基于SVM后端校正的含噪声机械故障分类方法

NOISY FAULT CLASSIFICATION OF MECHANICAL BASED ON SVM BACK-END CORRECTION

从模式识别角度对含噪声机械故障分类方法进行优化,在支持向量机的基础上提出一种新的一对一分类方法和后端校正方法。讨论了支持向量机的思想及其在机械故障分类中的应用,详细介绍了针对含噪声机械故障样本分类的新的一对一分类方法以及后端校正方法。结果表明,在噪声环境下,该方法能够显著提高机械故障分类精度。

Optimized the method of noisy fault classification of mechanical from the perspective of pattern recognition. Based on support vector machine （SVM） , proposed a new method for noisy fault classification of mechanical： one-versus-one classification and back-end correction. The theory of the SVM is discussed and we highlight the proposed one-versus-one classification and back-end correction method for noisy fault classification of mechanical. It proves that the method can improve the classification accuracy of mechanical fauh classification significantly in the noise environment.