基于VPMELM的滚动轴承劣化状态辨识方法

摘要

Aiming at the problem that only four simple mathematical models in the variable predictive mode based class discriminate(VPMCD) method can not reflect complex relationships among eigenvalues,it is found that the extreme learning machine (ELM) regression model,a complex and widely used one,can reflect relationships among eigenvalues.Here,combining with the advantage of ELM regression model and VPMCD method,a variable predictive mode-based extreme learning machine (VPMELM) method was proposed.It was applied to identify the deterioration state of rolling bearings.The test results showed that the identification method based on VPMELM can effectively to identify the deterioration state of rolling bearings.%针对变量预测模型模式识别方法(VPMCD)仅仅包含几种简单数学模型的问题,所建立的预测模型不足以反映特征值之间的复杂关系;极限学习机(ELM)回归模型是一种复杂且被广泛应用的模型,其模型可以反映特征之间的相互关系.结合极限学习机回归模型和VPMCD方法的优点,提出了一种基于极限学习机的变量预测模型(VPMELM)模式识别方法,并将该方法应用于滚动轴承劣化状态实验中.实验表明,基于VPMELM的辨识方法可以有效地对滚动轴承的劣化状态进行识别.