多元回归分析模型优化反导威胁评估  

Multiple Regression Analysis Model for Optimization of Antimissile Threat Assessment

作　　者：王莎[1] 张多林[1] WANG Sha ZHANG Duo-lin(AFEU, Air and Missile Defense College, Shaanxi Xi＇ an 710051, China)

机构地区：[1]空军工程大学防空反导学院,陕西西安710051

出　　处：《现代防御技术》2017年第2期118-123,共6页Modern Defence Technology

摘　　要：反导威胁评估是信息化条件下反导指挥控制领域的一项重要研究内容。在信息网络基础上,准确、高效地评判出多个来袭弹道导弹的威胁程度是亟需解决的问题。首先针对反导威胁评估的多属性问题,从物理域、信息域、认知域、空间态势4个角度考虑威胁评估的影响因素,建立反导威胁评估指标体系,然后以基于弹道目标威胁度评估的多元回归模型为基础,采用PSO优化算法实现对回归模型的参数估计,将最优参数带入回归模型中输出反导威胁度。仿真表明,该模型与PSO优化算法具有较好的准确性和时效性。Antimissile threat assessment is an important research content in the field of the informati- zation antimissile command and control. On the basis of information network, it is urgently needed to solve problems such as accurately and efficiently evaluating multiple incoming ballistic missile threat lev- el. At first, in view of the multi attributes of missile threat assessment, starting from four aspects as the physical domain, information domain, cognition domain, space situation to consider the influence factors of threat assessment, an anti-ballistic missile threat assessment index system is established. Then on the basis of multivariate regression model of ballistic target threat assessment, the regression model parame- ters are estimated by using the PSO optimization algorithm, the missile threat level is obtained by input- ting the optimal parameters. The simulation results show that the model and the PSO algorithm has better accuracy and timeliness.

关 键 词：反导作战 威胁评估 指标体系 多元回归分析 参数估计 PSO优化算法 

分 类 号：TJ761.7[兵器科学与技术—武器系统与运用工程] TP301.6[自动化与计算机技术—计算机系统结构]