and have shorter latencies. We evaluated our solution on a benchmark of 37500 random workloads, synthesized from realistic designs (i.e., topology, resource occupancy). For this testbench, on average, our heuristic produces optimum makespan solutions in 47.4% of the cases. It produces acceptable solutions for moderately constrained systems (i.e., the deadline falls within 10% of the optimum makespan) in 90.1% of the cases.     

Operational time-series data modeling via LSTM network integrating principal component analysis based on human experience

Today’s information technologies involve increasingly intelligent systems, which come at the cost of increasingly complex equipment. Modern monitoring systems collect multi-measuring-point and long-term data which make equipment health prediction a “big data” problem. It is difficult to extract information from such condition monitoring data to accurately estimate or predict health statuses. Deep learning is a powerful tool for big data processing that is widely utilized in image and speech recognition applications, and can also provide effective predictions in industrial processes. This paper proposes the Long Short-term Memory Integrating Principal Component Analysis based on Human Experience (HEPCA-LSTM), which uses operational time-series data for equipment health prognostics. Principal component analysis based on human experience is first conducted to extract condition parameters from the condition monitoring system. The long short-term memory (LSTM) framework is then constructed to predict the target status. Finally, a dynamic update of the prediction model with incoming data is performed at a certain interval to prevent any model misalignment caused by the drifting of relevant variables. The proposed model is validated on a practical case and found to outperform other prediction methods. It utilizes a powerful deep learning analysis method, the LSTM, to fully process big condition monitoring series data; it effectively extracts the features involved with human experience and takes dynamic updates into consideration.