Gaussian Process Regression Model Research Articles

Machine learning assisted improved desalination pilot system design and experimentation for the circular economy

Desalination is among the most feasible solutions to supply sustainable and clean drinking water in water scarcity areas. In this regard, Multi-Effect Desalination (MED) systems are particularly preferred for harsh feeds (high temperature and salinity) because of their robust mode of operation for water production. However, maintaining the efficient operation of the MED systems is challenging because of the large system design and variables' interdependencies that are sensitive to the distillate production. Therefore, this research leverages the power of machine learning and optimization to estimate the optimal operating conditions for the maximum distillate production from the MED system. In the first step, detailed experimentation is conducted for distillate production against hot water temperature (HWT) varying from 38 to 70 °C, and feed water temperature (FWT) is changed from 34 to 42 °C. Whereas, the feed flow rate (FFR) is investigated to be varied nearly from 3.6 to 8.7 LPM in the three stages, i.e., FFR-S1, FFR-S2 and FFR-S3. The compiled dataset is used to make the process models of the MED system by three ML-based algorithms, i.e., Artificial Neural Network (ANN), Support Vector Machine (SVM), and Gaussian Process Regression (GPR) under rigorous hyperparameters optimization. GPR exhibited superior predictive performance than those of ANN and SVM on R2 value of 0.99 and RMSE of 0.026 LPM. Monte Carlo technique-based variable significance analysis revealed that the HWT has the highest effect on distillate production with a percentage significance of 95.6 %. Then Genetic Algorithm is used to maximize the distillate production with the GPR model embedded in the optimization problem. The GPR-GA driven maximum distillate production is estimated on HWT = 70 ± 0.5 °C, FWT = 40 ± 2.5 °C, FFR-S1 = 6 ± 2.6 LPM, FFR-S2 = 7 ± 1 LPM and FFR-S3 = 7 ± 1. The ML-GA-based system analysis and optimization of the MED system can boost the distillate production that promotes operation excellence and circular economy from the desalination sector.

Utilizing undisturbed soil sampling approach to predict elastic modulus of cohesive soils: a Gaussian process regression model

Utilizing undisturbed soil sampling approach to predict elastic modulus of cohesive soils: a Gaussian process regression model

Bayesian dynamic modelling for probabilistic prediction of pavement condition

Significant funds have been allocated to maintain road networks each year in developed countries. Performance prediction is crucial for pavement management systems to adjust working plans and budget allocation. As a Bayesian nonparametric method, Gaussian process regression (GPR) is powerful in predicting nonlinear time series and quantifying uncertainty. However, it remains computationally intensive and fails to adapt to the time-varying characteristics. To address such issues, a dynamic GPR model is proposed for probabilistic prediction of the International Roughness Index (IRI) for flexible pavements. A moving window strategy is developed to substantially shrink the size of training data, which effectively alleviates computational cost and thus leads to a dynamic GPR. A genetic algorithm is then adopted to determine the optimal window size by considering the trade-off between computational efficiency and accuracy. A dataset acquired from Long-Term Pavement Performance (LTPP) is used to demonstrate the feasibility of the dynamic GPR. Its performance is compared to traditional GPR as well as dynamic and static Bayesian linear regression (BLR) models. The comparison results indicate that the proposed dynamic GPR can increase the accuracy by 0.86, 1.52, and 2.27 times for dynamic BLR, static GPR, and static BLR, respectively. It exhibits the best results in terms of accuracy and uncertainty metrics due to its nonlinear modelling and time-varying ability.

Towards autonomous learning and optimisation in textile production: data-driven simulation approach for optimiser validation

AbstractThe textile industry is a traditional industry branch that remains highly relevant in Europe. The industry is under pressure to remain profitable in this high-wage region. As one promising approach, data-driven methods can be used for process optimisation in order to reduce waste, increase profitability and relieve mental burden on staff members. However, approaches from research rarely get adopted into practice. We identify the high dimensionality of textile production processes leading to high model uncertainty as well as an incomplete problem formulation as the two main problems. We argue that some form of an autonomous learning agent can address this challenge, when it safely explores advantageous, unknown new settings by interacting with the process. Our main goal is to facilitate the adoption of promising research into practical applications. The main contributions of this paper include the derivation and formulation of a probabilistic optimisation problem for high-dimensional, stationary production processes. We also create a highly adaptable simulation of the textile carded nonwovens production process in Python that implements the optimisation problem. Economic and technical behavior of the process is approximated using both Gaussian Process Regression (GPR) models trained with industrial data as well as physics-motivated explicit models. This ’simulation first’-approach makes the development of autonomous learning agents for practical applications feasible because it allows for cheap testing and validation before physical trials. Future work will include the comparison of the performance of different agent approaches.

