摘要
采煤机齿轮传动部分是保障其稳定运转的关键,智能诊断方法各有优缺点,综合多种智能算法的混合智能方法有效保障了预测结果的准确性。研究了故障诊断的基本流程,采用多种特征提取方法建立采煤机齿轮传动的混合智能诊断模型,介绍了混合智能诊断模型中ANFIS分类器的基本结构以及各层之间的传递函数,确定了ANFIS的样本训练过程以及网络诊断流程。通过对比ANFIS和RBF神经网络的迭代次数证明了ANFIS的训练速度更快,对比单一ANFIS分类器和混合智能模型的分类准确率,验证了混合智能模型的预测准确率更高。
The gear transmission part of shearer is the key to ensure its stable operation. The intelligent diagnosis methods have their own advantages and disadvantages. The hybrid intelligent method combining multiple intelligent algorithms effectively guarantees the accuracy of the prediction results.The basic process of fault diagnosis was studied. The hybrid intelligent diagnosis model of shearer gear transmission was established by multiple feature extraction methods. The basic structure of ANFIS classifier and the transfer function between layers were introduced. The sample training process and network diagnostic process of ANFIS were determined. By comparing the iteration times of ANFIS and RBF neural network, it is proved that the training speed of ANFIS is faster. Compared with the classification accuracy of single ANFIS classifier and hybrid intelligent model, the prediction accuracy of hybrid intelligent model is higher.
作者
田林红
赵阳
Tian Linhong;Zhao Yang(Henan Polytechnic Institute,Nanyang 473000,China)
出处
《煤矿机械》
北大核心
2019年第11期155-158,共4页
Coal Mine Machinery
作者简介
田林红(1965-),河南新野人,副教授,硕士,主要从事数控机床故障诊断研究工作,电子信箱:tianlinhong@163.com.
