基于随机共振和经验模态分解的水力发电机组振动故障诊断

发布时间：2018-01-24 19:20

本文关键词： 随机共振 EMD 支持向量机故障诊断水力发电机组　出处：《水利学报》2017年03期 　论文类型：期刊论文

【摘要】：针对实际水力发电机组故障诊断中微弱信号难以检测引起故障诊断准确率低的难题,提出了一种基于随机共振(stochastic resonance,SR)和经验模态分解(Empirical Mode Decomposition,EMD)的微弱信号检测方法。首先,采用随机共振对振动信号进行降噪处理,提高信号的信噪比;继而对随机共振的双稳输出信号进行EMD分解,并采用能量法进行故障特征向量的提取,最后将其作为基于遗传算法优化支持向量机(GA-SVM)故障诊断模型的输入,实现故障模式的识别与诊断。仿真结果表明,该方法能够准确识别机组的异常情况,具有较高的故障诊断精度。
[Abstract]:In order to solve the problem of low accuracy caused by the difficulty of weak signal detection in fault diagnosis of practical hydroelectric generating units, a new fault diagnosis method based on stochastic resonance (SRM) stochastic resonance is proposed. SR) and empirical Mode decomposition (EMD) for weak signal detection. The signal to noise ratio of the vibration signal is improved by using the stochastic resonance to reduce the noise. Then the bistable output signal of stochastic resonance is decomposed by EMD and the fault eigenvector is extracted by energy method. Finally, it is used as the input of the GA-SVM fault diagnosis model based on genetic algorithm to realize the fault pattern recognition and diagnosis. The simulation results show that. This method can accurately identify the abnormal condition of the unit and has high fault diagnosis accuracy.
【作者单位】：西安理工大学;国网电力科学研究院;甘肃省电力科学研究院;
【基金】：国家自然科学基金项目(51279161) 陕西水利科技计划项目(2015slkj-04)
【分类号】：TV738
【正文快照】： 1研究背景水力发电机组一般工作在高温、高压和高转速等比较恶劣的环境条件下,较容易发生各种故障,严重影响了水电站的安全稳定运行[1-2]。据统计,水力发电机组80%以上的故障和事故与振动紧密相关,然而水力发电机组发生故障时引起的振动故障原因与征兆之间存在不确定性、非线

【相似文献】