Abstract:To investigate the statistical characteristics of wind and waves during typhoon processes, based on 40-year (1979-2018) hindcast wave data and reanalysis wind speed data, the wind speed and significant wave height data during periods of significant typhoon influence at three study sites in the northwestern South China Sea were extracted using trajectory tracking and time matching methods. The joint distribution characteristics of wind speed and wind direction, as well as significant wave height and wave direction, were analyzed, and the delay characteristics of wind and waves during typhoon processes were examined. Furthermore, based on the extreme value theory, the maximum significant wave height and maximum wind speed samples during each typhoon process were extracted. Based on the three-parameter Weibull distribution model and Copula model, the extrapolated values for 50-year, 100-year, 150-year, and 200-year return periods under univariate and joint return periods were calculated. The results indicate that in the northern part of the study area, strong waves exceeding 6.0 m driven by typhoons are mainly distributed in the 33.75°-123.75° sector, and the time difference between the peak wind speed and peak significant wave height in the study area is mainly concentrated in 0-20 h. Because a broader risk range is considered, the extrapolated values of the joint return periods are greater than those of the univariate return periods.

Abstract:To explore the characteristics of wind field models, by taking Super Typhoon “Yagi” in 2024 as an example, a parameter sensitivity analysis of the Yan Meng wind field model was conducted via numerical simulation. The results show that the surface wind speed calculated by the Yan Meng wind field model is positively correlated with the pressure profile constant and the maximum wind speed radius, and the pressure profile constant is positively correlated with the maximum wind speed radius. A smaller equivalent roughness height leads to a higher surface wind speed. The relative error of the surface wind speed calculation results of the Yan Meng wind field model can be stabilized within 20%, and the average value of the mean absolute error is within 2 m/s. The Yan Meng wind field model exhibits good adaptability in the study of typhoon wind fields along the coast of China, and it can be applied to the evaluation of typhoon disasters along the coast of China.

Abstract:Based on ERA5 reanalysis data, the spatial distribution characteristics of Stokes drift velocity in four seasons in adjacent seas of China from 1980 to 2023 were calculated and analyzed, and the variation characteristics of Stokes drift velocity under typical typhoon conditions were analyzed. The results show that the Stokes drift velocity in adjacent seas of China exhibits significant latitudinal differentiation and seasonal variation characteristics. Except in summer, the high-value areas of Stokes drift velocity are all located in the adjacent seas of the Taiwan Strait, China, where the average velocity can reach 0.1 m/s, and the direction is generally southward and southwestward. In summer, the high-value areas of Stokes drift velocity shift to the southwest, located in the southeast of the Indo-China Peninsula, with velocities ranging from 0.08 m/s to 0.09 m/s, and the direction is generally northward and northeastward. Affected by topography and monsoons, the Stokes drift in the Yellow Sea and Bohai Sea regions presents the characteristic of northward propagation along the coastline, which is closely related to the northward transport of floating objects on the sea surface. Under the influence of extreme weather such as typhoons, especially in the seas near the typhoon generation area, the Stokes drift velocity along the typhoon path increases by 1.5-2.3 times.

Abstract:Aiming at the strong random volatility of ocean energy generation, power characteristic indexes, including amplitude stability, duration stability, and the coefficient of variation, were calculated for wave and tidal current power generation. A feature vector constructed from these power characteristic indexes was then used to perform dimensionality reduction and clustering of the time-series data of ocean energy generation power, thereby obtaining typical output scenarios of ocean energy generation.With the goal of optimizing the consumption rate, a model for optimizing the allocation of ocean energy generation capacity is constructed and transformed into a linear programming problem for solution.The calculation results based on actual data show that the optimized configuration results after clustering of characteristic indicators are basically consistent with the measured output data scenario, which verifies the effectiveness of the method.

