Fault diagnosis method and system for blower of thermal power plant and electronic equipment

• Main Authors: LI JUN, SUN ZIWEI, DONG WEIBO, ZANG DIANZONG, LIU QINGSONG, ZHAO DONGHUI, HAN JIANHENG, LIU LIN, JIA ZHIJUN, WU LIXING, HE LIRONG, PEI HAORAN, WANG RONG, LIANG MANCANG, LIU SHUAI, MOU ZHIQIANG, XIAO LIANG, CHEN GANG, PENG DAOGANG, FAN GUOQIANG, QIU XIAOZHI, ZHAO ZHIHONG, ZHU QICHUN, SHI YANQIANG, LIU JI, CHANG YI
• Format: Patent
• Language: Chinese; English
• Published: 03.02.2023
• Subjects: BLASTING; CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; HEATING; LIGHTING; MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC ORINFRASONIC WAVES; MEASURING; MECHANICAL ENGINEERING; NON-POSITIVE DISPLACEMENT PUMPS; PHYSICS; POSITIVE DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS; TESTING; WEAPONS

The invention relates to a thermal power plant blower fault diagnosis method and system and electronic equipment. The method comprises the following steps: firstly, carrying out initial noise reduction on a non-linear and non-stationary original vibration signal, and removing part of random particle noise; secondly, through a latent factor model (LFM), modeling is carried out on the signals to obtain implicit features, and signal decomposition is carried out on the noise reduction signals to realize secondary noise reduction; and finally, based on the extracted Product Function (PF) component (i.e., the second PF component), a diagnosis result is obtained by using an efficient and extensible multi-fault classifier, so that the fault diagnosis of the blower of the thermal power plant is efficiently and accurately realized. 本发明涉及一种火电厂送风机故障诊断方法、系统及电子设备。本发明首先对非线性非平稳的原始振动信号进行初始降噪，去除部分随机颗粒噪声；其次，通过潜在因子模型(LatentFactorModel，LFM)，对信号进行建模得到隐含特征，并对降噪信号进行信号分解，实现二次降噪；最后，基于提取的乘积函数(ProductFunction，PF)分量(即第二PF分量)，利用高效可扩展的多故障分