Flow condition was simulated in a shaft tubular pump by using the Shear‐Stress Transport (SST) k‐ω turbulence model with high quality structured grids in design condition. Corresponding structural vibration characteristics were then analyzed based on two‐way coupled Fluid‐Structure Interaction (FSI) method. Fluid results showed that flow in the outlet flow passage was a combination of the axial flow and circumferential rotation motion. Time and frequency domain analysis of pressure pulsation of typical measure points indicated that larger pulsation amplitudes appeared in the tip of the blades and the main vibration source was the pressure pulsation induced by rotation of the blades. The fluid pulsation amplitudes decreased gradually along the flow direction, which can be ascribed to the function of fixed guide vane. Structural analysis of the blades in both pressure and suction side indicated that significant stress concentration was formed at the blade and hub connection near the leading edge. Maximum effective stress of the blades varied periodically, so prevention measures of the fatigue of blades should be taken. This research can provide important reference for the design of the tubular pump.

It is of significant value to understand the unsteady hydraulic features and pressure pulsation transmission path in the flow channel through a turbine for providing technical support for turbine design and optimization, as well as laying a foundation for analysis of the stability and the coupled vibration of the hydropower house. In this paper, a three-dimensional mechanics–hydraulics–concrete structure coupled numerical model was established to accurately simulate Francis hydraulic machinery, including the high-rotating turbine runner and fixed guide vane, the unsteady flowing water, the structure of the entire flow channel, as well as the dynamic interaction between them. Turbulent hydraulic features of flow condition and pressure pulsation in design operation were explored using the detached eddy simulation (DES) turbulence model. Then, a novel method was proposed to identify the fluid pressure pulsation transmission path based on the time-delayed transfer entropy method and wavelet theory. On basis of time and frequency analysis of pressure calculation results, investigation into identification of pressure pulsation transmission path was performed using the method of traditional transfer entropy and the method adopted in this paper. The pressure pulsation transmission features in the entire flow channel were revealed during operation of the large-scale Francis turbine. The research method and results could not only lay a basis for exploring the structural vibration regularity of the hydropower house but also provide a scientific reference for vibration reduction design of the hydropower house.

Research on the vibration response prediction and safety early warning is of great significance to the operation and management of pumping station engineering. In the current research, a hybrid prediction method was proposed to predict vibration responses of the pumping station based on a single model of autoregressive integrated moving average (ARIMA), a combined model of the adaptive network-based fuzzy inference system (ANFIS) and whale optimization algorithm (WOA). The performance of the developed models was studied based on the effective stress vibration data of the blades in a shaft tubular pumping station. Then, the D-S evidence theory was adopted to perform safety early warning of the operation state by integrating the displacement, velocity, acceleration and stress indicators of the vibration responses of the pumping station. The research results show that the proposed prediction model ARIMA–ANFIS–WOA exhibited better accuracy in obtaining both linear and nonlinear characteristics of vibration data than the single prediction model and hybrid model with different optimization algorithms. The D-S evidence fusion results quantitatively demonstrate the safe operation state of the pumping station. This research could provide a scientific basis for the real-time analysis and processing of data in pumping station operation and maintenance systems.

The working principle of a large hydropower station is to guide the high‐pressure water flow to impact the turbine to rotate and generate electricity. The high‐pressure water flow impacts the turbine blades, which forms complex high‐speed eddy currents in the spiral case and the draft tube and causes complicated vortex‐induced vibration problems. Traditionally used harmonic response methods and dynamic time‐history analysis methods are difficult to reflect this complex fluid‐solid dynamic coupling problem. In this paper, the bidirectional fluid‐structure interaction (FSI) simulation analysis theory for a large hydropower house is studied, and the analysis methods of geometric simulation, mechanical simulation, and vibration energy transmission path simulation are presented. A large‐scale 3D fluid‐hydraulic machinery‐concrete structure coupled model of a hydropower house is established to study the vortex‐induced vibration mechanism and coupled vibration law during transient unit operation. A comparison of the fluid results against the in‐site data shows good agreement. Structural responses of vibration displacement, velocity, and acceleration reveal coupled regularity of hydraulic machinery‐concrete structure‐fluid during blades rotating periods, and it comes to the conclusion that the turbine blade rotation is the main vibration source of the hydropower house. The research results can provide a scientific basis for the design and safe operation of the hydropower house.

In response to the problem of insufficient model training and low prediction accuracy caused by insufficient load data samples in the target domain, this paper proposes a method for predicting the vibration response of hydropower houses by integrating the CNN-BiLSTM deep learning model. The CNN is utilized to extract the spatiotemporal features of high-frequency transient vibrations, which are then fed as input to the BiLSTM structure for further learning. Subsequently, the proposed model is employed to predict the vibration acceleration of the dam foundation and corridor in the hydropower station. The research findings demonstrate that the proposed CNN-BiLSTM model significantly enhances the accuracy of vibration prediction when compared to LSTM, BiLSTM, and CNN-LSTM data prediction algorithms.

Accurate prediction of vibration responses in hydraulic structures during flood discharge is essential for ensuring safe and stable operation. This study develops a hybrid deep learning model that combines Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (BiLSTM), and a Channel Attention (CA) mechanism, optimized through Bayesian Optimization (BO), to predict dam gantry crane beam displacements. Time-lagged Pearson correlation and Maximum Information Coefficient (MIC) are applied to select the informative input features. The CNN-BiLSTM-CA model captures both spatial patterns and temporal dependencies in vibration signals. BO tunes model hyperparameters, while Partial Dependence (PD) analysis provides insight into how these parameters affect prediction accuracy. The model is validated using vibration data from an arch dam in Southwest China during flood discharge. Results show that CNN parameters have a greater impact on prediction accuracy than BiLSTM parameters, underscoring the importance of spatial feature extraction. Ablation studies confirm each component’s contribution. Compared with existing methods, the proposed model achieves superior accuracy with a Root Mean Square Error (RMSE) of 5.49, Mean Absolute Error (MAE) of 4.34, and correlation coefficient (R) of 99.42%. This framework provides a reliable and interpretable tool for predicting structural vibrations in hydraulic engineering under complex discharge conditions.