磷虾群算法优化支持向量机的威胁估计

发布时间：2018-03-16 18:20

本文选题：威胁估计　切入点：磷虾群　出处：《光学精密工程》2016年06期 　论文类型：期刊论文

【摘要】：为提高目标威胁估计的预测精度,在传统支持向量机优化方法的基础上,提出了采用磷虾群算法优化支持向量机的威胁估计方法。介绍了磷虾群算法和支持向量机的原理,并基于此采用磷虾群算法对支持向量机中的惩罚参数和核函数参数进行优化,寻找最优的惩罚参数和核函数参数;建立磷虾群优化支持向量机的目标威胁估计模型,并实现基于该模型的目标威胁估计算法。采集90组原始数据组成训练集、30组数据组成测试集,对该目标威胁估计算法进行仿真实验。实验结果显示,磷虾群算法优化支持向量机的预测误差为0.002 91,小于采用粒子群算法或萤火虫算法优化的支持向量机。结果表明,磷虾群优化支持向量机的目标威胁估计方法可以有效地完成目标威胁估计。
[Abstract]:In order to improve the prediction accuracy of target threat estimation, based on the traditional support vector machine (SVM) optimization method, a threat estimation method using krill swarm algorithm to optimize support vector machine (SVM) is proposed, and the principle of krill swarm algorithm and support vector machine (SVM) is introduced. Based on this, the penalty parameters and kernel function parameters in support vector machine are optimized by using the krill swarm algorithm, the optimal penalty parameters and kernel function parameters are found, and the target threat estimation model of krill swarm optimization support vector machine is established. The target threat estimation algorithm based on this model is implemented. 90 groups of raw data are collected to form a training set and 30 groups of data constitute a test set. The simulation experiment of the target threat estimation algorithm is carried out. The experimental results show that, The prediction error of krill swarm optimization support vector machine is 0.002.91, which is smaller than that of particle swarm optimization or firefly algorithm. The target threat estimation method based on optimal support vector machine (SVM) for krill population can effectively accomplish the target threat estimation.
【作者单位】：中国科学院长春光学精密机械与物理研究所;中国科学院大学;
【基金】：国家自然基金青年基金(No.61205143)
【分类号】：TP18

【相似文献】