Search

Hydrodynamic models are often used to obtain insights into potential dike breaches, because dike breaches can have severe consequences. However, their high computational cost makes them unsuitable for real-time flood forecasting. Machine learning models are a promising alternative, as they offer reasonable accuracy at a significant reduction in computation time. In this study, we explore the effectiveness of a Long Short-Term Memory (LSTM) neural network in fast flood modelling for a dike breach in the Netherlands, using training data from a 1D–2D hydrodynamic model. The LSTM uses the outflow hydrograph of the dike breach as input and produces water depths on all grid cells in the hinterland for all time steps as output. The results show that the LSTM accurately reflects the behaviour of overland flow: from fast rising and high water depths near the breach to slowly rising and lower water depths further away. The water depth prediction is very accurate (MAE = 0.045 m, RMSE = 0.13 m), and the inundation extent closely matches that of the hydrodynamic model throughout the flood event (Critical Success Index = 94%). We conclude that machine learning techniques are suitable for fast modelling of the complex dynamics of dike breach floods. [ABSTRACT FROM AUTHOR]

Recent flood disasters caused by extreme meteorological events highlight the need of fast and reliable tools for flooding forecast. For our purposes, the danger associated with floods is embodied in a single risk-level flag which considers both local water depth and velocity. The methodology here derived is applied and validated for the case study of the St. Lucia island in the eastern Caribbean Sea that experiences flash flooding as a result of combined intense rainfall and steep slopes, difficult to predict with traditional early-warning systems. A multi-layer perceptron neural network is trained on a high-fidelity dataset generated through full two-dimensional shallow water simulations of real and synthetic events. The dataset is validated against social markers obtained from real events. The predictive capabilities of the neural network model are tested on the out-of-box case of the Dean and Tomas hurricanes and compared with the solutions of the shallow water solver. The surrogate solver allows a significant speed-up in the prediction time with respect to traditional CFD (seconds vs hours), showing a high precision and accuracy, with accuracy, precision and F1-score above 0.99. [ABSTRACT FROM AUTHOR]

A high failure rate is associated with fracture plates in proximal humerus fractures. The causes of failure remain unclear due to the complexity of the problem including the number and position of the screws, their length and orientation in the space. Finite element (FE) analysis has been used for the analysis of plating of proximal humeral fractures, but due to computational costs is unable to fully explore all potential screw combinations. Surrogate modelling is a viable solution, having the potential to significantly reduce the computational cost whilst requiring a moderate number of training sets. This study aimed to develop adaptive neural network (ANN)‐based surrogate models to predict the strain in the humeral bone as a result of changing the length of the screws. The ANN models were trained using data from FE simulations of a single humerus, and after defining the best training sample size, multiple and single‐output models were developed. The best performing ANN model was used to predict all the possible screw length configurations. The ANN predictions were compared with the FE results of unseen data, showing a good correlation (R2 = 0.99) and low levels of error (RMSE = 0.51%–1.83% strain). The ANN predictions of all possible screw length configurations showed that the screw that provided the medial support was the most influential on the predicted strain. Overall, the ANN‐based surrogate model accurately captured bone strains and has the potential to be used for more complex problems with a larger number of variables. [ABSTRACT FROM AUTHOR]

Surrogate-Assisted Evolutionary Optimisation algorithms are a specialized brand of optimisers developed to undertake problems with computationally expensive fitness functions. These algorithms work by building a cheap approximation or model of the exact function and using it in the evaluation of solutions within the optimisation process. This use of modelling techniques within optimisation, while offers a practical reduction in function calls, brings along with it some additional questions. This paper starts with a description of the key elements of surrogate-assisted evolutionary optimisation algorithms as they are outlined throughout the literature, and then, proceeds to rearrange these elements using a novel blueprint of the field. The proposed blueprint can be used to represent any surrogate-assisted evolutionary algorithm in a way that illustrates its principles and components in a non-vague manner. In addition, a survey of the most prominent works in the field is conducted using this novel blueprint. Finally, a number of challenges and perspectives are listed before the paper is concluded. [ABSTRACT FROM AUTHOR]

