摘要
针对网络故障诊断中的模式识别问题,提出一种基于多重提升(Multi Boost)的优化支持向量机集成学习方法.首先,利用自适应的荷尔蒙调节遗传算法(HMGA),对支持向量机基分类器进行建模参数优化;然后,通过构建Multi Boost集成学习方法将多个基分类器集成,建立以多分类器优化集成为核心的故障诊断系统.实验结果表明,所提出的方法在网络故障诊断中,迭代次数少、建模时间短,并且能够明显提高故障分类的准确率.
For pattern recognition in the network fault diagnosis, an optimal SVM ensemble learning method based on MultiBoost is proposed. Firstly, the parameters of SVM-base-classifier are optimized by using the adaptive hormone modulation genetic algorithm（HMGA）. Then, multi-base-classifiers are integrated by using the MultiBoost algorithm. Finally, with multiple ensemble optimal classifiers as the core, the fault diagnosis system is established. Simulation results show that the proposed method can not only reduce the number of iteration and lower the computing cost, but also improve the fault diagnosis accuracy of the network fault diagnosis system.
出处
《控制与决策》
EI
CSCD
北大核心
2015年第1期81-85,共5页
Control and Decision
基金
国家自然科学基金项目(60974063
61175059)
河北省自然科学基金项目(F2014205115)
河北省高等学校科学技术研究项目(Q2012053)
