Gradient Model Research Articles

Drainage dynamics: From capillary to viscous fingering under different wettability

The displacement of two-phase fluids within porous media is a critical process in various engineering and industrial fields. However, the inherent instability of the fluid interface often leads to an irregular morphology of the displacement front, which can impact the efficiency of displacement and limit its application in engineering activities. In this study, the color gradient model based on the lattice Boltzmann method was employed to simulate displacement, with the aim of investigating the dynamic evolution of the displacement front under the combined effects of wettability and injection velocity, specifically during the capillary to viscous fingering transition. The results demonstrate that the impact of wettability on the displacement front morphology increases as the injection velocity decreases. The displacement process with the maximum fingering distribution range and compact morphology is found at the midpoint of the capillary-viscous fingering crossover zone. Furthermore, an in-depth analysis of the phase field evolution and pressure difference changes between the inlet and outlet unveiled the pore-filling mechanisms associated with the unique variation characteristics of the displacement front and the fluid topological structure. These comprehensive pore-scale findings offer theoretical insight supporting the engineering application of the two-phase flow in porous media from multiple perspectives.

Concentration gradient effect in liquid thermoelectric device composed of organic-solvent-added aqueous solution containing K4[Fe(CN)6]/K3[Fe(CN)6

A liquid thermoelectric conversion device (LTE), which has a simple device structure composed of low-cost materials, is attracting attention. Here, we investigated the additive effect of methanol (MeOH), acetone, and tetrahydrofuran (THF) on an LTE composed of a K4[Fe(CN)6]/K3[Fe(CN)6] aqueous solution. We ascribed to the enhancement of the thermal voltage V and electrochemical Seebeck coefficient α to precipitation of K4[Fe(CN)6] and resultant concentration gradient of [Fe(CN)6]3− between the hot and cold electrodes. The enhancement of V and α is quantitatively reproduced by the concentration gradient model incorporating experimentally-obtained solubility s(T) of K4[Fe(CN)6]. We further demonstrated that precipitation enhances the maximum power by as much as 152% in a MeOH-added LTE composed of 0.35 M K4[Fe(CN)6]/K3[Fe(CN)6] aqueous solution at ΔT = 10 K.

A Target-Aware Well Path Control Method Based on Transfer Reinforcement Learning

Summary In the actual construction process, well path control is a challenging task mainly due to the inevitable well deflection caused by geological factors, drilling tools as well as borehole enlargement. Most conventional well path control methods focus on elaborate mechanism model construction. The methods are typically constructed on the basis of certain constraints or assumptions, which reflect their limited ability to accurately capture the actual drilling process, low level of intelligence, poor anti-interference performance, and weak adaptive capacity. To address these challenges, this paper proposes a target-aware well path control method that integrates reinforcement learning and transfer learning. The proposed method employs a deep deterministic policy gradient model based on the prioritized experience replay mechanism and leverages transfer learning to accelerate model learning. This enables the construction of a target-aware well path adaptive control system with strong anti-interference capability. The proposed target-aware control method of well path based on reinforcement learning and transfer learning can accurately track the preset trajectory in diverse geological environments, reach the target area with high precision, and make reasonable trajectory optimization decisions with measurement while drilling (MWD) even when the target trajectory does not match the actual distribution of the reservoir. This approach exhibits excellent anti-interference and adaptive abilities.

Wind turbine wake superposition under pressure gradient

We investigate the effect of pressure gradient on the cumulative wake of multiple turbines in wind tunnel experiments spanning across a range of adverse pressure gradient (APG), zero pressure gradient (ZPG), and favorable pressure gradient (FPG). Compared to the upstream-most turbine, the in-wake turbines exhibit lower (higher) wake velocity in APG (FPG) than in the ZPG. The maximum velocity deficit shows a lesser difference for the in-wake turbine between different cases compared to the upstream-most one. This is linked to the effect of the wake of the upstream turbine. Conversely, the wake width varies more for the in-wake turbines. A new analytical approach to model the cumulative wake velocity deficit is proposed. This approach extends the application of the analytical pressure gradient model to multiple turbine wakes. Specifically, the new approach explicitly accounts for the effect of the pressure gradient induced by the wake of the upstream turbine on the wake of the downstream one. The new method is compared to the linear summation approach and experimental data. It agrees well with the experiments and outperforms the linear summation approach.

