ID 原文 译文
9104 介绍了故障安全分析演示平台所涉及的关键技术实现方法,包括虚拟样机的建立与验证、基于故障仿真的故障模式与影响分析、基于蒙特卡罗仿真的概率风险分析以及基于广义回归神经网络的关键参数预警,并采用多种软件环境集成构建了演示平台。 Introduced fail-safe analysis demonstration method the key technology involved, including the establishment of the virtual prototype and validation, based on the fault simulation of failure mode and effect analysis, probability risk analysis based on monte carlo simulation, and key parameters of early warning based on generalized regression neural network, and use a variety of software environment integration platform is constructed.
9105 操舵系统的案例研究结果表明,所提出的技术方法能够有效缓解历史故障数据不足的问题,演示平台能够满足机电液系统故障安全分析的需要。 Steering system of the case study results show that the proposed method can effectively alleviate the problem of insufficient historical failure data, the demo platform can meet the needs of the mechanical and electrical hydraulic system failure safety analysis.
9106 为了充分利用稀疏表示分类算法中重构残差包含的特征信息,将重构残差的波段信息反馈到测试样本中,自适应增强样本的稀疏特征提取。 In order to make full use of sparse representation classification algorithm of reconstructing the characteristic information of the residual contains will reshape the residual band information feedback to the test sample, sample adaptive enhancement of sparse feature extraction.
9107 但反馈调整过程可能会出现特征过拟合的问题,为了进一步提高算法的稳定性和分类精度,提出了紧耦合像元生成算法(close coupled set of pixels,CCSP)来平滑特征分布以解决过拟合问题,并最终提出了基于紧耦合像元的自适应增强类内稀疏表示高光谱图像分类方法(close coupled set of pixels-based adaptive boosting class-wise sparse representation classifier,CCSP-ABCWSRC)。 But feedback adjustment process characteristics may arise from the problem of fitting, to further improve the stability and classification accuracy of the algorithm, proposed the tightly coupled as a yuan generation algorithm (close coupled is the set of pixels, CCSP) to smooth the characteristics of distribution, in order to solve the problem, a fitting and eventually was proposed based on adaptive enhancement class tightly coupled like yuan sparse representation in hyperspectral image classification method (close coupled is the set of pixels - -based adaptive boosting class - wise sparse representation classifier, CCSP - ABCWSRC).
9108 在Indian Pines,University of Pavia,Salinas三个高光谱数据集上的实验结果表明,提出的算法对高光谱图像进行了稳定有效的分类并且其分类精度优于同类算法。 In Indian Pines, the University of Pavia, goodness in three hyperspectral data sets on the experimental results show that the proposed algorithm for classification of hyperspectral image has carried on the stable and effective and its classification precision is superior to the similar algorithm.
9109 研究了在进行多目标跟踪时机会数字阵列雷达(opportunistic digital array radar,ODAR)的功率资源管理问题。 Studied the opportunity digital array radar in multiple target tracking (opportunistic digital array radar, ODAR) power resource management problems.
9110 针对复杂多变的环境和未知的目标信息所导致的不确定性,建立了基于随机和模糊机会约束规划(chance-constraint programming,CCP)的多目标稳健功率资源管理模型。 In view of the complicated and changeable environment and uncertainty, as a result of the unknown target information, based on random and fuzzy chance constrained programming (chance - the constraint programming, CCP) multi-objective robust power resource management model.
9111 模型引入随机变量表征雷达总发射功率,引入模糊变量表征每个目标的RCS,以贝叶斯克拉美罗界(Bayesian Cramer Rao lower bound,BCRLB)作为目标跟踪精度的衡量标准。 Model introduced stochastic variable to characterize radar total transmit power, the introduction of fuzzy variables represent each target RCS, by luo bei Ye Sike Latin America (Bayesian Cramer Rao lower bound, BCRLB) as a measure of target tracking precision.
9112 将随机模拟和模糊模拟都嵌入到遗传算法(genetic algorithm,GA)当中,从而预测出下一时刻满足给定置信水平的各目标最优的功率分配,然后根据求解出来的功率分配情况,利用无迹卡尔曼滤波器(unscented Kalman filter,UKF)进行目标跟踪。 Stochastic simulation and fuzzy simulation is embedded into the genetic algorithm (based algorithm, GA), so as to predict the next time meet the goal of a given confidence level, the optimal power allocation, and then, according to the solving of power allocation situation of no trace Kalman filter (unscented Kalman filter, UKF) for target tracking.
9113 最后,通过仿真实验验证了算法的有效性和稳定性。 At last, the validity of the algorithm is verified by simulation experiment and stability.