Abstract Accurately describing the evolution of water droplet size distribution in crude oil is fundamental for evaluating the water separation efficiency in dehydration systems. Enhancing the separation of an aqueous phase dispersed in a dielectric oil phase, which has a significantly lower dielectric constant than the dispersed phase, can be achieved by increasing the water droplet size through the application of an electrostatic field in the pipeline. Mathematical models, while being accurate, are computationally expensive. Herein, we introduced a constrained machine learning (ML) surrogate model developed based on a population balance model. This model serves as a practical alternative, facilitating fast and accurate predictions. The constrained ML model, utilizing an extreme gradient boosting (XGBoost) algorithm tuned with a genetic algorithm (GA), incorporates the key parameters of the electrostatic dehydration process, including droplet diameter, voltage, crude oil properties, temperature, and residence time as input variables, with the output being the number of water droplets per unit volume. Furthermore, we modified the objective function of the XGBoost algorithm by incorporating two penalty terms to ensure the model’s predictions adhere to physical principles. The constrained model demonstrated accuracy on the test set, with a mean squared error of 0.005 and a coefficient of determination of 0.998. The efficiency of the model was validated through comparison with the experimental data and the results of the population balance mathematical model. The analysis shows that the initial droplet diameter and voltage have the highest influence on the model, which aligns with the observed behaviour in the real-world process.

Hypoxic zones and associated nitrate pollution from farms, cities and industrial facilities is driving declines in water quality that affect ecosystems, economies and human health in major rivers and coastal areas worldwide. In the Mississippi River, the United States Environmental Protection Agency set a goal of reducing nitrogen loading 20% by 2025, but estimating progress towards this goal is difficult because data from in‐stream gauges and laboratory samples are too sparse. Satellites have the potential to provide sufficient data across the Mississippi River, if a key methodological challenge can be overcome. Satellites provide data from visible light, but nitrates are only observable with ultraviolet light. We address this methodological challenge by using a two‐step surrogate modelling procedure to link optical data and nitrates in the Lower Mississippi River. First, we correlate in situ nitrate measurements to common water quality parameters, particularly turbidity and chlorophyll, using data from water sensors installed at Baton Rouge, Louisiana, USA, and a long‐term dataset from Louisiana State University. Second, we correlate these water quality data to satellite estimates of water quality parameters. We found a correlation between these water quality parameters and nitrate concentrations, as indicated by a coefficient of determination, when the relationship was viewed in nonlinear parameter space. The spatial extent of the correlation was tested with an upstream nitrate sensor 140 km north of the estimation location. These results provide proof of concept that we can develop models that use satellite data to provide large‐scale monitoring of nitrates across the Mississippi River Basin and other impaired rivers, globally. [ABSTRACT FROM AUTHOR]

This study aims to provide a comprehensive assessment of single-objective and multi-objective optimisation algorithms for the design of an elbow-type draft tube, as well as to introduce a computationally efficient optimisation workflow. The proposed workflow leverages deep neural network surrogates trained on data obtained from numerical simulations. The use of surrogates allows for a more flexible and faster evaluation of novel designs. The success history-based adaptive differential evolution with linear reduction and the multi-objective evolutionary algorithm based on decomposition were identified as the best-performing algorithms and used to determine the influence of different objectives in the single-objective optimisation and their combined impact on the draft tube design in the multi-objective optimisation. The results for the single-objective algorithm are consistent with those of the multi-objective algorithm when the objectives are considered separately. Multi-objective approach, however, should typically be chosen, especially for computationally inexpensive surrogates. A multi-criteria decision analysis method was used to obtain optimal multi-objective results, showing an improvement of 1.5% and 17% for the pressure recovery factor and drag coefficient, respectively. The difference between the predictions and the numerical results is less than 0.5% for the pressure recovery factor and 3% for the drag coefficient. As the demand for renewable energy continues to increase, the relevance of data-driven optimisation workflows, as discussed in this study, will become increasingly important, especially in the context of global sustainability efforts.

