摘要
柴油发电机在早期故障阶段,故障引起的振动信号往往很微弱,容易被背景噪声和其他正常振动成分掩盖。这些微弱的故障特征难以从复杂信号中分离出来,导致早期故障难以被及时发现。因此,提出基于频谱的柴油发电机振动故障诊断方法。首先针对柴油发电机振动信号增强问题,采用最小熵解卷积方法通过优化信号的熵值,突出信号中的有效成分,抑制噪声干扰;利用自回归线性预测方法依据信号的自回归特性,对信号进行预测和重构,进一步提升信号的质量;借助小波变换保留信号的特征信息。结合特征参数,计算特征与各放电故障类型的马氏距离,实现对柴油发电机振动故障诊断。通过对比实验证明,该方法能够显著提高故障检测的准确性,为开关柜的安全运行提供有力保障。
In the early stages of diesel generators,the vibration signals caused by faults are often weak and easily masked by background noise and other normal vibration components.These weak fault features are difficult to separate from complex signals,making it difficult to detect early faults in a timely manner.Therefore,a frequency spectrum based diesel generator vibration fault diagnosis method is proposed.Firstly,in order to enhance the vibration signal of diesel generators,the minimum entropy deconvolution method is adopted to optimize the entropy value of the signal,highlight the effective components in the signal,and suppress noise interference;Using autoregressive linear prediction methods to predict and reconstruct signals based on their autoregressive characteristics,further improving the quality of the signals;Utilize wavelet transform to preserve the characteristic information of the signal.By combining feature parameters,calculate the Mahalanobis distance between features and various types of discharge faults to achieve vibration fault diagnosis of diesel generators.Through comparative experiments,it has been proven that this method can significantly improve the accuracy of fault detection and provide strong guarantees for the safe operation of switchgear.
作者
罗晓青
Luo Xiao-qing(CNOOC Energy Development Equipment Technology Co.,Ltd.,Zhanjiang Branch,Guangdong Zhanjiang 524000)
出处
《内燃机与配件》
2025年第6期83-85,共3页
Internal Combustion Engine & Parts
关键词
频谱分析
复杂信号
振动故障
特征信息
Spectrum analysis
Complex signals
Vibration malfunction
Feature information