Integrated form-position measurement of large-aperture transparent elements based on stereoscopic phase measuring deflectometry

Abstract In the field of ultra-precision manufacturing and measurement, sub-aperture stitching is a widely employed technique for the measurement of complex optical components. Those influencing factors like system errors and numerical bias introduce notable stitching errors in the overlapped areas between sub-apertures, consequently degrading measurement accuracy. In addition, the surfaces forms, refractive indices, and positions are difficult to be specified simultaneously. In this paper, an integrated measurement method based on the stereo deflectometry is proposed, which can measure the surface form, refractive index, and relative position of large-aperture transparent components together. This method utilizes a Gaussian process regression model to decouple and predict the refractive index and relative positions of the upper and lower surfaces. Then, a robust sub-aperture stitching technique with an adaptive distance function is employed to correct the positioning errors of sub-apertures. The feasibility and effectiveness of this method are demonstrated. The final stitched surface results exhibit a measurement error of 320 nm and a thickness deviation of 28 µm.

Coupling Global Parameters and Local Flow Optimization of a Pulsed Ejector for Proton Exchange Membrane Fuel Cells

Proton exchange membrane fuel cells (PEMFCs), as an important utilization of hydrogen energy, contribute to the sustainable development of global energy. Pulsed ejectors have a high potential for improving the hydrogen utilization of PEMFCs in the full operating range by circulating unconsumed hydrogen. In this study, a pulsed ejector applied to a 120 kW fuel cell was designed, and the flow characteristics were analysed using computational fluid dynamics (CFD). Based on the data from the CFD model, the global optimization of the ejector was carried out using the Gaussian process regression (GPR) surrogate model and the grey wolf optimization (GWO) algorithm. The local structure was then further optimized using an adjoint method coupling streamlining modification that takes into account the local flow characteristics. The CFD results showed that, under a fixed structure, increasing the pressure difference between the secondary flow and the ejector outlet would promote boundary layer separation, shorten the shockwave chain length, change the effective flow area of the secondary flow, and lower the entrainment ratio (ER). The analytical results from the GPR model indicated significant interactions among the structural parameters. The globally optimized ejector using GPR and GWO improved the hydrogen entrainment ratio from 1.42 to 3.12 at the design point. Furthermore, the results of streamlining local optimization show that the entrainment ratio increased by 1.67% at the design point and increased by up to 3.99% over the full operating range compared to the optimized ejector by global optimization.

Prediction of mechanical properties of LPBF built part based on process monitoring and Gaussian process regression

As a highly promising technology in additive manufacturing, the laser powder bed fusion has only limited application due to its low reproducibility. In this study, the image information of the 316L specimen after laser scanning and powder paving of each layer was acquired by a complementary metal–oxide–semiconductor industrial camera. The important features were selected, extracted and quantificated by analyzing the tensile test results. Finally, combined with the laser power, the quantified features were as input of a Gaussian process regression model based on optimization algorithm of grid search to predict the tensile strength of 316L specimen. The results show that the quantized image features have a significant improvement on the regression effect, and the coefficient of determination (R 2) is improved from 63% to 90.57% compared to using only the laser power as input.

Prediction and optimisation of gasoline quality in petroleum refining: The use of machine learning model as a surrogate in optimisation framework

Prediction and optimisation of gasoline quality in petroleum refining: The use of machine learning model as a surrogate in optimisation framework

Effect of Human Head Shape on the Risk of Traumatic Brain Injury: A Gaussian Process Regression-based Machine Learning Approach.