This study aims to provide a comprehensive assessment of single-objective and multi-objective optimisation algorithms for the design of an elbow-type draft tube, as well as to introduce a computationally efficient optimisation workflow. The proposed workflow leverages deep neural network surrogates trained on data obtained from numerical simulations. The use of surrogates allows for a more flexible and faster evaluation of novel designs. The success history-based adaptive differential evolution with linear reduction and the multi-objective evolutionary algorithm based on decomposition were identified as the best-performing algorithms and used to determine the influence of different objectives in the single-objective optimisation and their combined impact on the draft tube design in the multi-objective optimisation. The results for the single-objective algorithm are consistent with those of the multi-objective algorithm when the objectives are considered separately. Multi-objective approach, however, should typically be chosen, especially for computationally inexpensive surrogates. A multi-criteria decision analysis method was used to obtain optimal multi-objective results, showing an improvement of 1.5% and 17% for the pressure recovery factor and drag coefficient, respectively. The difference between the predictions and the numerical results is less than 0.5% for the pressure recovery factor and 3% for the drag coefficient. As the demand for renewable energy continues to increase, the relevance of data-driven optimisation workflows, as discussed in this study, will become increasingly important, especially in the context of global sustainability efforts. • Efficient draft tube design workflow based on DNN surrogates introduced. • Design flexibility ensured through B-spline parameterisation. • MOEA/D and L-SHADE identified as the best-performing algorithms. • A single-objective optimisation approach can be viable depending on the goals. • Multi-objective optimisation provides measurable performance improvements. [ABSTRACT FROM AUTHOR]

This work presents the development of a model-based online damage identification system for a 13th century rammed earth (RE) tower in the Alhambra, the Muhammad Tower. The system is fed with continuous data from an ambient vibration-based monitoring system and a meteorological station. Ambient vibrations are continuously processed through Operational Modal Analysis (OMA), and environmental effects are minimised via statistical pattern recognition. The normalized modal signatures are used to update the stiffness properties of certain parts of the tower through inverse model calibration. To do so, a high-fidelity three-dimensional finite element model (FEM) of the tower is developed. Since its computational burden precludes conducting online calibration, the FEM is bypassed by a light Kriging surrogate model (SM). In this light, the developed SM-assisted system identification constitutes a long-term Structural Health Monitoring (SHM) system outputting quasi-real-time series of modal properties and local stiffness parameters, so providing full damage assessment (detection, localization and quantification). The presented results refer to a time period of three months since January until March 2022. Numerical results and discussion are reported concerning the characterization and removal of environmental effects, and synthetic damage scenarios through non-linear simulations are used to validate the developed damage identification system. [ABSTRACT FROM AUTHOR]

Hydrodynamic models are often used to obtain insights into potential dike breaches, because dike breaches can have severe consequences. However, their high computational cost makes them unsuitable for real-time flood forecasting. Machine learning models are a promising alternative, as they offer reasonable accuracy at a significant reduction in computation time. In this study, we explore the effectiveness of a Long Short-Term Memory (LSTM) neural network in fast flood modelling for a dike breach in the Netherlands, using training data from a 1D–2D hydrodynamic model. The LSTM uses the outflow hydrograph of the dike breach as input and produces water depths on all grid cells in the hinterland for all time steps as output. The results show that the LSTM accurately reflects the behaviour of overland flow: from fast rising and high water depths near the breach to slowly rising and lower water depths further away. The water depth prediction is very accurate (MAE = 0.045 m, RMSE = 0.13 m), and the inundation extent closely matches that of the hydrodynamic model throughout the flood event (Critical Success Index = 94%). We conclude that machine learning techniques are suitable for fast modelling of the complex dynamics of dike breach floods.

Wind fences are commonly used to mitigate dust emission from coal stockpiles in an open storage yard. There has substantial progress in the trial-and-error study of traditional straight fences, but only limited research focused specifically on the shelter effect of curved fence and its shape optimization. This paper introduces an innovative wind fence with curved deflector, and applies the aerodynamic optimization to enhance its shelter efficiency. Computational fluid dynamics (CFD) techniques are employed to simulate the flow field around a triangular coal stockpile behind the fence using RNG k-ε turbulence model. The simulation results have confirmed the role of curved fences in reducing wind erosion through comparative analysis of different fence geometries. In order to efficiently and automatically explore the optimal design parameters including fence porosity and external shape, the aerodynamic optimization via surrogate modelling is implemented. Two conflicting objectives namely minimization of total absolute surface pressure coefficients and reduction of peak velocity ratios around the stockpile are considered. The weighted sum method is utilized to solve the proposed multi-objective model. The optimization results suggest that the optimal curved fence is associated with a significant improvement of shelter performance in comparison with the traditional straight fence. [ABSTRACT FROM AUTHOR]

