基于模糊粒度相关向量机的缓变故障参数区间预测  

Interval Prediction of Gradual Fault Parameter Based on Fuzzy Granular RVM

作　　者：曲晓燕[1] 张勇亮[2] 吕晓峰[1] 马羚[1] 

机构地区：[1]海军航空工程学院兵器科学与技术系,山东烟台264001 [2]海军装备研究院,北京100161

出　　处：《计算机与现代化》2017年第8期5-12,共8页Computer and Modernization

摘　　要：为解决相关向量机(RVM)在多样本故障预测中存在的预测精度下降和运算效率低等不足,提出一种基于模糊粒度相关向量机(FGRVM)的缓变故障参数区间预测方法。首先,对参数初始时间序列进行模糊信息粒化(FIG),并采用基于指数相似度的约简方法对训练集中的冗余数据进行约简;然后,使用自适应极值扰动和自适应变异对简化粒子群算法(ts PSO)进行改进,并以4折交叉验证误差最小为优化目标,采用该粒子群算法(its PSO)实现相关向量机核宽度的自适应选择;最后,采用训练好的RVM预测缓变故障参数的变化区间。以3个经典的大规模、非线性、带噪声的时间序列及国航某航空发动机排气温度变化量(DEGT)为例对该方法的预测性能进行验证,仿真结果表明,相对于RVM和模糊粒度支持向量机(FGSVM),该方法在预测精度和运算效率上都有较大的提高。To solve the shortages of Relevance Vector Machine（ RVM） in fault prediction for large-scale samples,such as the fall of prediction accuracy,low computation efficiency and so on,a method of Fuzzy Granular RVM（ FGRVM） used for interval prediction of gradual fault parameters was proposed. Firstly,the original time series of parameters were granulated by fuzzy information granulation（ FIG）,and then the redundant data in the training set was reduced by a reduction approach based on the exponent similarity. Secondly,the Simple Particle Swarm Optimization（ ts PSO） algorithm was improved by adaptive extremum disturbed and adaptive mutation,and the kernel width of RVM was selected automatically by the improved ts PSO（ its PSO） algorithm with the optimization target to minimize the error of 4-fold cross validation. Finally,the changeable intervals of gradual fault parameters were predicted by the well-trained RVM. The prediction performance of the method was proved by simulations on three classic large-scale,nonlinear time series with noisy data and Delta Exhaust Gas Temperature（ DEGT） of an aero-engine in Air China,and the experiment results show that,compared with RVM and Fuzzy Granular Support Vector Machine（ FGSVM）,the prediction accuracy and computation efficiency are improved greatly by the method proposed in the paper.

关 键 词：缓变故障 区间预测 模糊信息粒化 相关向量机 约简 粒子群算法 

分 类 号：V263.6[航空宇航科学与技术—航空宇航制造工程] TP206.3[自动化与计算机技术—检测技术与自动化装置]