Abstract:In view of the current lack of systematic analysis and quantitative research on the load mechanism and pipeline stress evolution during the trenching and lowering of in-service subsea pipelines, OrcaFlex software was used to numerically simulate the secondary trenching and lowering of in-service subsea pipelines, and the influence of key load variations on pipeline stress under different trenching depths was quantitatively analyzed. The results indicate that temperature difference, pipeline pressure difference, and wave-current load show a linear or nonlinear positive correlation with pipeline stress; the pipeline residual tension has little effect on the pipeline stress; the existence of soil resistance can effectively reduce the stress effect of wave-current load. Under different trenching depths, the maximum variation rate of pipeline equivalent stress under the same temperature difference can reach 130 MPa/m, and the maximum stress variation rate under the same pressure difference approaches 175 MPa/m.

Abstract:In view of the local scour problem around pile groups in cohesive sediment beds, the local scour characteristics of a single pile and 5×5 pile groups under various pile spacing conditions were investigated through laboratory flume experiments. In the experiments, 0.038 mm quartz powder was used to simulate silty cohesive sediment. Combined with geotechnical parameter measurements and three-dimensional topographic scanning technology, the variation law of scour depth over time, the integral characteristics of pile group scour, and the influence of pile spacing on scour efficiency were analyzed. Moreover, the differences in scour mechanisms between cohesive and non-cohesive sediments and their physical driving factors were discussed. The results show that the temporal variation of local scour depth in cohesive sediment can be divided into an initial scour stage, a scour development stage, and a scour equilibrium stage. The scour depth behind the pile is generally greater than or equal to that in front of the pile, which is significantly different from the local scour characteristics around piles in non-cohesive sediment. Furthermore, the local scour efficiency of pile groups decreases with the increase in pile spacing; the maximum scour depth reaches nearly six times the pile diameter at three times the pile spacing, which is 2.4 times the local scour depth of a single pile. The local scour depth of the piles in the first three rows is generally smaller than that of the downstream piles, and the scour depth tends to be stable after the third row.

Abstract:To address the problem that non-vertical installation of suction pile can easily lead to shallow soil instability and wellhead overturning, a study on verticality control of suction piles with different diameters and installation depths was conducted through a combination of theoretical analysis and simulation tests. A calculation method for verticality control of suction piles in deepwater oil and gas well construction was proposed based on the structural dimensions of the suction pile and the properties of the formation soil and was validated through simulation tests. The results show that the diameter of the suction pile, the penetration depth of the suction pile, and the properties of the seabed soil are the main factors affecting the verticality control of the suction pile. Under the same other conditions, a smaller diameter of the suction pile and a shallower penetration depth pose higher verticality control requirements on the wellhead. To ensure the stability of the deepwater seabed wellhead, the inclination angle of the suction pile wellhead shall not exceed 6°.

Abstract:Based on the HSS constitutive model, a three-dimensional finite element model of a tripod bucket foundation in a soft clay seabed was established; the bearing characteristics of the tripod bucket foundation under horizontal cyclic loading were analyzed, and a multi-parameter sensitivity analysis on the accumulated deformation of the foundation was conducted using the range analysis method. The results indicate that during cyclic loading, the axial force transmitted to the top of the buckets via the tripod frame contributes the most to resisting the external total overturning moment. Under the serviceability limit state, the foundation stiffness degrades to 96.3% of its initial value, and the vertical displacement difference between the front and rear buckets is relatively small, resulting in a slow growth rate of the accumulated rotation angle. Under the fatigue limit state, the foundation stiffness degrades to 86.5% of its initial value; the vertical displacement difference between the front and rear buckets increases correspondingly, and the accumulated rotation angle continuously increases with the number of cycles. As the spacing between the foundation buckets increases, the rotation center becomes deeper, and the foundation tends to exhibit a translational movement. The accumulated deformation of the foundation is highly sensitive to the loading frequency and bucket spacing, followed by the initial shear modulus of the soil and the aspect ratio of the buckets, while it is least sensitive to the threshold shear strain of the soil.

