Steam turbine generator unit health degree evaluation and prediction method based on multi-parameter fusion

• Main Authors: PU CHENJIE, YOO KYUNG-MIN, FU DONGLIANG, SHI YIFAN
• Format: Patent
• Language: Chinese; English
• Published: 24.10.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention relates to a steam turbine generator unit health degree evaluation and prediction method based on multi-parameter fusion, and the method comprises the steps: firstly, designing an evaluation index system of a unit according to the common monitoring parameters of a steam turbine generator unit, employing an entropy weight method to carry out the weighting of each characteristic parameter based on a certain amount of monitoring historical data, and obtaining the health degree of the steam turbine generator unit; the method comprises the steps that firstly, feature parameters are obtained, weights of the feature parameters and weights of subsystems are obtained through calculation, then the health degree of each subsystem and the overall unit is evaluated based on real-time data, a historical trend sequence of the health degree is formed, then the health degree sequence is used for training of an SVR regression model, a health degree prediction model is obtained, and the health degree at the next m