Twin-box decks are being extensively used in the design of long-span bridges due to their superior flutter performance. Although the significant role played by the gap distance has been previously addressed in the frame of experimental studies, there is still a lack of understanding about the complex interplay between box geometry, gap distance and the aerodynamic force coefficients and flutter derivatives. In the present work, firstly, a surrogate model is developed, considering three design inputs for the geometry of the deck along with the angle of attack, providing the force coefficients as output. Afterwards, the work is then extended by developing another surrogate model, considering as inputs the reduced velocity and the same geometric variables of the deck, with the outputs being the flutter derivatives. The methodology, comprising the selection of the design domain, the definition of the samples, the CFD-based evaluation of the samples' response and the construction of the surrogate through the application of the neural network-based radial basis method, is reviewed. The surrogate models enable a quantitative description of the impact caused in the force coefficients and the flutter derivatives by modifications in the geometry of the twin-box deck. It has been found that flutter derivatives H1*, H2* and A2* are strongly dependent on the gap distance.

The railgun is an electromagnetic device that converts electrical energy to mechanical energy for accelerating the projectiles to hyper velocities, which is the main reason that railguns are becoming an increasingly popular topic of interest among the military, defense industry and scientific communities. Output force is used for evaluating the performance and the effectiveness of a railgun which is highly depended on the Railgun geometric design parameters. When a new railgun is designed, drawing the geometry, simulation and analysis stages take a long time; this study has aimed to provide an Artificial Intelligence based surrogate model for a railgun design to prevent time consuming at these stages and to make design optimization process computationally efficient. For this reason, in this paper three different rail geometries have been combined and simulated with three different armature types by using Ansys Maxwell which is based on 3-D Finite Element Method. Within the scope of this study; first of all, one of the best rail and armature pair was selected according to the efficiency. Secondly, a dataset with inputs and outputs was created by changing the geometric variables of the selected pair. Thirdly, using this dataset; six different surrogate models were trained, tuned and tested. Railgun's output force was predicted with minimum symmetric mean absolute percentage error (SMAPE) of 1.89% at the end of tests. Finally, Particle Swarm Optimization (PSO) was carried out with the surrogate model that gave the best result for modelling of a railgun design with 8.0 kNewton required output force. These optimization results were compared with the Ansys simulation outputs and found to be hand to hand. Thus, herein, a computationally efficient method for design optimization of railgun designs has been achieved using surrogate based modelling techniques. [ABSTRACT FROM AUTHOR]

Consideration of uncertainties in bridge structural engineering is a growing topic because it provides valuable information on the reliability of bridges in time while addressing life‐cycle aspects. Utilization of nonlinear finite element method (NLFEM) and Monte Carlo (MC) type simulation is essential for modelling of the concrete bridge random behaviour and its reliability assessments. This contribution describes keystones of a complex methodology for statistical, reliability and risk analyses of concrete structures including advanced nonlinear finite element analysis, uncertainty propagation and degradation modelling. It presents virtual simulation used on the way from assessment of experimental results to reliability analysis of a bridge structure, based on randomization of nonlinear fracture mechanics finite element analysis. Keystones of complex methodology are material parameters identification, nonlinear computational modelling, uncertainties propagation, surrogate modelling, degradation modelling, and safety formats. The paper describes methodology and for a routine complex assessment utilizing developed software tools and practical application of the approach for selected case studies of concrete bridges from Austria and the Czech Republic. [ABSTRACT FROM AUTHOR]

