Predicting the oscillation evolution of hydraulic turbines is crucial to ensuring their safe and stable operation. However, accurately and effectively predicting the oscillation evolution of hydraulic turbines can be challenging due to their non-stationary and nonlinear characteristics. Therefore, this article proposes an oscillation evolution prediction model for hydroelectric units based on Optimal variational modal decomposition-Bilateral gated recurrent neural networks (OVMD-BIGRU). Firstly, the decomposition parameters of Optimal variational modal decomposition (OVMD) are determined using the central frequency observation method and the residual index minimization criterion, avoiding subjective factors. Then perform Variational modal decomposition (VMD), normalize each modal component, and establish a BIGRU model for prediction. Finally, reverse normalize and stack the modal results to obtain the final predicted vibration trend of the unit. This article designs a comparative experiment based on the unit data of a domestic hydraulic power station, and the findings indicate that the proposed model has good performance and high prediction accuracy.
Hydro-Turbine Governing System (HTGS) is a key component in hydropower station, which converse hydropower into electricity, the stable operation of HTGS is of vital importance to a hydropower station. This paper aims to study the fault diagnosis method under noisy environment and propose an Improved Generalized Predictive Control (IGPC) strategy under fault operation conditions. Firstly, the Extended Kalman Filter (EKF) is adopted to characterize the faults of HTGS based on the nonlinear mathematical model of HTGS. Each EKF corresponds to a certain fault mode, which composes a fault archive. Meanwhile, the EKFs are able to effectively filter the external noise. The detection and recognition of faults are accomplished by the generalized residual model, the probability of a certain fault is quantified by the deviation between the observation of EKFs and the state variables of real HTGS. Then, the probability of each fault is injected into IGPC to reconstruct the CARIMA of HTGS, hence the CARIMA always follow the real state of HTGS even under fault operation conditions. Based on the online reconstruction of CARIMA, the robustness of GPC is further strengthened. Simulation result indicates that the proposed diagnosis model is effective for the detection and recognition of known faults under noise contaminated environment. In addition, the proposed IGPC is able to effectively stabilize the HTGS under fault operation conditions.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.