Computational head injury models are promising tools for understanding and predicting traumatic brain injuries. However, most available head injury models are "average" models that employ a single set of head geometry (e.g., 50th-percentile U.S. male) without considering variability in these parameters across the human population. A significant variability of head shapes exists in U.S. Army soldiers, evident from the Anthropometric Survey of U.S. Army Personnel (ANSUR II). The objective of this study is to elucidate the effects of head shape on the predicted risk of traumatic brain injury from computational head injury models. Magnetic resonance imaging scans of 25 human subjects are collected. These images are registered to the standard MNI152 brain atlas, and the resulting transformation matrix components (called head shape parameters) are used to quantify head shapes of the subjects. A generative machine learning model is used to generate 25 additional head shape parameter datasets to augment our database. Head injury models are developed for these head shapes, and a rapid injurious head rotation event is simulated to obtain several brain injury predictor variables (BIPVs): Peak cumulative maximum principal strain (CMPS), average CMPS, and the volume fraction of brain exceeding an injurious CMPS threshold. A Gaussian process regression model is trained between head shape parameters and BIPVs, which is then used to study the relative sensitivity of the various BIPVs on individual head shape parameters. We distinguish head shape parameters into 2 types: Scaling components ${T_{xx}}$, ${T_{yy}}$, and ${T_{zz}}$ that capture the breadth, length, and height of the head, respectively, and shearing components (${T_{xy}},{T_{xz}},{T_{yx}},{T_{yz}},{T_{zx}}$, and ${T_{zy}}$) that capture the relative skewness of the head shape. An overall positive correlation is evident between scaling components and BIPVs. Notably, a very high, positive correlation is seen between the BIPVs and the head volume. As an example, a 57% increase in peak CMPS was noted between the smallest and the largest investigated head volume parameters. The variation in shearing components ${T_{xy}},{T_{xz}},{T_{yx}},{T_{yz}},{T_{zx}}$, and ${T_{zy}}$ on average does not cause notable changes in the BIPVs. From the Gaussian process regression model, all 3 BIPVs showed an increasing trend with each of the 3 scaling components, but the BIPVs are found to be most sensitive to the height dimension of the head. From the Sobol sensitivity analysis, the ${T_{zz}}$ scaling parameter contributes nearly 60% to the total variance in peak and average CMPS; ${T_{yy}}$ contributes approximately 20%, whereas ${T_{xx}}$ contributes less than 5%. The remaining contribution is from the 6 shearing components. Unlike peak and average CMPS, the VF-CMPS BIPV is associated with relatively evenly distributed Sobol indices across the 3 scaling parameters. Furthermore, the contribution of shearing components on the total variance in this case is negligible. Head shape has a considerable influence on the injury predictions of computational head injury models. Available "average" head injury models based on a 50th-percentile U.S. male are likely associated with considerable uncertainty. In general, larger head sizes correspond to greater BIPV magnitudes, which point to potentially a greater injury risk under rapid neck rotation for people with larger heads.

Enabling efficient regional seismic fragility assessment of multi‐component bridge portfolios through Gaussian process regression and active learning

