Abnormality warning method, device and equipment for water-turbine generator set and medium

• Main Authors: PEI JUN, XU KAIWEI, LIANG YEQUAN, LIU HUAN, WANG SIJIE, CHEN YUNYUN, HUANG XIANGWEI, WAN BO, HU WENXING, GONG YU, YANG MINGXUAN, YU YAXIONG, LUO SHUSHENG, CHEN XUTENG, ZHANG BONING
• Format: Patent
• Language: Chinese; English
• Published: 03.10.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC ORINFRASONIC WAVES; MEASURING; PHYSICS; TESTING

The invention discloses an abnormity warning method, device and equipment for a water-turbine generator set and a medium. The method comprises the following steps: acquiring current vibration data of a target water-turbine generator set in real time, and performing data preprocessing on the current vibration data to obtain standard current vibration data; performing feature extraction on the standard current vibration data through a preset vibration data transformation method to obtain a current spectral kurtosis feature; inputting the current spectral kurtosis feature into a pre-trained abnormal alarm cluster confirmation model, and determining a clustering cluster type corresponding to the current spectral kurtosis feature; and if the cluster type is an abnormal alarm cluster, carrying out abnormal alarm operation on the target water-turbine generator set. The problem that it is difficult to accurately give an alarm to the water-turbine generator set according to the collected vibration data is solved, the