Decarbonizing buildings is crucial in addressing pressing climate change issues. Buildings significantly contribute to global greenhouse gas emissions, and reducing their carbon footprint is essential to achieving sustainable and low-carbon cities. Retrofitting buildings to become more energy efficient constitutes a solution. However, building energy retrofits are complex processes that require a significant number of simulations to investigate the possible options, which limits comprehensive investigations that become infeasible to carry out. Surrogate models can be vital in addressing computational inefficiencies by emulating physics-based models and predicting building performance. However, there is a limited focus on investigating feature engineering and selection methods and their effect on the model's performance and optimization. Feature selection methods are considered effective with interpretable models such as multi-variate linear regression (MVLR) and multiple adaptive regression splines (MARS) for achieving stable prediction stability. This study proposes a modelling framework to create, optimize, and improve the performance of surrogate predictive models for energy consumption, carbon emissions, and the associated cost of building energy retrofit processes. The investigated feature selection methods are wrapper and embedded methods such as backward-stepwise feature selection (BSFS), recursive feature elimination (RFE), and Elastic Net embedded regularization in order to provide insights into the model's behavior and optimize the model's performance. The most accurate surrogate models developed achieved a mean absolute percentage error (MAPE) of 0.2–1.8% compared to the used test data. In addition, when calculated for a million samples, all developed surrogate models reduced the computational time by one-thousand-fold compared to physics-based models. The study's findings pave the way towards low-computational accurate models that can comprehensively predict building performance in near real-time, ultimately leading to identifying decarbonization measures at scale. [ABSTRACT FROM AUTHOR]

Neural network (NN) ensembles can reduce large prediction variance of NN and improve prediction accuracy. For highly non-linear problems with insufficient data set, the prediction accuracy of NN models becomes unstable, resulting in a decrease in the accuracy of ensembles. Therefore, this study proposes a prediction frequency-based ensemble that identifies core prediction values, which are core prediction members to be used in the ensemble and are expected to be concentrated near the true response. The prediction frequency-based ensemble classifies core prediction values supported by multiple NN models by conducting statistical analysis with a frequency distribution, which is a collection of prediction values obtained from various NN models for a given prediction point. The prediction frequency-based ensemble searches for a range of prediction values that contains prediction values above a certain frequency, and thus the predictive performance can be improved by excluding prediction values with low accuracy and coping with the uncertainty of the most frequent value. An adaptive sampling strategy that sequentially adds samples based on the core prediction variance calculated as the variance of the core prediction values is proposed to improve the predictive performance of the prediction frequency-based ensemble efficiently. Results of various case studies show that the prediction accuracy of the prediction frequencybased ensemble is higher than that of Kriging and other existing ensemble methods. In addition, the proposed adaptive sampling strategy effectively improves the predictive performance of the prediction frequency-based ensemble compared with the previously developed space-filling and prediction variance-based strategies. [ABSTRACT FROM AUTHOR]

Magnesium (Mg) and its alloys are promising materials for temporary bone implants due to their mechanical properties and biocompatibility. The most challenging aspect of Mg-based implants involves adapting the degradation rate to the human body, which requires extensive in vitro and in vivo testing. Given that in vivo tests are significantly more labour-intensive than in vitro and ethics prohibit direct experiments on animals or humans, attempts are commonly undertaken to infer conclusions on in vivo degradation behavior from in vitro experiments. However, there is a wide gap between these tests, and in vitro testing is often a poor predictor of in vivo outcomes. In the development of biodegradable Mg-based implants, considerable efforts are being made to reduce the overall time and cost of in vitro and in vivo testing. Finding a suitable alternative to predict the degradation of Mg alloys, however, remains challenging. We present computational modelling as a possible alternative to bridge the gap between in vitro and in vivo testing, thus reducing overall cost, duration and number of experiments. However, traditional modelling approaches for complex biodegradable systems are still rather time-consuming and require a clear definition of the relations between input parameters and the model result. In this study, Kriging surrogate models based on the peridynamic in vitro degradation model were developed to simulate the degradation behavior for two main alloys, Mg-5Gd and Mg-10Gd, for both in vitro and in vivo cases. Using Kriging surrogate models, the simulation parameters were calibrated to the volume loss data from in vitro and in vivo experiments. In vivo degradation of magnesium has one order of magnitude higher apparent diffusion coefficients than in vitro degradation, thus yielding the higher volume loss observed in vivo than in vitro. On the basis of the diffusivity of the Mg2+ ions modeled under in vitro degradation, Kriging surrogate models were able to simulate the in vivo degradation behavior of Mg-xGd with a ratio between 0.46 and 0.5, indicating that the surrogate-modelling approach is able to bridge the gap between in vitro and in vivo degradation rates for Mg-xGd implants.