Shallow and deep geothermal water sources identification in Unai geothermal field, Gujarat, India with applications of Magnetotelluric (MT)

The aim of this study is to conduct and examine the results of a Magnetotelluric (MT) survey in Unai, Gujarat, India for geothermal prospect identification. At Unai, the data was acquired, processed, and interpreted in the form of 2D and 3D MT surveys. Data were acquired at 56 stations in uniformly spaced profiles along the ENE-WSE direction (geological strike direction) with a station spacing of 2 km. Qualitative and quantitative data understanding has been done for geothermal prospect identification. Deep and shallow 1D and 2D models of electrical resistivity were made, which suggests the presence of a geothermal aquifer. This is corroborated by polar and skew tipper diagrams. The one-dimensional Occam models and two-dimensional conjugate gradient models confirm the presence of a low resistivity anomaly near the surface ranging from the depth of 100–700 m. This suggests the presence of a shallow geothermal zone. Apart from the shallow geothermal prospect, this study also delineates a deep conductive body which may be attributed to a matured geothermal reservoir. For a better understanding of the prospect geoelectrical cross sections are analysed from a depth of 10 m to 15 km. The MT cross-sections show the presence of geological features such as a series of faults. These faults form horst and graben in nature and may act as a good trap for the shallow reservoirs.

Mechanism investigation of highly selective inhibitors toward phosphodiesterase 5 and 6 via the in vitro calculation and simulation.

Introduction: In clinical practice, phosphodiesterase 5 (PDE5) inhibitors are commonly used to treat erectile dysfunction and pulmonary arterial hypertension. However, due to the high structural similarity between PDE5 and Phosphodiesterase 6 (PDE6), there is a risk that existing drugs will cause off-target effects on PDE6 resulting in visual disorders such as low visual acuity and color blindness. Previous research on the selectivity of PDE5 inhibitors focused on marketed drugs such as sildenafil and tadalafil. Methods: In this study, a highly selective PDE5 inhibitor, ligand3, was used as the subject, and molecular docking, molecular dynamics simulations, MM-GBSA, alanine scanning, and independent gradient model analysis were employed to investigate the biological mechanism underlying the selectivity of PDE5 inhibitors. Results and Discussion: The present work revealed that the binding mode of ligand3 to the PDE5A and PDE6C targets was distinctly different. Ligand3 exhibited stronger coulombic forces when binding to PDE5A, while showing stronger van der waals forces when binding to PDE6C. Ligand3 binds more deeply at the active site of PDE5A than at PDE6C, allowing its side chains to effectively bind to the critical TYR612, whereas in the case of the shallow binding to PDE6C, ligand3 lacks a similar effect. Mechanism investigations of highly selective inhibitors through computational simulation might provide an insight into potent treatment of drugs.

An experimental and computational exploration on noncovalent interactions, spectroscopic and nonlinear optical properties of novel creatininium cyanoacetate single crystal

A novel creatininium cyanoacetate (CCA) single crystal is synthesized by the slow evaporation technique and characterized by various experimental techniques such as FT-IR, FT-Raman and UV–visible spectral methods. Similar techniques are also investigated by the computational methods using Gaussian 09 software. The structural parameters, natural bond orbital (NBO), frontier molecular orbital (FMO), molecular electrostatic potential (MEP) and nonlinear optical (NLO) investigations are performed utilizing the DFT method with B3LYP/6-311G(d,p) basis set. The polarizable continuum model (PCM) has been used to identify the effect of various solvents (DMSO, ethanol, methanol, water) on the energy gap, excitation and NLO properties of CCA molecule. Fukui analysis is performed to identify the sites prone to nucleophilic and electrophilic attacks. Different topological methods including electron localization function (ELF), localized orbital locator (LOL), independent gradient model (IGM) and interaction region indicator (IRI) has been performed to quantify the noncovalent interaction and to analyse the weak interaction region. The UV–visible analysis indicates that the CCA crystal was optically transparent and has lower cut-off wavelength. The hole-electron analysis is performed to identify the characteristics of six excitations. The third-order NLO activity of the CCA compound is performed using the Z-scan technique and is compared with some related NLO materials. The substantial NLO response suggests that the CCA molecule can be significant for its utilization in advanced NLO applications.