Abstract:In response to the “23·7” extreme storm-flood event in the Haihe River basin, the Beixindian watershed of the Qingshui River was selected as the study area. Based on the grid Xin’anjiang model, a dual retention grid model was developed by coupling surface and subsurface retention modules. The “23·7” flood process was retrospectively simulated, and the rainfall-soil water-groundwater response relationship and the regulation and storage effects of the surface and subsurface retention reservoirs were analyzed. The results show that for 11 historical flood events from 1980 to 2000, the dual retention grid model achieves a 100% qualification rate for runoff depth and a 64% qualification rate for flood peaks, and the proportion of events with a deterministic coefficient greater than 0.5 is 64%, which are superior to the grid Xin’anjiang model and the existing forecasting schemes in the study basin. For the “23·7” flood process, the simulated errors of flood peak and runoff depth by the dual retention grid model are constrained within 4%, with a deterministic coefficient of 0.91. The spatiotemporal variation of soil moisture has a regulating effect on the formation and recession processes of the “23·7” flood. The surface retention module of the dual retention grid model has a flood-retention and peak-reduction effect, and the subsurface retention module improves the agreement of the recession curve.

Abstract:To address the problems that the water level of tidal river networks exhibits nonlinear fluctuations under the coupling effects of tides, runoff, and sluice-pump regulation, and that topographic differences of water systems result in low accuracy of holistic unified simulation and high cost and complexity of individual station modeling, an intelligent water level prediction method for tidal river networks was proposed based on a multi-scale fusion spatiotemporal attention network (MSTANet) with spatiotemporal clustering. By integrating multi-scale temporal features and spatial dependencies, this method enhanced the simulation capability for water level evolution processes under multiple disturbances, such as tides and sluice-pump regulation; meanwhile, spatiotemporal clustering was introduced to achieve watershed subdivision modeling, effectively mitigating the performance degradation caused by spatial heterogeneity. The validation in Wennan Water Control Area of Shanghai shows that this method has good simulation accuracy and computational efficiency and possesses engineering practicability and application prospects.

Abstract:In response to the issues of single simulation results and lack of interpretability when using data-driven models to replace traditional hydrodynamic models for rapid forecasting, an LSTM model for joint simulation of multiple factors was constructed based on a multi-task learning framework, and the fitted physical relationships among factors were incorporated into the loss function as constraint terms. Furthermore, the interpretability of the model was explored by combining the comprehensive feature evaluation results of random forest, Spearman rank correlation coefficient method, and principal component analysis (PCA). Based on a database of simulation results of 17 flood events in Tunxi District by a two-dimensional hydrodynamic numerical model, the LSTM models with and without physical constraints were trained, respectively. The results indicate that the computational speed of the LSTM model with physical constraints does not decrease significantly, which can realize multi-factor simulation within 1.s. For the overall simulation of multiple flood events, the LSTM model with physical constraints performs better, and the simulation accuracies of water level, discharge, flow velocity, and inundation area are improved by 20.38%, 17.03%, 4.02%, and 14.07%, respectively. The LSTM model with physical constraints demonstrates a stronger capability in capturing small floods and higher accuracy in simulating regular floods, but its performance still needs to be improved under extremely complex conditions.

Abstract:By considering the superposition (reduction) effect of disaster losses, an adaptive index formula for multivariate compound disaster systems was derived, and an adaptive evaluation index system for multivariate compound disaster systems was constructed from the two major dimensions of disaster prevention capacity and disaster response capacity. The entropy weight-TOPSIS method is used to evaluate adaptability of each region, and ArcGIS software is utilized to map the adaptability zoning. The application results of the Guangdong-Hong Kong-Macao Greater Bay Area case show that for the “rainstorm-landslide” and “rainstorm-collapse” multivariate compound disaster systems in the Guangdong-Hong Kong-Macao Greater Bay Area, the regions with high adaptability are mainly concentrated in the southern area; the regions with relatively high adaptability are mainly a few areas in the central part; the regions with medium and relatively low adaptability are mainly distributed in most areas of the central part and the southwestern and eastern areas; the regions with low adaptability are mainly located in the northwestern and northeastern areas, respectively. When facing multivariate compound disaster systems, the regional adaptive index is significantly lower than that for single disasters.

