基于支持向量机的矿井通风机故障诊断系统的研究

The study of fault diagnosis system of mine ventilator based on support vector machine

针对矿井通风机故障诊断过程中样本数据有限的特点,本文提出了一种基于支持向量机的矿井通风机故障诊断方法。支持向量机是一种机器学习方法,它对有限的样本数据具有良好的学习能力。采用支持向量机对样本数据通过学习形成故障分类器,利用该分类器对故障样本进行测试,得出矿井通风机的故障诊断结果。结果表明:支持向量机对有限样本数据具有良好的推广能力,最终达到了快速并准确地诊断矿井通风机故障的目的。

Abs Aimed at the limited sample data in the process of mine ventilator fault diagnosis, this paper proposes a fault diagno- sis method of mine ventilator based on support vector machine. Support vector machine is a machine learning approach, which has a good learn ability to the limited sample data. Using support vector machine form fault classifiers by learning through the reduced da- ta, then obtained fault diagnosis results of mine ventilator by using the classifier to test the fault samples. The results show that sup- port vector machine has good promotion ability to small samples, it achieving ultimately the purposes of rapid and accurate diagno- sis the mine fan faults.