基于PCA-KNN的给水泵振动预警方法研究

Research on Early Warning Method of Feed Pump Vibration Based on PCA-KNN

基于某燃气-蒸汽联合循环机组高压给水泵,研究了基于PCA-KNN(主成分分析-K近邻算法)的给水泵振动预警方法。在实际工程仿真中,通过对比直接采用KNN算法进行给水泵振动预测的结果,采用PCA-KNN算法,不仅预测精度相近,而且计算速度得到了大幅提升。此外,还基于PCA-KNN算法,设计了一种将振动参数预警与振动劣化预警相结合的振动异常预警模式。

Based on high-pressure feed pump of a gas-steam combined cycle unit,the PCA-KNN-based early warning method for feed pump vibration was discussed.In simulating practical engineering,comparing feed pump’s KNN algorithm-based vibration predictions show that,PCA-KNN has similar prediction accuracy and the calculation speed cab be greatly improved.In addition,based on the PCA-KNN algorithm,the abnormal vibration’s warning mode which having the vibration parameter warning combined with the vibration degradation warning was designed.