【摘要】 针对陀螺仪输出样本不平衡和噪声干扰大的特点,提出一种使用DSP实现非平衡FSVM陀螺仪故障诊断方法。系统核心算法FSVM的隶属度函数设计由非平衡特征因子和去噪模糊因子两部分组成,用于解决样本不平衡和噪声干扰大导致SVM分类精度降低的问题。首先,FSVM的训练阶段在上位机上实现,采集陀螺仪的无故障和故障信号,经过小波包特征提取后得到训练样本,然后由FSVM训练得到分类识别函数,最后将分类识别函数移植到DSP硬件平台上实现FSVM的测试应用。速率陀螺故障诊断的实验结果表明,该实现方法不仅提高了故障样本的诊断精度,而且满足系统实时性要求,具有一定的实用价值。更多还原

【Abstract】 According to the features of unbalanced samples and high noise from gyro output signals,an unbalanced FSVMfault diagnosis method of gyro is presented based on DSP.The membership function of FSVMalgorithm of system consists of unbalanced characteristic factor and de-noising fuzzy factor,which is used to solve the classification accuracy reducing problem of SVMresulting from unbalanced samples and high noise. Firstly,the training phase of FSVMis implemented in host computer. The normal and fault signals of gyro are collected,and then the samples are obtained through the method of wavelet package feature extraction. Then,the classification function is transplanted,which is obtained by FSVMtraining model for DSP to achieve the test and recognition phase. The fault diagnosis experiment results of rate gyro shows that the implementation means not only improves the diagnostic accuracy of fault samples,but also meets the real-time requirements. So it has a certain practical value.更多还原