Hydrological simulation has long been a challenge because of the computationally intensive and expensive nature of complex hydrological models. In this paper, a surrogate modelling (SM) framework is presented based on the Hadoop cloud for approximating complex hydrological models. The substantial model runs required by the design of the experiment (DOE) of SM were solved using the Hadoop cloud. Polynomial chaos expansion (PCE) was fitted and verified using the high-fidelity model DOE and was then used as a case study to investigate the approximation capability in a Soil and Water Assessment Tool (SWAT) surrogate model with regard to the accuracy, fidelity, and efficiency. In experiments, the Hadoop cloud reduced the computation time by approximately 86% when used in a global sensitivity analysis. PCE achieved results equivalent to those of the standard Monte Carlo approach, with a flow variance coefficient of determination of 0.92. Moreover, PCE proved to be as reliable as the Monte Carlo approach but significantly more efficient. The proposed framework greatly decreases the computational costs through cloud computing and surrogate modelling, making it ideal for complex hydrological model simulation and optimization. HIGHLIGHTS Our surrogate modelling framework reduces the computational cost of simulations.; The design of the experiment was parallelized on a Hadoop cloud.; PCE was fitted and verified using a high-fidelity model.; The approximation ability of PCE in the SWAT surrogate model was investigated.;

Majority of the past research on application of machine learning (ML) in earthquake engineering focused on contrasting the predictive performance of different ML algorithms. In contrast, the emphasis of this paper is on the use of data to boost the predictive performance of surrogates. To that end, a novel data engineering methodology for seismic collapse risk assessment is proposed. This method, termed the automated collapse data constructor (ACDC), stems from combined understanding of the ground motion characteristics and the collapse process. In addition, the data‐driven collapse classifier (D2C2) methodology is proposed which enables conversion of the collapse data from a regression format to a classification format. The D2C2 methodology can be used with any classification tool, and it allows estimation of seismic collapse capacities in a way analogous to the incremental dynamic analysis. The proposed methodologies are tested in a case study using decision trees (XGBoost) and neural network classifiers with an extensive dataset of collapse responses of a 4‐story and an 8‐story steel moment resisting frames. The results suggest that the ACDC methodology allows for dramatic improvement of the predictive performance of data‐driven tools while at the same time significantly reducing data requirements. Specifically, the proposed method can reduce the number of ground motions required for collapse risk assessment from at least forty, as traditionally used, to less than twenty motions. Moreover, interpretation of feature importance conforms with the engineering understanding while revealing a novel, period‐dependent measure of ground motion duration. All data and code developed in this research are made openly available. [ABSTRACT FROM AUTHOR]

Characterization of an existing building's energy-related features is critical to inform maintenance and retrofit decisions. However, existing field-scale characterization methods tend to be labour intensive, invasive, and require high fidelity longitudinal data gathered through tightly regulated experiments. This highlights the need for a low cost, scalable, and efficient screening method. This paper puts forward a surrogate model-based approach to rapidly estimate energy-related building features. To this end, EnergyPlus models for 12 midrise office archetypes, all with a rectangular footprint, are developed. Ten thousand variants of each archetype are generated by altering envelope, causal heat gain, and heating, ventilation, and air conditioning operation features. A unique load signature is derived for each variant's heating and cooling energy use. The parameters of the load signatures are clustered, then each cluster is associated with a set of plausible energy-related features. The accuracy of the results was evaluated using five test buildings not seen by the algorithm. The method could effectively identify building features with reasonable accuracy and no significant degradation in performance across all 12 archetypes. [ABSTRACT FROM AUTHOR]

Physics-informed neural network (PINN) architectures are recent developments that can act as surrogate models for fluid dynamics in order to reduce computational costs. PINNs make use of deep neural networks, where the Navier-Stokes equation and freestream boundary conditions are used as losses of the neural network; hence, no simulation or experimental data in the training of the PINN is required. Here, the formulation of PINN for fluid dynamics is demonstrated and critical factors influencing the PINN design are discussed through a low Reynolds number flow over a cylinder. The PINN architecture showed the greatest improvement to the accuracy of results from the increase in the number of layers, followed by the increase in the number of points in the point cloud. Increasing the number of nodes per hidden layer brings about the smallest improvement in performance. In general, PINN is much more efficient than computational fluid dynamics (CFD) in terms of memory resource usage, with PINN requiring 5–10 times less memory. The tradeoff for this advantage is that it requires longer computational time, with PINN requiring approximately 3 times more than that of CFD. In essence, this paper demonstrates the direct formulation of PINN without the need for data, alongside hyperparameter design and comparison of computational requirements. [ABSTRACT FROM AUTHOR]