Abstract:By considering the seasonal regulation characteristics of hybrid pumped storage power stations, a long-and short-term nested scheduling model for hydro-wind-photovoltaic-storage complementary systems was established with the objective of maximizing power generation benefits. In view of the high-dimensional features in solving the model, the dynamic programming successive approximation algorithm was improved by introducing the pumping-generation conversion model of hybrid pumped storage power stations, and the impact of hybrid pumped storage power stations on the operational benefits of the hydro-wind-photovoltaic-storage complementary system was quantitatively analyzed. Case study results indicate that for hybrid pumped storage power stations built relying on hydropower stations with seasonal or higher-level regulation capabilities, a seasonal regulation scheduling scheme of hybrid pumped storage is recommended. Compared with the scheme without hybrid pumped storage, this scheme can effectively reduce the curtailment of renewable energy (99.37%) and increase the power generation benefits of the system (4.34%). By pumping water from the downstream reservoir to the upstream reservoir in advance, additional flood control storage capacity can be provided for the downstream reservoir, reducing the water curtailment of cascade reservoirs (93.35%). Compared with the daily regulation scheme of hybrid pumped storage, this scheme can improve the power generation in dry periods and increase the comprehensive energy conversion efficiency of the system (6.88%).

Abstract:To address the multi-attribute, multi-level, and uncertain evaluation problem of capacity allocation schemes for hydro-photovoltaic complementary systems, a three-level evaluation index system covering five factor layers of security, environmental protection, economy, technicality, and sociality was constructed, and a comprehensive evaluation model based on combined weighting and cloud matter-element analysis method was proposed. In this model, subjective weights were determined through correction by the AHP-cloud model; objective weights were obtained by introducing the improved entropy weight method; weight fusion was realized based on Euclidean distance. Moreover, the cloud matter-element theory was applied to calculate the membership degrees of each scheme to evaluation grades and the comprehensive scores, expanding the grade boundaries from rigid thresholds to flexible distributions, thereby more reasonably depicting the random fluctuations and fuzzy transitions in the evaluation process. The results of the case verification for the clean energy base in the Xizang section of the Lancang River indicate that the model exhibits a good differentiation effect in the comparison and selection of multiple capacity allocation schemes. The evaluation results are consistent with the engineering operation characteristics, and the conflicts among multi-dimensional indicators can be effectively coordinated. The comprehensive evaluation model considers both the comprehensiveness of weight determination and the robustness of grade judgment, and it can provide scientific and reliable decision support for the capacity allocation of clean energy bases.

Abstract:To address the problem that the reactive power compensation configuration of new energy power systems is mostly based on a single operating scenario and is often solved by single-objective optimization algorithms, a reactive power compensation configuration scheme considering multiple operating scenarios and multiple optimization objectives simultaneously was proposed. Considering the randomness of wind power output, typical wind power output scenarios were generated by the Monte Carlo sampling method and K -means clustering method. An optimal configuration model of reactive power compensation devices under multiple typical wind power output scenarios was established by taking system voltage deviation, active power loss, and configuration cost of reactive power compensation devices as optimization objectives, taking installation locations and compensation capacities of reactive power compensation devices as optimization variables, and setting power flow, node voltages, and power source outputs as constraints. At the same time, a multi-objective particle swarm optimization algorithm was used to solve the optimal configuration model to obtain the optimal installation locations and compensation capacities of reactive power compensation devices. The simulation analysis on the IEEE 39-node system with wind power shows that under multiple operating scenarios of the system, the proposed reactive power compensation configuration method can effectively reduce system voltage deviation and active power loss, and the configuration schemes are more diverse.

