Aiming at the optimal dispatching problem of cascade hydropower stations in the lower reaches of the Jinshajiang River during the drawdown period, based on the analysis of the reservoir water balance, water level, discharge flow rate, and unit output, a daily-scale refined scheduling model was established by maximizing the total power generation of cascaded reservoirs and considering the actual scheduling requirements. The model was solved using the DPSA-POA algorithm. The effectiveness of the model and solution method was verified by the actual case. The results show that the proposed scheduling model can make full use of water resources and maximize the power generation in the drawdown period under the premise of ensuring the safe operation of each reservoir. Thus, it provides theoretical basis and technical support for the refined scheduling management of the cascaded reservoirs.

Addressing the issues of single model algorithm, low accuracy, and poor generalization in existing shield tunneling speed prediction methods, this study proposes a shield tunneling speed prediction approach to improve prediction accuracy based on Variational Mode Decomposition (VMD), Dung Beetle Optimizer (DBO), and Stacking ensemble learning. Firstly, to obtain more effective data, VMD is applied to decompose and reconstruct the original data to obtain denoised construction parameter data for subsequent model prediction. Secondly, based on the ensemble learning strategy, Support Vector Regression (SVR), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) models are selected as base learners, while Gaussian Process Regression (GPR) is chosen as the meta-learner to construct a Stacking ensemble learning prediction model with higher prediction accuracy and stronger generalization ability. Thirdly, to further enhance prediction accuracy, DBO is employed to optimize the hyperparameters of the ensemble learning model. Finally, this prediction method is applied to the shield tunneling construction of a water diversion tunnel project in Henan Province and compared with other prediction methods. Compared to other single models (SVR, RF, XGBoost), the results indicate that the proposed method achieves higher prediction accuracy, with average accuracy improvements of 7.76%, 6.70%, and 4.97%, respectively, providing a new approach for shield tunneling speed prediction.

To accurately calculate the effective range of vibration rods in reinforced concrete, this study proposes a new prediction model for the effective radius of vibration rods based on the rheological theory of fresh concrete. Under the assumptions that the vibrated concrete exhibits pseudoplastic fluid characteristics and the flow in the steel mesh is equivalent to seepage through porous media, an expression for the effective radius of the vibration rod in reinforced concrete is derived. The empirical parameters in the expression are then empirically fitted in conjunction with experimental results. By comparing the model calculations with experimental values, it shows that, except for a few cases with low vibration intensity of the rod, the overall prediction accuracy of the effective radius prediction model is higher. Considering that the vibration intensity of the rods used in concrete pouring at construction sites is generally high, the model can effectively guide on-site compaction operations and quantitatively evaluate the impact of steel bar layout schemes on the vibration range of the rod.

The long-term prediction of concrete dam deformation is an important requirement for maintaining its structural integrity during actual operation. To improve the accuracy of long-term deformation prediction of concrete, a long-term dam deformation prediction model based on multi-layer perceptron (MLP) and ecoder-decoder (Ecoder-Decoder) architecture, MLP-Ecoder-Decoder (MED), was constructed. This model captured the long-term dependence of dam deformation and environmental loads through a deep auto-correlation (Deep-Auto-Correlation) mechanism, and used time series decomposition and deep auto-correlation mechanism for multi-step deformation prediction. The model was used to predict the deformation of a 250 m height arch dam in Qinghai Province under complex environmental conditions. The results show that the MED model effectively improves the prediction accuracy and has a strong advantage in extracting long-term time features.

In order to further improve the control performance of water turbine, a control strategy based on the adaptive chaotic particle swarm optimization variable domain fuzzy PID (CAS-PSO-VFPID) was proposed for the turbine regulation system. Firstly, a model of nonlinear hydraulic turbine regulation system was established, and a variable domain fuzzy controller was constructed according to the system model. Then, the adaptive chaotic particle swarm optimization (CAS-PSO) was used to optimize and design the variable domain fuzzy controller, and the CAS-PSO-VFPID controller was obtained. Finally, the applicability of the nonlinear turbine regulation system model was verified by simulation. The multiple control strategies under different working conditions were compared and simulated with Whale Algorithm Optimization Variation Domain Fuzzy PID (WOA-VFPID), Standard Particle Swarm Optimization PID (PSO-PID), Standard Particle Swarm Optimization Optimization Variable Domain Fuzzy PID (PSO-VFPID). The simulation results show that the system convergence speed and optimization ability of the CAS-PSO-VFPID control strategy are fast, which can effectively improve the response speed and accuracy of the hydraulic turbine regulation system, and make the system have better dynamic stability.