AbstractRegional seismic fragility assessment of bridge portfolios must address the embedded uncertainties and variations stemming from both the earthquake hazard and bridge attributes (e.g., geometry, material, design detail). To achieve bridge‐specific fragility assessment, multivariate probabilistic seismic demand models (PSDM) have recently been developed that use both the ground motion intensity measure and bridge parameters as inputs. However, explicitly utilizing bridge parameters as inputs requires numerous nonlinear response history analyses (NRHAs). In this situation, the associated computational cost increases exponentially for high‐fidelity bridge models with complex component connectivity and sophisticated material constitutive laws. Moreover, it remains unclear how many analyses are sufficient for the response data and the resulting demand model to cover the entire solution space without overfitting. To deal with these issues, this study integrates Gaussian process regression (GPR) and active learning (AL) into a multistep workflow to achieve efficient regional seismic fragility assessment of bridge portfolios. The GPR relaxes the probability distribution assumptions made in typical cloud analysis‐based PSDMs to enable heteroskedastic nonparametric seismic demand modeling. The AL leverages the varying standard deviation to select the least but most representative bridge‐model‐ground‐motion sample pairs to conduct NRHA with much‐improved efficiency. Both independent and correlated multi‐output GPRs are proposed to deal with bridge portfolios with seismic demand correlations among multiple components (column, bearing, shear key, abutment, unseating, and joint seal). Considering a single benchmark highway bridge class in California as the case study, the AL‐GPR framework and the associated component‐level fragility results are investigated in terms of their efficiency, accuracy, and robustness. The fragility results show that 70 AL‐selected samples would enable the GPR to derive bridge‐specific fragility models comparable to the ones using the multiple stripes analysis approach with 1950 ground motions considered for each individual bridge. The AL‐GPR model also successfully captures the physics of how bridge span length, deck area, column slenderness, and steel reinforcement ratio would change the damage state exceedance probabilities of different bridge components. The efficiency of AL stems from the fact that, with the multi‐output independent GPR, a stable and reliable fragility model can be achieved using 50 AL‐selected samples compared to at least 270 randomly chosen samples. The proposed methodology advances the state of the art in enabling more efficient and reliable regional seismic fragility assessment of multi‐component bridge portfolios.

Non-oscillatory fracture mode partition for interface crack under mixed-mode loading

The oscillatory crack-tip field in the classical interface fracture mechanics solution often leads to complications in identifying the mode mixity in bimaterial interface fracture problems. In this paper, a non-singular series solution for the continuum problem of a crack on the cohesive-spring interface under mixed-mode loading is presented and compared to the classical oscillatory solution. A Gaussian process regression model was also developed to enable quick evaluation of the phase angle for interface delamination characterization by using mixed-mode bending fracture test.

Machine-Learning-Based Kinetic-Free Hybrid Gray-Box Model to Predict Sulfuric-Acid-Catalyzed Esterification Reaction.

The sulfuric-acid-catalyzed esterification reaction of 2-butanol and propionic anhydride is a vital industrial process. In this paper, several experiments are conducted via reaction calorimetry to validate that both sulfuric and propionic acids have discernible catalytic effects on the reaction. This finding complicates the accurate description of the reaction kinetics through traditional methods. So this paper turns to a kinetic-free black-box model, Gaussian process regression (GPR) model via 24 experiments, as a more adaptable approach. Besides, the best GPR model is combined with traditional heat balance model to generate a hybrid gray-box model, which can give complete knowledge of reaction process. The hybrid gray-box model finally achieves maximum root-mean-square error (RMSE) of 0.0069 for conversion, 0.6535 for temperature, and 10.6087 for heat flow rate, underscoring its pretty good predictive ability.

A physical simulation-machine learning model for optimal process schemes in laser-based directed energy deposition process

The major challenge faced is the definition of optimal process variables for rich quality of fabricated parts in laser-based directed energy deposition (DED-LB) processes. However, predicting the optimal process scheme using machine learning models is still challenging owing to the need for a large amount of training experimental data with high costs in DED-LB. In view of this, a physical simulation (PS)-machine learning (ML) model using a relatively small accurate data set of 31 simulation data for optimal process schemes is proposed in DED-LB process in this study. Firstly, a powder-scale high precision phenomenon model incorporating the mass transfer, phase transformations and heat transfer and using Lagrangian particle model to add mass is developed to depict the DED-LB process. Then, a Gaussian process regression (GPR) agent model is established to rapidly and accurately predict the geometry and dilution rate of deposition tracks under different manufacturing parameters based on the high precision simulation results. Finally, a PS-ML model for process parameter optimization is developed using the particle swarm algorithm (PSO), and the optimized parameters are experimentally validated. The results show that the developed powder-scale model and GPR model results are consistent with the Inconel 718 alloy experimental results. The proposed PS-ML model can increase the accuracy of the ML models even with a small simulation data set. None of PS-ML model optimization results has a relative error of more than 3% to the experimental results, and the dilution rate is reduced by up to 61.66% compared to the experimental design parameters without optimization. The proposed physical simulation-machine learning model in this study enables inexpensive and accurate optimization of DED-LB process parameters.

