金字塔渐进融合低照度图像增强网络

针对现有低照度图像增强网络对不同尺度特征信息存在感知与表达能力不足的问题,提出金字塔渐进融合低照度图像增强网络模型。网络对图像进行多次下采样操作以组成特征金字塔,通过在特征金字塔的三个不同分支上加入跳跃连接,将不同尺度的特征图进行相互融合。通过精细恢复模块进一步提取精炼信息,将特征图恢复到正常的光照图像。结果表明,该网络模型不但能有效地提升低照度图像的整体亮度,而且能很好地保持图像中的细节信息和清晰的物体边缘轮廓,同时能够有效地抑制图像中的暗部噪声,使增强后的图像整体画面真实自然。     

注意力机制量化剪枝优化方法

面向基于注意力机制模型的巨大计算和访存开销问题,研究量化和剪枝协同优化的模型压缩技术,提出针对注意力机制中查询、键、值、概率共四个激活值矩阵的对称线性定点量化方法。同时,提出概率矩阵剪枝方法和渐进式剪枝策略,有效降低剪枝精度损失。在不同数据集上的实验结果表明,针对典型基于注意力机制模型BERT,在较低或者没有精度损失的情况下该优化方法可达到4位或8位定点量化、0.93～0.98的稀疏度,大幅度降低模型计算量,为加速量化稀疏模型的推理奠定良好的基础。     

Energy‐efficient scheduling for sustainable manufacturing systems with renewable energy resources

Environmentally friendly energy resources open a new opportunity to tackle the problem of energy security and climate change arising from wide use of fossil fuels. This paper focuses on optimizing the allocation of the energy generated by the renewable energy system to minimize the total electricity cost for sustainable manufacturing systems under time‐of‐use tariff by clipping the peak demand. A rolling horizon approach is adopted to handle the uncertainty caused by the weather change. A nonlinear mathematical programming model is established for each decision epoch based on the predicted energy generation and the probability distribution of power demand in the manufacturing plant. The objective function of the model is shown to be convex, Lipchitz‐continuous, and subdifferentiable. A generalized benders decomposition method based on the primal‐dual subgradient descent algorithm is proposed to solve the model. A series of numerical experiments is conducted to show the effectiveness of the solution approach and the significant benefits of using the renewable energy resources.     

防空兵指挥决策系统及其运行环境的影响

决策者、决策辅助人员和指挥决策信息系统是部队指挥决策系统的基本组成部分,防空兵指挥自动化和决策支持系统则是指挥决策信息系统的一部分,环境因素和决策信息是影响防空兵指挥决策系统的重要因素.从系统的观点阐述了防空兵指挥决策系统与内外环境因素的关系,提出在新技术条件下,系统信息是防空兵指挥决策系统的又一组成要素.从这个观点出发,可以准确、全面地了解和分析系统环境因素对防空兵指挥决策系统影响,最终达到提高防空兵指挥决策效能的目的.     

导航装备寿命周期费用分析

费用在高科技装备中的地位越来越重要.运用寿命周期费用理论对导航装备进行分析,通过实例得到某型导航装备的经济寿命,所得结论对部队的导航装备管理具有一定的参考价值.     

Nonperiodic preventive maintenance for repairable systems

As a complex system with multiple components usually deteriorates with age, preventive maintenance (PM) is often performed to keep the system functioning in a good state to prolong its effective age. In this study, a nonhomogeneous Poisson process with a power law failure intensity is used to describe the deterioration of a repairable system, and the optimal nonperiodic PM schedule can be determined to minimize the expected total cost per unit time. However, since the determination of such optimal PM policies may involve numerous uncertainties, which typically make the analyses difficult to perform because of the scarcity of data, a Bayesian decision model, which utilizes all available information effectively, is also proposed for determining the optimal PM strategies. A numerical example with a real failure data set is used to illustrate the effectiveness of the proposed approach. The results show that the optimal schedules derived by Bayesian approach are relatively more conservative than that for non‐Bayesian approach because of the uncertainty of the intensity function, and if the intensity function are updated using the collected data set, which indicates more severe deterioration than the prior belief, replacing the entire system instead of frequent PM activities before serious deterioration is suggested. © 2010 Wiley Periodicals, Inc. Naval Research Logistics, 2010     

高炮组网战术信息分发系统(AAANTIDS)作战结构

首先概要分析了高炮组网战术信息分发系统(AAANTIDS)的定义、功能和作战运用设想,据此,针对高炮在未来防空反导作战中的运用方式,具体从未配备指挥自动化系统情形下高炮群组网作战、配备指挥自动化系统情形下高炮团(旅)组网作战、防空导弹与高炮混编组网作战3个方面分析了AAANTIDS的作战结构,为AAANTIDS进一步的概念建模、研制发展提供了总体思路与框架.     

装备保障兵棋动态评估方法研究

针对装备保障兵棋推演中的方案裁决和分析评估问题,对装备保障效能动态评估方法进行了研究.从分析装备保障兵棋要素入手,构建基于兵棋地图的评估模型.该兵棋地图评估模型能够充分利用地图中的静态地理环境信息和动态战场环境信息,对装备保障方案进行综合分析并给出评估结果,为装备保障方案的制定、调整和执行提供依据.     

军用工程机械急造军路机群优化配置研究

针对军用工程机械遂行构筑急造军路任务过程中,面临的多型装备、多种任务、多点同时作业的机群配置难题,基于整数线性规划理论建立了军用工程机械机群优化配置模型,并运用改进的离散粒子群算法得出最优的配置方案.实例分析表明,该机群优化配置方法可以快速、高效地提出军用工程机械机群优化配置方案,为充分发挥军用工程机械的保障能力提供了...     

飞机涡扇发动机振动源信号数目的确定及盲分离

目前使用的大多数盲源分离方法都依赖于观测传感器数量等于信号源数目这样一个基本假设.主要针对传感器数量大于源信号数量情况,对涡扇发动机振动信号盲源分离问题进行研究,先后采用三种算法进行试验,结果证明了奇异值降维方法和基于四阶累积量的源数估计方法的可行性和可靠性.