A robust optimum solution is defined as one that is optimum and maintains its feasibility regardless of the values taken by uncertain parameters within a bounded interval. Solving for a robust optimum solution, however, can be computationally costly. To reduce the computational cost, a new robust optimization method, surrogate feasibility testing–cutting robust optimization (SFTC-RO), is proposed that solves two optimization subproblems. In the first subproblem, a scenario robust optimization problem is solved using a local optimization technique. In the second subproblem, a feasibility check and constraint cut of the feasible domain are performed via operations on a surrogate model of the constraints. The two subproblems are iteratively solved until convergence at a robust optimum solution. Results from numerical and engineering examples show that on average the proposed method can obtain robust optimal solutions at a lower computational cost and with better scalability than an existing method from the literature. [ABSTRACT FROM AUTHOR]

Magnesium (Mg) and its alloys are promising materials for temporary bone implants due to their mechanical properties and biocompatibility. The most challenging aspect of Mg-based implants involves adapting the degradation rate to the human body, which requires extensive in vitro and in vivo testing. Given that in vivo tests are significantly more labour-intensive than in vitro and ethics prohibit direct experiments on animals or humans, attempts are commonly undertaken to infer conclusions on in vivo degradation behavior from in vitro experiments. However, there is a wide gap between these tests, and in vitro testing is often a poor predictor of in vivo outcomes. In the development of biodegradable Mg-based implants, considerable efforts are being made to reduce the overall time and cost of in vitro and in vivo testing. Finding a suitable alternative to predict the degradation of Mg alloys, however, remains challenging. We present computational modelling as a possible alternative to bridge the gap between in vitro and in vivo testing, thus reducing overall cost, duration and number of experiments. However, traditional modelling approaches for complex biodegradable systems are still rather time-consuming and require a clear definition of the relations between input parameters and the model result. In this study, Kriging surrogate models based on the peridynamic in vitro degradation model were developed to simulate the degradation behavior for two main alloys, Mg-5Gd and Mg-10Gd, for both in vitro and in vivo cases. Using Kriging surrogate models, the simulation parameters were calibrated to the volume loss data from in vitro and in vivo experiments. In vivo degradation of magnesium has one order of magnitude higher apparent diffusion coefficients than in vitro degradation, thus yielding the higher volume loss observed in vivo than in vitro. On the basis of the diffusivity of the Mg 2 +  ions modeled under in vitro degradation, Kriging surrogate models were able to simulate the in vivo degradation behavior of Mg-xGd with a ratio between 0.46 and 0.5, indicating that the surrogate-modelling approach is able to bridge the gap between in vitro and in vivo degradation rates for Mg-xGd implants. [ABSTRACT FROM AUTHOR]

Phased Array Antenna (PAA) technology plays an important role in fields such as radar, 5G and satellite or any application which requires wide bandwidth and high gain. However, achieving such design is a difficult and complex task that requires an accurate calculation and combination of results obtained for varying phase and amplitude of each unit and coupling effects between these elements of the PAA structure is a task that can only be obtained using full wave EM simulation tools. This comes at the price of a significant increase for the computational cost of the design process which is a well-known drawback of forward EM modeling of microwave stages most especially in case of repetitive analysis’s such as yield analyses or optimization tasks. Data-driven surrogate models have emerged as a powerful and versatile solution that bridges the gap between computationally expensive simulations and rapid, reliable prediction models suitable for deployment in applications such as optimization and/or yield analyses. Herein, for having a high-performance broadband PAA for millimeter band in a computationally efficient manner, artificial intelligence based surrogate model assisted optimization approach is deployed. A series of state-of-the-art surrogate modeling algorithms are deployed to create a surrogate model of the studied PAA design for the prediction of radiation pattern characteristic with respect to the input phase values of each array element. As a result, a drastic reduction in computational time of almost 90% for the optimization of three PAA designs is achieved. Thus, the proposed approach offers promising avenues for further exploration in computational electromagnetics, most especially in simulation expensive problems with complex designs.