A novel decoupled approach combining invertible cross-entropy method with Gaussian process modeling for reliability-based design and topology optimization

Design optimization considering the presence of uncertainties in parameters poses an extremely challenging problem. The source of difficulties comes with reliability-based formulations, where addressing the probabilistic problem exhausts the large computing efforts for failure estimations of the structure violating limit-state functions (LSFs). This paper proposes a novel decoupled approach for effectively solving reliability-based design optimization (RBDO) problems, namely an invertible cross-entropy (iCE) method advantageously combined with a Gaussian process regression (GPR) model, termed as iCE-GPR. The GPR model is applied to approximate the spectrum of LSFs under random parameters. Furthermore, to enhance the accurate prediction of the system failure probability, an active learning process is applied to systematically refine the GPR model by adding new learning points in the region with the largest uncertainty and high-reliability sensitivity through the maximization of an expected feasibility function (EFF). Based on the updated GPR model, the failure probability is estimated by a cost-effective cross-entropy (CE) method without any calls to the actual performance function. To perform the decoupling optimization process with the reliability analysis, the novel iCE, based on the CE method, is developed to update the most probable point (MPP) assigned for the next deterministic optimization process in determining the new optimal design. The method iteratively performs the deterministic optimization process based on the MPP underpinning LSFs sequentially updated by the active learning process. The proposed iCE-GPR method fast-converges the optimal design and significantly alleviates computational burdens associated with reliability analyses. The proposed method is also applied to solve a reliability-based topology optimization (RBTO) problem. Four numerical examples for both the RBDO and RBTO problems are provided to illustrate efficiency and robustness of the proposed iCE-GPR method.

Integrating MODFLOW and machine learning for detecting optimum groundwater abstraction considering sustainable drawdown and climate changes

Detecting optimum groundwater abstraction is a challenging objective. This study proposes a new approach for identifying the optimum pumping rates by integrating MODFLOW with machine learning (ML). The adopted strategy has been applied to the Oligocene and Eocene aquifers in the western desert of Minia, Egypt. MODFLOW-2005 model has been utilized to simulate various management scenarios, including current and expected climate scenarios of two Representative Concentration Pathways (RCPs): RCP 4.5 and RCP 8.5. The steady-state and transient MODFLOW models were calibrated using the groundwater levels of 37 observation wells with a coefficient of correlation (R2) of 0.986 and 0.988, respectively. The developed model has been run for 80 years, from 2020 to 2100. The results of MODFLOW demonstrated a drawdown in the groundwater level in the two aquifers, about 40, 41, and 42 m, owing to current climate conditions and scenarios of RCP 4.5 and RCP 8.5, respectively, in 2100. Generated data based on the calibrated model has been used to develop the ML models. ML models have been applied to predict the pumping rates using the well coordinates, aquifer type, evapotranspiration, the operating period in days, and drawdown as inputs. For this purpose, four ML models, Gaussian Process Regression (GPR), Support Vector Regression (SVR), Tree, and Artificial Neural Network (ANN)) were developed and evaluated. All ML models demonstrated an effective performance during the training and validation processes, particularly the GPR model, which has the best ability to predict the pumping rates with a correlation coefficient, RMSE, and Mean Absolute Error (MAE) equal to 1.0, 6.53, and 8.04, respectively. Therefore, the GPR model has been utilized to identify the optimum pumping rates. The findings of the GPR model revealed that the optimum abstraction of the Oligocene aquifer is 602266, 625329, and 635061 m3/day, while the Eocene aquifer’s optimal groundwater abstraction is 10229801, 1027906, and 102936 m3/day, considering the current climate, RCP 4.5, and RCP8.5, respectively. Anticipated findings are supposed to be instrumental for decision-makers, offering valuable insights into sustainable groundwater management. The novel integration methodology promotes the efficient use of groundwater, contributing to achieving sustainable development goals (SDGs).

