Study on fault prediction method of quantum SBO-Kalman-LSTM hydrogenerator

In order to meet the new requirements of Turbine Unit Failure Prediction (TUFP) in the era of big data, a Kalman filter is introduced for filtering processing, and a TUFP-Kalman-LSTM deep learning prediction network is proposed by integrating Long Short Term Memory (LSTM). In order to obtain the best combination of hyper-parameters of deep learning networks, the Quantum Satin Bowerbird Optimizer (QSBO) algorithm is introduced, and a TUFP-Kalman-LSTM network hyperparameter optimization method based on QSBO algorithm is established. A new method for fault prediction of TUFP-Kalman-LSTM-QSBO turbine units is constructed, and applied to the online monitoring data of unit #2 of a hydropower station in Hubei Province in July. The actual operation results show that the method can effectively predict the vibration trend of the hydro generator set and realize the intelligent prediction of the hydro generator.