基于 BWO-VMD-FE 与 ELM的往复压缩机气阀故障诊断研究
Research on Fault Diagnosis of the Reciprocating Compressor Valve Based on BWO-VMD-FE and ELM
黄欣悦 1别锋锋 1李倩倩 1缪新婷 1黄文庆 1邢雨1
作者信息
- 1. 常州大学 机械与轨道交通学院,江苏常州 213164;江苏省绿色过程装备重点实验室,江苏常州 213164
- 折叠
摘要
如何提取往复压缩机气阀故障信号中的冲击特征并识别蕴含在内的非线性与非平稳特性,一直是本研究领域的热点和难点.为了有效提取往复压缩机振动信号中的特征信息,本文提出了一种基于改进型变分模态分解(Variational Mode Decomposition,VMD)和模糊熵(Fuzzy Entropy,FE)结合的特征提取方法.首先,利用白鲸优化算法(Beluga Whale Optimization,BWO)对变分模态分解算法进行优化,自适应优选其核心参数(模态个数K和惩罚因子α),由此针对往复式压缩机气阀振动信号分解得到K个模态分量(Intrinsic Mode Function,IMF),将IMFs进行信号重构,通过模糊熵值分析构成特征向量集,最后通过构建的极限学习机(ELM)分类模型完成气阀工况模式识别.数值仿真和实验模拟研究结果表明,该方法具有良好的有效性、适用性及准确性.
Abstract
Extracting the shock features in the fault signals of reciprocating compressor valves and identifing the non-linear and non-stationary characteristics involved have been the hot and difficult issues in this research field.In order to effectively extract the characteristic information in the reciprocating compressor signal,a feature extraction method based on improved variational mode decomposition(VMD)and fuzzy entropy(FE)is proposed in this paper.Firstly,the beluga whale optimization(BWO)algorithm is used to optimize the variational mode decomposition algorithm,and the core parameters(number of modes Kand penalty parameters α)are adaptively optimized,on the basis,the reciprocating compressor air valve vibration signal is decomposed into K intrinsic mode functions(IMFs).Then the IMFs are reconstructed into a new signal,and the feature vector set is formed through fuzzy entropy analysis.Finally,the air valve working condition mode recognition is completed by the constructed extreme learning machine(ELM)classification model.Numerical and experimental simulation results show that the proposed method has good effectiveness,applicability and accuracy.
关键词
往复式压缩机/变分模态分解/模糊熵/极限学习机/模式识别Key words
reciprocating compressor/variational mode decomposition(VMD)/fuzzy entropy(FE)/extreme learning machine(ELM)/mode recognition引用本文复制引用
出版年
2024
NETL
NSTL
万方数据