According to the prototype characteristic curve of the pump set, there will be a deviation between the theoretical value and the actual value when guiding the actual scheduling of the pumping station, which will affect the accuracy of daily scheduling and energy consumption evaluation. Therefore, taking Liyuzhou pumping station as an example, this paper proposes a method combining the measured data of the pumping station and the law of similarity to correct the prototype characteristic curve of the pump set. The flow optimization distribution model is established in terms of the actual operating conditions, and the difference of optimal flow distribution scheme of pumping station between the characteristic curve of the pump set before and after the correction is analyzed. The results show that the deviation between the actual operating head and the theoretical head is mainly concentrated in 8%-11%, and the deviation between the actual operating efficiency and the theoretical efficiency is mainly concentrated in 6%-9% when the prototype characteristic curve of the pump set is used. When the flow optimization distribution scheme is selected through the corrected characteristic curve of the pump set, the results are more in line with the actual operation situation, which can effectively strengthen the guiding role of the flow optimization distribution scheme of the pumping station on the actual scheduling.

The tailwater level of hydropower station is a critical parameter for calculating the unit's output. When influenced by the downstream reservoir's backwater effect, discrepancies often arise between the designed tailwater curve and the actual observed values, leading to increased errors in the output-flow calculations. Utilizing the latest historical observation data, this study explores the relationship between the tailwater level of BHT Hydropower Station, its discharge, and the water level of the downstream XLD Reservoir. A Bayesian optimized long short-term memory (BO-LSTM) prediction model is developed based on multi-scenario analysis. The applied effect is analyzed under conditions of peak load and flood discharge. The results indicate that when the water level of XLD exceeds 585 meters, the tailwater level of BHT Hydropower Station is significantly influenced. Compared to the nonlinear curve fitting method, the BO-LSTM model based multi-scenario analysis demonstrates a substantial improvement in accuracy, with an average absolute error (M_MAE) reduced by 68.1%. The BO-LSTM model more accurately captures the fluctuations and changes in water levels under various operating conditions. The research results have important significant for refined operation of hydropower stations.

Aqueducts are common water conveyance structures in water diversion projects, and accurate prediction of aqueduct deformation is crucial for ensuring the stable operation of water conservancy projects. For this purpose, taking the Liaohe Aqueduct in the South-to-North Water Diversion Project as an example, five different linear additive models, namely elastic net regression, multiple linear regression, stepwise regression, ridge regression and LASSO regression, were established based on the long-term deformation monitoring data of the aqueduct. The prediction results of the aqueduct's deformation behavior by the five different linear additive models were compared. The results indicate that as the prediction time increases, the prediction accuracy of different linear additive models gradually decreases. The LASSO model selects the optimal regularization parameter through cross-validation, achieving variable selection simplification and minimizing model complexity. Additionally, it is verified that the training length affects the prediction performance of multiple linear regression and stepwise regression. The findings of this study provide valuable references for selecting prediction model of aqueduct deformation.

It is the key issues of reasonably and accurately predicting the thrust and torque of tunnel boring machines (TBM) to realize the intelligent control of TBMs. This paper proposes a two-stage prediction method of knowledge-data-driven spatio-temporal stacked convolutional network (KD-NTS-GAT). Firstly, based on expert knowledge and the NTS-NOTEARS method, a new information fusion technique is proposed. The discrete expert experience and the continuous NTS-NOTEARS indicators is mapped and smoothly fused through clustering. The causal relationships among the key operating parameters of the TBM is quantitatively extracted to improve the authenticity of the causal relationships significantly. Then, causality is further combined as a prior knowledge with stacked convolutional network deep learning model for predicting thrust and torque of TBM. Taking the bid Ⅳ of Xinjiang Water Conveyance Tunnel Project as an example, a comparative analysis of the KD-NTS-GAT method and the pure data-driven method shows that the KD-NTS-GAT has better prediction capability on thrust and torque. The conclusions can provide a reference for the intelligent control of TBM construction.

Aiming at the complex karst environment of the lower reservoir of a pumped storage power station, a three-dimensional finite element model for seepage analysis was established to simulate the main buildings and karst passageways in the reservoir area. "The depth of seepage control curtain at the base of the dam is 0.5 times of the pre-dam head, the depth of seepage control curtain on both sides of the dam is 3 m below the 3 Lu line, the length of seepage control curtain on the right side of the dam is 200 m, and a single-row curtain is set up" is used as the preliminary seepage control scheme. The seepage field, infiltration slope and infiltration volume of the reservoir area were calculated. The preliminary seepage control scheme met the specification requirements. But the seepage rate and infiltration slope were close to the critical value. The four indexes of curtain depth on both sides of the bank, curtain depth at the base of the dam, length of curtain on the right bank, and double-row curtain were changed to optimize seepage control scheme. The impact of the changes of the indexes on the infiltration volume of the reservoir area was investigated so that the optimization of the seepage control scheme was put forward. The analysis results show that on the basis of the preliminary scheme, the double-row curtain is set up at the dam base, the infiltration flow at the dam base is reduced by 267.3 m³/d, and the effect of seepage control in the reservoir area is remarkable.