Abstract: The present disclosure discloses a method for capturing a trip sign of a turbine due to a high bearing temperature based on correlation and a device therefor. By combining a temperature of a target bearing and related operating parameters thereof, this method can capture possible abnormal trip online. According to the present disclosure, it is not necessary to add additional detection equipment, and it does not need to establish a complex physical model for turbine bearings, and only the historical data of the operating parameters of the temperature of the target bearing and generator set operating parameters related to the temperature of the target bearing are required to complete the establishment of the model for capturing abnormal sign before the trip, which is convenient for popularization and application.

Abstract: The present disclosure discloses a method for capturing a trip sign of a turbine due to a high bearing temperature based on correlation and a device therefor. By combining a temperature of a target bearing and related operating parameters thereof, this method can capture possible abnormal trip online. According to the present disclosure, it is not necessary to add additional detection equipment, and it does not need to establish a complex physical model for turbine bearings, and only the historical data of the operating parameters of the temperature of the target bearing and generator set operating parameters related to the temperature of the target bearing are required to complete the establishment of the model for capturing abnormal sign before the trip, which is convenient for popularization and application.

Abstract: The present invention provides a method for predicting remaining useful life of bearings based on a gated recurrent neural network, comprising the following steps: S1. obtaining full life cycle vibration signals of bearings, extracting the vibration distribution features and creating a training set of gated recurrent neural network; S2. constructing a gated recurrent neural network model, and introducing an attention mechanism that directly calculates weights to improve the integrity of extracting temporal information; S3. adding a Bayesian layer to construct a nonlinear mapping relationship between temporal information and remaining useful life; S4. taking vibration signals of a test bearing as input, the output result of the gated recurrent neural network model is the remaining useful life of the test bearing at the current time. The present invention does not need to add additional neural network layers, which avoids the problem of increasing the complexity of the model.