Water turbine rotating shaft state monitoring method and system based on a neural network

• Main Author: QIAO LIZHONG
• Format: Patent
• Language: Chinese; English
• Published: 10.05.2019
• Subjects: BLASTING; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HEATING; LIGHTING; MACHINES OR ENGINES FOR LIQUIDS; MECHANICAL ENGINEERING; OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR; PHYSICS; PRODUCING MECHANICAL POWER; WEAPONS; WIND, SPRING WEIGHT AND MISCELLANEOUS MOTORS

The invention relates to a water turbine rotating shaft state monitoring method and system based on a neural network. The method comprises the following steps: 1) establishing and training a neural network-based rotating shaft temperature prediction model according to historical data of the rotating shaft of the water turbine; 2) determining a fault threshold of the rotating shaft of the water turbine through a nuclear density estimation method; and 3) predicting the temperature residual distribution characteristic of the rotating shaft through the trained rotating shaft temperature predictionmodel by using the real-time data of the rotating shaft of the water turbine, and comparing the rotating shaft temperature residual distribution characteristic with the fault threshold to judge whether a fault occurs or not. According to the method, a correlation coefficient method is used for selecting parameters having certain correlation with the rotating shaft temperature for modeling, invalid values in data are remo