A Gaussian process approach for rapid evaluation of skin tension

Skin tension plays a pivotal role in clinical settings, it affects scarring, wound healing and skin necrosis. Despite its importance, there is no widely accepted method for assessing in vivo skin tension or its natural pre-stretch. This study aims to utilise modern machine learning (ML) methods to develop a model that uses non-invasive measurements of surface wave speed to predict clinically useful skin properties such as stress and natural pre-stretch. A large dataset consisting of simulated wave propagation experiments was created using a simplified two-dimensional finite element (FE) model. Using this dataset, a sensitivity analysis was performed, highlighting the effect of the material parameters and material model on the Rayleigh and supersonic shear wave speeds. Then, a Gaussian process regression model was trained to solve the ill-posed inverse problem of predicting stress and pre-stretch of skin using measurements of surface wave speed. This model had good predictive performance (R2 = 0.9570) and it was possible to interpolate simplified parametric equations to calculate the stress and pre-stretch. To demonstrate that wave speed measurements could be obtained cheaply and easily, a simple experiment was devised to obtain wave speed measurements from synthetic skin at different values of pre-stretch. These experimental wave speeds agree well with the FE simulations, and a model trained solely on the FE data provided accurate predictions of synthetic skin stiffness. Both the simulated and experimental results provide further evidence that elastic wave measurements coupled with ML models are a viable non-invasive method to determine in vivo skin tension. Statement of significanceTo prevent unfavourable patient outcomes from reconstructive surgery, it is necessary to determine relevant subject-specific skin properties. For example, during a skin graft, it is necessary to estimate the pre-stretch of the skin to account for shrinkage upon excision. Existing methods are invasive or rely on the experience of the clinician. Our work aims to present an innovative framework to non-invasively determine in vivo material properties using the speed of a surface wave travelling through the skin. Our findings have implications for the planning of surgical procedures and provides further motivation for the use of elastic wave measurements to determine in vivo material properties.

Surrogate modeling with non-stationary-noise based Gaussian process regression and K-Fold ANN for systems featuring uneven sensitivity distribution

In the realm of aircraft design, there is a prevalent need for surrogate modeling techniques capable of efficiently and accurately modeling objects with high evaluation costs and uneven sensitivity distributions, such as those encountered in the nonlinear buckling of thin-walled structures and aerodynamics at transonic speeds. However, the non-stationary Gaussian Process Regression (GPR) model requires extensive hyperparameters or judicious assumptions, limiting its applications in such tasks, and the commonly used K-Fold methodology suffers from inherent instability. Addressing these issues, this paper proposes an active learning method for surrogate modeling in systems featuring uneven sensitivity distribution, integrating non-stationary-noise GPR and K-Fold Artificial Neural Network (ANN) methodology, namely NSN-GPR-KF. Instead of relying on a non-stationary kernel function, the algorithm regards noise as a non-stationary factor within a stationary GPR framework, with noise levels estimated via the K-Fold ANN methodology, allowing GPR to efficiently adapt to non-stationary systems while avoiding the aforementioned requirements for non-stationary GPR. Case validations, performed on two test functions with pronounced uneven spatial sensitivity, demonstrate that NSN-GPR-KF outperforms commonly used stationary GPR and pure K-Fold methodology. It synthesizes the exploration capabilities of GPR with the exploitation proficiency of the K-Fold methodology, achieving comparable global accuracy with reduced sample sizes—up to 26% and 22% less than the other two algorithms, respectively. The algorithm was successfully applied to predict the buckling load of thin-walled pipe beams, demonstrating accelerated convergence and approximately 25% greater accuracy compared to pure K-Fold model. These results suggest that NSN-GPR-KF can serve as an efficient and precise modeling tool for engineering tasks with high evaluation costs and complex sensitivity distributions.

Adaptive switching and routing protocol design and optimization in internet of things based on probabilistic models

Through smart cities, Intelligent Transportation Systems (ITS), the agricultural sector, and wearable devices, the Internet of Things (IoT) has revolutionized several human interests. Through the development of new cluster tasks, the Decision-Making System (DMS) of Cluster Heads (CHs), and improving the accuracy of traffic prediction and reliability of transportation, the present study intends to improve the energy depletion of IoT devices. The paper explores the subject of data flow optimization using Fuzzy Assisted Cuckoo Search Optimization (FACSO), traffic flow using Gaussian Process Regression (GPR), and CH prediction using the Stochastic Optimization Algorithm (SOA). Optimizing network lifetime while minimizing Energy Consumption (EC) is feasible through the practical application of the SOA, GPR, and FACSO models. Increasing End-to-End Delay (EED), Network Throughput (NT), and energy efficiency can be rendered feasible through a real-time DMS regarding routing employing a novel approach referred to as FACSO. This approach has enhanced the efficacy and reliability of Wireless Sensor Networks (WSN). With up to 500 nodes and an EC of 0.3451 J, the experiment's findings demonstrate that a proposed SOA-FACSO model achieves superior EED.