Abstract:To resolve the contradiction between massive monitoring data of twin models and high-fidelity rapid response, a rapid calculation method for dam deformation based on a POD-RBFNN reduced-order model was proposed taking an arch dam as an example. Based on static response data under different water level conditions obtained through finite element simulations, dominant modes and modal coefficients were extracted using the proper orthogonal decomposition (POD) method. Combined with a radial basis function neural network (RBFNN), a high-accuracy surrogate mapping between water level parameters and modal coefficients was established, achieving rapid and accurate prediction of the static displacement field of the arch dam. Calculation results show that under the condition of considering only hydrostatic pressure and structural self-weight, the prediction results of the POD-RBFNN reduced-order model agree well with finite element (FEM) simulation results, with an average relative error of 0.35% and an average absolute error of only 0.003 2 mm. Meanwhile, the calculation efficiency is improved by 3 004 times compared with the finite element simulation.

Abstract:In view of the problems that existing research lacks an analysis on the optimal depth of cutoff walls in ultra-deep multilayer foundations containing an impermeable layer, and the influence laws of seepage under high external water pressure remain unclear, a generalized vertical anti-seepage model was established based on a hydropower project. By analyzing the seepage fields of the dam foundation under different cutoff wall depths, the influence laws of cutoff wall depth variations on the seepage field of the dam foundation in ultra-deep complex multilayer foundations were discussed, and on this basis, an optimal cutoff wall depth satisfying the standards of both soil seepage stability and seepage discharge control was proposed. The results indicate that the presence or absence of an impermeable layer in ultra-deep complex multilayer foundations significantly influences the evolution laws of the maximum hydraulic gradient at the cutoff wall tip. When the dam foundation does not contain an impermeable layer, the maximum hydraulic gradient at the cutoff wall tip shows a U-shaped trend with an increase in the penetration ratio and changes significantly when the penetration ratio is less than 0.2 or greater than 0.7. When the dam foundation contains an impermeable layer, above the impermeable layer, the maximum hydraulic gradient at the cutoff wall tip shows a U-shaped trend with an increase in the penetration ratio and changes significantly when the penetration ratio is less than 0.2 or greater than 0.7; within the impermeable layer, the maximum hydraulic gradient at the cutoff wall tip decreases rapidly with an increase in the penetration ratio; below the impermeable layer, the maximum hydraulic gradient at the cutoff wall tip increases slowly with an increase in the penetration ratio and increases rapidly when the penetration ratio is greater than 0.7. Considering the possible inhomogeneity and uncertainty in the deep parts of the ultra-deep overburden foundation, combined with existing engineering cases, a final cutoff wall depth of 200 m is recommended.

Abstract:By taking the Yellow River Delta as a prototype, an idealized model featuring typical river mouth bar topography was constructed using Delft3D. The model was used to simulate and analyze the influence of vegetation on the topographic evolution of the river mouth bar in the Yellow River Delta under water and sediment regulation. The results show that with vegetation, the sediment deposition volume of the river mouth bar after the water and sediment regulation is reduced by 63.7% compared to the scenario without vegetation, which contrasts with the general understanding that vegetation promotes sediment trapping and accumulation. The flow resistance effect of the vegetation on the river mouth bar alters the flow partitioning ratio between the bifurcated channels and the river mouth bar, reducing the flow discharge over the bar by 70%-85%. This redirects sediment-laden flows primarily through the river channel to the sea, leading to a decrease in local sediment supply and trapping on the river mouth bar. Furthermore, the geomorphic effects of vegetation differ significantly between the water-regulation period and the sediment-regulation period. In the absence of vegetation, vertical accretion and lateral expansion of the river mouth bar mainly occur during the sediment-regulation period, characterized by low discharge and high sediment concentration, with almost no deposition during the water-regulation period. In contrast, when vegetation is present, significant vertical accretion of the river mouth bar occurs during the water-regulation period, while its lateral expansion mainly takes place during the sediment-regulation period and is confined to the direction of sediment supply.

