采用SVM的磨粒分类识别方法研究

Debris recognition method based on multi-support vector machine

为克服传统磨粒识别分类器训练时需要大量特征样本的缺点,设计一种基于多元支持向量机(Multi-Support Vector Machine,Multi-SVM)的磨粒识别分类器。支持向量机(SVM)是一种新的机器学习方法,在小样本和高维二元分类方面有非常突出的优点。实验证明,依据此优点设计的多元支持向量机磨粒分类器模型,不仅可以在小样本情形下对模型进行快速训练,而且可以快速识别多种磨粒类型,同时识别率也比传统的神经网络方法有较大提高,从而达到了提高设备监测和故障诊断效率的目的。

For solving the defect of traditional classificatory with many samples, a new classificatory recognizing debris based on Multi-SVM （ Multi-Support Vector Machine） is proposed. SVM is a new machine study method which has excellent advantages in small-sample and multi-dimension binary classification. The new Multi-SVM classificatory can be studied in few samples rapidly to recognize several kinds of new debris. At the same time, the experiments showed that recognizing correct rate increased more greatly compared with traditional BP method.