The expansion of offshore wind farms, driven by better offshore wind conditions and fewer spatial limitations, has promoted the growth of this technology. This study focuses on the design of jacket support structures for Offshore Wind Turbines, which are suitable for deeper waters. However, the structural analysis required for designing these structures is computationally intensive due to multiple load cases and numerous checks. To reduce this computational cost, artificial-neural-network-based surrogate models capable of estimating the feasibility of a jacket structure acting as the support structure for any given wind turbine at a specific site are developed. A synthetic dataset generated through random sampling and evaluated by a structural model is utilized for training and testing the models. Two kind of models are compared: one is trained to estimate global feasibility, while the other estimates compliance with each of the structural partial requirements. Also, several assembly methods are proposed and compared. The best-performing model shows great classification metrics, with a Matthews Correlation Coefficient of 0.674, enabling an initial assessment of the structural feasibility. The low computational cost of artificial neural networks compared to structural models makes this surrogate model useful for accelerating otherwise prohibitive parametric studies or optimization processes.

As the threat of natural disasters to structures intensifies, risk assessment of infrastructure has gained much importance. Fragility curves are essential tools in predicting disaster-related losses and making disaster mitigation decisions. In this paper, we propose a new method to efficiently derive accurate fragility curves for structures with high levels of nonlinearity or complexity, addressing the computational challenges of conventional finite element reliability analysis (FERA). To reduce the computational cost for calculating probability of failure in FERA, the proposed method utilizes the first-order reliability method (FORM). However, even with this approach, the computational cost of deriving the fragility curve may remain high; therefore, a surrogate model is used to further reduce costs. By training the surrogate model using the initial structural damage probabilities for a few hazard intensities, an optimal starting point can be calculated for the subsequent FORM analysis. During the fragility analysis, the surrogate model can be updated sequentially to increase the efficiency of FORM analysis continuously. In particular, the training process of the surrogate model requires no separate or additional finite element analysis because it is constructed using previous FERA results. The accuracy and efficiency of the proposed method are tested using conventional FERA and Monte Carlo simulations through a hypothetical short-column example. In addition, fragility curves are derived through a bridge flood fragility assessment considering the scour and seismic vulnerability assessment of a buried gas pipeline considering soil-structure interactions. The derived fragility curves closely match those derived using the conventional FERA, and the computational costs are reduced by 36.54 % and 52.38 %, respectively, compared with the conventional FERA, confirming its cost-effectiveness.

Ground heat flux (G0) is a key component of the land-surface energy balance of high-latitude regions. Despite its crucial role in controlling permafrost degradation due to global warming, G0 is sparsely measured and not well represented in the outputs of global scale model simulation. In this study, an analytical heat transfer model is tested to reconstruct G0 across seasons using soil temperature series from field measurements, Global Climate Model, and climate reanalysis outputs. The probability density functions of ground heat flux and of model parameters are inferred using available G0 data (measured or modeled) for snow-free period as a reference. When observed G0 is not available, a numerical model is applied using estimates of surface heat flux (dependent on parameters) as the top boundary condition. These estimates (and thus the corresponding parameters) are verified by comparing the distributions of simulated and measured soil temperature at several depths. Aided by state-of-the-art uncertainty quantification methods, the developed G0 reconstruction approach provides novel means for assessing the probabilistic structure of the ground heat flux for regional permafrost change studies. Ground heat flux is the energy that goes into or comes out from belowground that controls the soil freeze-thaw process in high-latitude regions. Its changes under climate warming will influence variations in the soil's seasonal thawing depth and permafrost thickness and spatial extent. Available data on ground heat flux are very sparse from both direct field measurements and large-scale model outputs in the Arctic. This study combines detailed modeling and uncertainty quantification methods to accurately reconstruct the ground heat flux from shallow soil temperature observations and estimates from predictive models, which are more readily available for the Arctic. Since the approach relies on several assumptions, we also quantify the uncertainty of the estimated ground heat flux. The reconstructed ground heat fluxes using the method developed in this study match well with the fluxes observed or derived from the predictive model. The soil properties inferred from the developed process are also consistent with the values observed for typical soils. Ground heat flux is reconstructed from various types of shallow soil temperature and auxiliary data using an analytical heat transfer model Uncertainty quantification methods are applied to infer model parameters and increase simulation efficiency drastically The efficacy of the proposed ground heat flux reconstruction framework is shown by agreement between simulation and observation

There is a complex multifactorial coupling effect among the damages of various protection structures on slopes. Existing research focused on the health assessment of individual structures is often insufficient in representing the overall health status of the protection engineering system. Considering the characteristics of expansive soil slope protection engineering, this study proposes a health diagnosis method using combined weights and binary K-means clustering algorithm. The method quantifies the damage data of protection structures based on subjective and objective weights, and clusters the data by combining the binary K-means method and target vector layer to obtain the diagnosis results. Furthermore, an XGBoost-based surrogate diagnosis model is constructed to omit the repetitive modelling process in practical applications to achieve dynamic diagnosis. The proposed method is validated to an expansive soil slope in Gaochun district, Nanjing. The results show that the proposed method can accurately evaluate protection engineering with different degrees of damage; the surrogate model follows the same weight assignment process as the diagnostic method to establish reliable prediction. Based on the proposed method, damage coupling effects between individual protection structures are captured, and targeted maintenance and repair can be implemented. The proposed method can be further extended to other slope engineering.