Abstract:To address the overly conservative predictions of empirical formulas and the lack of physical mechanism guidance in purely data-driven models for predicting the local scour depth around bridge piers, a multilayer perceptron (PSI-MLP) model based on coupled physics-informed and symbolic regression constraints was developed. In the PSI-MLP model, the HEC-18 equation and the Melville scour formula were introduced into the standard multilayer perceptron framework as physical constraints, explicit equations extracted from data by the symbolic regression algorithm were combined as additional constraints,and the loss function integrates data loss, physical loss, and symbolic regression. With extreme gradient boosting, support vector machine, and purely data-driven multilayer perceptron models as benchmark models, the performance differences between the PSI-MLP model and each benchmark model were comparatively analyzed from multiple dimensions of prediction accuracy, model interpretability, and physical consistency. Meanwhile, based on 56 groups of field-measured engineering data, the cross-scale extrapolation and generalization abilities of the PSI-MLP model in practical engineering scenarios were analyzed. The results show that the root mean square error, coefficient of determination, and Kling-Gupta efficiency coefficient of the PSI-MLP model on the test set are 0.234, 0.88, and 0.928, respectively. Compared with the benchmark models, the root mean square error decreases by up to 34.8%, and the coefficient of determination and Kling-Gupta efficiency coefficient increase by up to 16.4% and 25.9%, respectively. The decision logic of the PSI-MLP model is consistent with hydrodynamic theories. The relative error of the PSI-MLP model is less than 25% under 71.4% of the working conditions, which is superior to the traditional empirical formulas and benchmark models. The coupled constraint mechanism of physical information and symbolic regression can improve the adaptability and prediction stability of the model under complex engineering-scale extrapolation conditions.

Abstract:To improve the drainage performance of iron tailings dams, permeability tests were conducted on geotextile filters in iron tailings dams under different vacuum negative pressures. The particle migration laws within the filter and the variation laws of the pore structure of the geotextiles were analyzed, and the influence of vacuum negative pressure on the permeability characteristics of the iron tailings-geotextile filter was investigated. The results indicate that vacuum negative pressure can significantly improve the permeability of the iron tailings-geotextile filter. Under stable seepage conditions, among the three types of geotextiles, staple fiber, namely filament and filter cloth geotextiles, the permeability coefficient of the filament geotextile filter is the highest, followed by that of the filter cloth geotextile filter, and the staple fiber geotextile filter exhibits the lowest permeability coefficient. The permeability coefficient of the staple fiber geotextile filter increases significantly starting from a vacuum negative pressure of 70 kPa and then increases with the increase of vacuum negative pressure. For the filament and filter cloth geotextile filters, the maximum permeability coefficients are obtained at a vacuum negative pressure of 70 kPa, increasing by 94.6 and 14.2 times, respectively, compared with the stable permeability coefficients without vacuum action. The improvement effect of vacuum action on the permeability characteristics of the geotextile filter is influenced by the fiber thickness and fabric thickness of the geotextiles. For geotextiles composed of coarser fibers or those with greater thickness, the improvement effect of vacuum action on the permeability characteristics of the filter is more pronounced.

Abstract:To achieve rapid and accurate estimation of the shear strength indicators of calcareous sand under emergency conditions, drained triaxial tests were conducted to investigate the strength and dilatancy characteristics of uncemented calcareous sand. By varying the relative density and effective confining pressure of specimens, the factors influencing the internal friction angle of uncemented calcareous sand were analyzed, and empirical formulas were proposed to accurately predict the peak internal friction angle and dilatancy angle of uncemented calcareous sand. The test results show that the peak internal friction angle and dilatancy angle of calcareous sand are significantly affected by relative density. The formula parameters modified based on the test data can improve the prediction reliability for the peak internal friction angle and dilatancy angle of uncemented calcareous sand. The prediction errors of the peak internal friction angle are mostly less than 25%. The empirical formulas can be used for rapid estimation of the shear strength indicators of uncemented calcareous sand specimens with a post-consolidation relative density ranging from 31% to 83% and a mean effective normal stress at the stress peak ranging from 100 kPa to 1 600 kPa.