基于EEMD-SD振动信号分析的水电机组故障预警方法

Early Warning of Hydropower Unit Faults based on Vibration Signal Analysis with Ensemble Empirical Mode Decomposition and Standard Deviation

为确保水电机组安全运行及电网稳定,探讨利用水电机组振动信号实现机组故障早期预警,提出了基于集合经验模态分解和标准差水电机组振动信号特征量提取方法,并引入K均值和欧氏距离构建水电机组振动信号异常状态预警指标,量化水电机组振动信号状态异常变化程度,融合多测点状态信息形成水电机组振动故障综合预警指标,实现了水电机组振动故障预警,证明所提特征量和预警方法有效。

To ensure the safe operation of hydropower units and the stability of the power grid, early warning of the unit faults based on vibration signal analysis is explored. An extraction method of the vibration signal characteristics of the units based on ensemble empirical mode decomposition(EEMD) and standard deviation(SD) is proposed. The K-mean and the Euclidean distance are adopted to construct the abnormal state early warning indicator for the vibration signal. It can quantify the degree of the abnormal change of the vibration state of the units, and combine the state information from multiple measuring points to form a comprehensive early warning indicator. Therefore, the early warning of hydropower units based on vibration signal analysis is realized. The effectiveness of the proposed method is verified with practical measurement data.