The development of floating offshore wind turbines (FOWTs) is gradually moving into deeper offshore areas with more harsh environmental loads, and the corresponding structure response should be paid attention to. Safety assessments need to be conducted based on the evaluation of the long-term extreme response under operating conditions. However, the full long-term analysis method (FLTA) recommended by the design code for evaluating extreme response statistics requires significant computational costs. In the present study, a power response prediction method for FOWT based on an artificial neural network algorithm is proposed. FOWT size, structure, and training algorithms from various artificial neural network models to determine optimal network parameters are investigated. A publicly available, high-quality operational dataset is used and processed by the Inverse First Order Reliability Method (IFORM), which significantly reduces simulation time by selecting operating conditions and directly yielding extreme response statistics. Then sensitivity analysis is done regarding the number of neurons and validation check values. Finally, the alternative dataset is used to validate the model. Results show that the proposed neural network model is able to accurately predict the extreme response statistics of FOWT under realistic in situ operating conditions. A proper balance was achieved between prediction accuracy, computational costs, and the robustness of the model.

The long-term safety of pressure-resistant structures used in deep-sea equipment may be threatened by creep deformation. The creep deformation behavior of a pressure-resistant structure made of different titanium alloys, Ti-6Al-4V and Ti-4Al-2V, at room temperature is investigated in this research. The kinetics and mechanisms underlying creep deformation in these materials is explained by proposing an improved constitutive model considering the effects of stress level, loading rate and environmental temperature field, offering crucial information for optimizing design parameters and guaranteeing the lifespan of the structure. Model parameters are determined for the two types of titanium alloys based on tensile creep testing results and validated through a simulation of the experimental process. In this study, a material creep model was used to predict the long-term deformation of large pressure-resistant titanium structures to ensure safe long-term operation. The safety factor used in the model is 1.5. Finite element analyses are conducted for the creep behavior of the pressure-resistant structure under real operating circumstances based on the creep constitutive model. The simulation predicts stress distribution, strain evolution, and deformation size over long periods of time by integrating complicated geometries, boundary conditions, and material characteristics. The present research can provide basic information for the local impacts of creep deformation on the inside of facilities, which helps refine design strategies to reduce possible damage risks.

Surface-focused monitoring of karst hillslopes can overlook subsurface critical-zone pathways, leaving a fundamental question unresolved: do nitrate routing and export magnitude share the same functional controls? We monitored surface runoff (SR), subsurface runoff at the soil–epikarst interface (SsR), and epikarst runoff at the epikarst–bedrock interface (EpR) from a karst hillslope platform in southwest China. Forty-seven erosive rainfall events across 10 plots were analysed for runoff generation, dissolved nitrogen, water isotopes (δ2H-H2O and δ18O-H2O), and nitrate isotopes (δ15N-NO3⁻ and δ18O-NO3⁻). EpR had the highest runoff coefficient (0.18 ± 0.15), while SR had the highest dissolved N concentrations. This concentration–export decoupling caused EpR to dominate annualized NO3⁻-N export (0.44 ± 0.33 kg N ha⁻1 yr⁻1) through a volume-driven mechanism in which a large runoff coefficient offset a lower solute concentration. Water and nitrate isotopes indicated pathway-dependent water and nitrate regimes, consistent with storage-water displacement rather than simple dilution in EpR. Hierarchical partitioning showed that critical-zone structure was the primary control on NO3⁻-N partitioning to EpR (46.8% of explained variance), whereas meteorological factors dominated total NO3⁻-N export (69.9%). Rainfall intensity further redistributed pathway contributions and amplified land-use differences, with maize exporting approximately ten times more NO3⁻-N than pasture under storm rainfall. These findings demonstrate that critical-zone architecture predisposes nitrate routing as a structural template, while rainfall activates hydrological connectivity and governs export magnitude. Multi-interface monitoring is therefore essential for accurately assessing nitrate loss and groundwater contamination risk in karst hillslopes.

For the Great Lakes Basin, the establishment of an ecological network, coupled with resilience assessment research, is pivotal for bolstering regional ecological resilience and facilitating effective ecological conservation initiatives. However, current research has overlooked the dynamic changes of ecological networks undergo under external disturbances and the varying impacts of different ecological sources and corridors on resilience. This paper took Dongting Lake Basin as an example, constructing an ecological network based on circuit theory. It further integrates complex network topology indicators to establish a resilience assessment framework. This framework was used to analyze resilience dynamics under scenario simulation, to identify key ecological spaces within the basin. Our findings indicated that forest land plays a significant role in the composition of these ecological sources. 70% of the ecological pinchpoints were concentrated on the four corridors, and ecological barriers were mainly interspersed between ecological sources. For the simulation of resilience dynamics, key ecological sources and corridors were instrumental in determining system resilience. Notably, the key sources were predominantly situated in the central watershed, demonstrating robust connectivity to the peripheral sources. Meanwhile, successive failures had an amplifying impact on single failures, resulting in an increase in toughness loss by 8.57% and 13.69% for failures involving 3 and 4 nodes respectively. This research contributes to the prioritization of essential ecological spaces within the basin, offering invaluable references for decision-making in the targeted optimization and management of ecological construction areas in the Dongting Lake basin.

Nitrogen (N) availability is a critical determinant of ecosystem productivity in karst forests, yet the spatial patterns of soil N pools and their underlying microbial mechanisms at the rock-soil interface remain poorly understood across different lithologies. This study aims to elucidate how limestone and dolostone lithologies, coupled with the rock-soil interface effect, regulate N cycling processes. We investigated soil N fractions and N-cycling functional gene abundances in both bulk and rock-soil interface soils within secondary forests overlying dolostone and limestone in a typical karst region. Four treatments were compared: limestone bulk soil (LB), limestone interface soil (LI), dolostone bulk soil (DB), and dolostone interface soil (DI). Two-way ANOVA revealed that the rock-soil interface significantly enriched all measured N parameters (p < 0.05–0.001), with a more pronounced effect in limestone soils where NH4+, NO3−, TDN, and DON were significantly elevated. Conversely, lithology exerted a profound impact on NO3− and SMN, which were generally higher in dolostone than in limestone sites. Interestingly, while N pools were higher in certain dolostone fractions, the abundances of most N-cycling functional genes (e.g., nifH, ureC, narG, and nirS) were markedly higher in limestone soils, except for AOA amoA, which peaked in dolostone soils (p < 0.001). Random forest and Mantel’s tests identified that soil reactive minerals ((Ca+Mg)e) and enzymatic stoichiometry (BG/(LAP+NAG)) were the primary determinants of soil N pools, whereas N functional gene abundances were more sensitive to amorphous minerals ((Fe+Al)o), microbial biomass carbon (MBC), and viral diversity (Virusesα). Our results demonstrate that the rock-soil interface acts as a biogeochemical hotspot for N accumulation and microbial activity, though the magnitude of this effect is modulated by lithology. These findings suggest that reactive minerals and enzymatic stoichiometry are key mediators of N cycling, providing new insights into the geochemical controls of soil fertility in fragile karst carbonate landscapes.