基于相关向量机的故障预测模型仿真研究  

Simulation Research on Fault Prognostic Model Based on RVM

作　　者：吴坤[1] 康建设[1] 曹文斌[1] 

机构地区：[1]军械工程学院装备指挥与管理系,河北石家庄050003

出　　处：《计算机仿真》2015年第11期6-9,41,共5页Computer Simulation

基　　金：国家自然科学基金(61271153)

摘　　要：针对相关向量机模型中的核函数优化问题,在单核模型的基础上,通过构造线性组合核函数的方法,综合全局性核函数和局部性核函数的映射特性,提出一种组合核函数相关向量机预测模型,并以连续测试函数仿真分析了上述模型的预测性能,最后以具有渐变故障趋势的某型雷达发射机速调管监测数据为例验证了模型性能。实验结果表明,与其它支持向量机、灰色理论等方法相比,具有更高的预测精度及可靠性,并有一定的工程应用前景。To solve the optimization problem of kernel function of RVM model, we presented a RVM prediction model of compounding kernels based on the single kernel model, which synthesizes the mapping characteristics of global and local kernel function after constructing linear compounding kernel functions. The performance of the prediction model was validated by simulation analysis of continual testing function and gradually varied radar transmitter monitoring data. The experiment results demonstrate that the proposed method has better prediction accuracy and reliability than support vector machine or grey theory models. And the method can be applied well in various engineering domains.

关 键 词：组合核函数 相关向量机 故障预测 

分 类 号：TP206[自动化与计算机技术—检测技术与自动化装置]