ID	原文	译文
7694	计算机仿真与半实物系统实测结果验证了所提解耦方法的正确性及有效性。	Computer and the hardware-in-the-loop simulation system measured results verify the correctness and validity of the proposed decoupling method.
7695	基于字典学习模型能真实反映雷达高分辨距离像(radar high resolution range profile,HRRP)潜在结构特征和统计建模算法可有效解决HRRP姿态敏感性问题的特点,运用统计建模划分HRRP角域,对鉴别字典的原子选取和判别优化问题开展研究。	Based on dictionary learning model can reflect radar high resolution range like (radar high resolution range profile, HRRP) potential structural features and statistical modeling algorithm can effectively solve the problem of HRRP attitude sensitivity characteristics, use of statistical modeling dividing HRRP Angle domain, to identify the dictionary discriminant optimization selection of atoms and conduct research.
7696	首先提出了基于概率主分量分析的最大概率差值算法,自适应划分HRRP角域获取帧界线。	Based on probabilistic principal component analysis is proposed first maximum probability differential algorithm, adaptive dividing HRRP Angle domain access frame line.
7697	其次,利用帧界线对应功率谱特征构成初始化鉴别字典,在鉴别字典基础上优化判别准则,引入原子稀疏相似误差约束最优字典更新实现测试样本分类。	Secondly, using the frame line corresponding power spectrum characteristics constitute the initialization identify dictionary, on the basis of identifying the dictionary optimization criterion, the introduction of atomic sparse similar error constraints, the optimal dictionary update test sample classification.
7698	雷达实测数据的实验结果验证了该算法可提高目标识别率,同时对噪声干扰具有很好的鲁棒性。	Radar measured data of the experiment results show that the algorithm can improve the target recognition rate, and has good robustness to noise interference.
7699	针对后续备件需求预测误差大的问题,提出一种基于粗糙集理论修正的后续备件指数平滑预测方法。	According to following spare parts demand prediction error is big problem, put forward a kind of modified subsequent spare parts based on rough set theory is exponential smoothing forecast method.
7700	根据备件需求数据呈现的趋势,通过拟合确定指数平滑法的次数和平滑系数。	According to the data showed a trend of the spare parts demand, determined by the number of exponential smoothing and fitting of smoothing coefficient.
7701	从装备在使用过程中影响备件需求数据波动的因素出发,提出了不依赖于基本预测方法的改进预测思路。	From the equipment in use process influence factors of spare parts demand data fluctuation, puts forward the improved prediction method for predicting is not dependent on the basic train of thought.
7702	构建基于粗糙集理论的修正模型。	Build a correction model based on rough set theory.
7703	结合算例,对比分析所提方法的优越性,结果表明修正方法可以显著提高预测精度,提出的改进方法不涉及基本预测方法内部特性且无需引入其他辅助方法,通用性较强。	Combined with examples, comparison and analysis the superiority of the proposed method, the results show that the modified method can significantly improve the prediction accuracy, put forward the improvement method does not involve the basic prediction methods internal features without introducing other auxiliary method, strong commonality.