USING MACHINE LEARNING TO PREDICT THE MECHANICAL PROPERTIES OF PET FIBER-REINFORCED CONCRETE

Fiber-reinforced concrete enhances structural integrity by improving post-cracking behavior and ductility, yet traditional fibers can be costly and have a significant environmental impact. Recently, polyethylene terephthalate (PET) fiber, derived from recycled plastic bottles, has been used as a reinforcing material in concrete to enhance its mechanical properties. Utilizing advanced machine learning algorithms, this research analyzes a comprehensive dataset collected from existing literature on the mechanical properties of PET fiber-reinforced concrete (PFRC), including concrete density, water-to-binder ratio, fine and coarse aggregates, and fiber parameters such as volume fraction and aspect ratio of PET fibers. Through the application of Polynomial Regression, Exponential Gaussian Process Regression (GPR), and Cubic Support Vector Machine (SVM) models, the study effectively predicts PFRC’s mechanical behavior, showcasing high R² values indicating the models’ predictive strength: 0.93 for compressive, 0.92 for flexural, and 0.92 for tensile properties. This investigation highlights the exceptional accuracy of Exponential GPR and Cubic SVM models in capturing complex, non-linear relationships within the material, demonstrating machine learning’s capability to advance the sustainable development of construction materials. This study provides significant insights into optimizing the use of recycled fibers for concrete reinforcement.

Adaptive Design of Experiments to Fit Surrogate Gaussian Process Regression Models Allows Fast Sensitivity Analysis of the Input Waveform for Patient-Specific 3D CFD Models of Liver Radioembolization

Background and objectivePatient-specific 3D computational fluid dynamics (CFD) models are increasingly being used to understand and predict transarterial radioembolization procedures used for hepatocellular carcinoma treatment. While sensitivity analyses of these CFD models can help to determine the most impactful input parameters, such analyses are computationally costly. Therefore, we aim to use surrogate modelling to allow relatively cheap sensitivity analysis. As an example, we compute Sobol's sensitivity indices for three input waveform shape parameters. MethodsWe extracted three characteristic shape parameters from our input mass flow rate waveform (peak systolic mass flow rate, heart rate, systolic duration) and defined our 3D input parameter space by varying these parameters within 75%-125% of their nominal values. To fit our surrogate model with a minimal number of costly CFD simulations, we developed an adaptive design of experiments (ADOE) algorithm. The ADOE uses 100 Latin hypercube sampled points in 3D input space to define the initial design of experiments (DOE). Subsequently, we re-sample input space with 10,000 Latin Hypercube sampled points and cheaply estimate the outputs using the surrogate model. In each of 27 equivolume bins which divide our input space, we determine the most uncertain prediction of the 10,000 points, compute the true outputs using CFD, and add these points to the DOE. For each ADOE iteration, we calculate Sobol's sensitivity indices, and we continue to add batches of 27 samples to the DOE until the Sobol indices have stabilized. ResultsWe tested our ADOE algorithm on the Ishigami function and showed that we can reliably obtain Sobol's indices with an absolute error <0.1. Applying ADOE to our waveform sensitivity problem, we found that the first-order sensitivity indices were 0.0550, 0.0191 and 0.407 for the peak systolic mass flow rate, heart rate, and the systolic duration, respectively. ConclusionsAlthough the current study was an illustrative case, the ADOE allows reliable sensitivity analysis with a limited number of complex model evaluations, and performs well even when the optimal DOE size is a priori unknown. This enables us to identify the highest-impact input parameters of our model, and other novel, costly models in the future.