基于LS-SVM算法的实兵对抗演习作战效能评估

Combat Effectiveness Evaluation of Real Combat Exercise Based on LS-SVM

在实战化实兵对抗演习训练中,经验驱动的作战效能评估方法存在主观性强、评估效率低的问题,对于部队提升实战化训练水平、强化实战能力带来不利影响。针对上述问题,提出数据驱动的实兵对抗演习作战效能评估方法,引入最小二乘支持向量机算法(LS-SVM)构建评估模型,通过演习数据对模型进行训练,建立效能指标和作战效能的非线性映射复杂关系。实验表明,基于LS-SVM算法的实兵对抗演习效能评估方法具有更高的效率和精度。

In the actual combat and combat training,experience driven combat effectiveness evaluation method is subjective and the efficiency is lower,which can adversely affect the strength of actual combat training and actual combat capability. Aiming at the problems above,a data-driven performance evaluation method is proposed and the LS-SVM to construct the combat effectiveness evaluation model is introduced. Then the model is trained through historical exercise data to establish the complex relationship between performance index and combat effectiveness. Experiments show the datadriven performance evaluation method based on LS-SVM algorithm has higher efficiency and accuracy.