Stokes shift prediction of fluorescent organic dyes using machine learning based hybrid cascade models

Fluorescent organic dyes are widely used in various fields, including science and technology, research and development, medicine, and drug delivery. Multitudinous attempts have been made by experimentalists to develop such fluorescent organic dyes with the desired Stokes shift property at negligible cost and time. For quickly and accurately predicting the Stokes shift property of fluorescent organic dye, we proposed eight hybrid models based on the combination of nine single machine-learning models. To fulfill the objective, we considered a dataset of 3066 fluorescent organic materials and evaluated the performance of each model using three evaluation parameters: mean absolute error (MAE), root mean squared error (RMSE), and the coefficient of determination (R2). The hybrid cascade model of Extreme Gradient Boosting Regression and Light Gradient Boosting Machine Regression (XGBR + LGBMR) performed best for Stokes shift prediction, with MAE of 13.83 nm, RMSE of 19.95 nm, and R2 of 86.18 %. The prediction performance of all the undertaken models was validated by the experimental data of four xanthene dyes (Rh-19, Rh–B, Rh-6G, and Rh-110). In this regard, the XGBR + DTR (Extreme Gradient Boosting Regression + Decision Tree Regression) model was the best performer, with errors ranging from 5 to 13 nm for four dyes. The resultant errors are much smaller than the recently reported synthesized material with an error of 30 nm. The proposed models allow for rapid and cost-effective screening of a wide range of fluorescent organic dyes, which assists the researchers in gaining prior knowledge of materials and accelerates the discovery of new materials.

A nonlocal strain gradient model for buckling analysis of laminated composite nanoplates using CLPT and TSDT

A nonlocal strain gradient model for buckling analysis of laminated composite nanoplates using CLPT and TSDT

Experimental study of viscous oil-water core-annular flow in horizontal pipe

Transporting viscous oils through pipelines is usually challenging due to their very high viscosity. Significant friction reduction in the transportation of viscous oils through pipelines using an ecologically friendly way is the most appealing feature of the core annular Flow (CAF) technique. An experimental study has been conducted to investigate the flow patterns and pressure gradient of viscous oil-water two-phase flow in a horizontal acrylic pipe with a 50 mm inner diameter. The test section was equipped with a fluid lubrication injector element to artificially provide CAF. The pressure gradient measurements were focused on the CAF pattern. The prediction models of the CAF pressure gradient were reviewed and evaluated. An empirical prediction model of CAF pressure gradient was developed, relying on both the present and previously published experimental results. The flow patterns were classified into five patterns. It was observed that the CAF occupied a large region in the constructed flow pattern map and its frictional pressure gradient closely resembles that of water flowing alone. It was found that the developed empirical model provided quite good predictions of the experimental pressure gradients and significantly outperformed the original model. The reviewed prediction models in this area for CAF either overpredicted or underpredicted the experimental pressure gradients. Some of these models considered the extensive effect of viscous oil adhered to the inner wall of the pipe, while others neglected the effect entirely.

Comprehensive solubility study and inter-molecular interactions on fenbendazole dissolved in some aqueous aprotic and protic co-solvent solutions

The isothermal shake-flask saturation method was used to experimentally determine the mole-fraction solubility of fenbendazole in three aqueous co-solvent blends of 1,4-dioxane/acetone/NMP encompassing the range of 283.15 to 328.15 K. The solubility of fenbendazole in various blended solvents ranks NMP + water (4.852 × 10−2) > 1,4-dioxane + water (14.81 × 10−4) > acetone + water (2.285 × 10−4) with composition of 1,4-dioxane (acetone or NMP) of 1 at 298.15 K. The solubility data rises monotonically with 1,4-dioxane/acetone/NMP concentration at the same temperature. X-ray power diffraction images show that no solvate production or crystal transition appeared throughout the studies. The solubility to solvent composition and temperature was satisfactorily associated by the modified van't Hoff-Jouyban-Acree, quantitative structure–property relationship, and Jouyban-Acree models with relative average deviations of no higher than 7.05 % and root-mean-square deviation of no higher than 1.225 × 10−3. In addition, the 1,4-dioxane/acetone/NMP + water blends studied in this paper and the methanol/ethanol/EG/DMF + water blends previously reported both used the extended Hildebrand solubility approach to quantitatively describe the solubility behavior at 298.15 K. The average relative deviations were kept under 4.06 % on both cases. As stated by the examination of the linear solvation energy relationship, the solubility parameter and dipolarity-polarizability of solutions have a significant influence on the solubility variation. The effective method of inverse Kirkwood-Buff integrals was used to investigate the preferred solvation of fenbendazole at 298.15 K. In blends with intermediate and rich ethanol/DMF/1,4-dioxane/acetone/NMP composition areas, the preferred solvation parameters of ethanol/DMF/1,4-dioxane/acetone/NMP displayed positive values, indicating the preferential solvation of fenbendazole by the co-solvents. Thermodynamic examination of the entropy-enthalpy compensation and dissolution parameters for fenbendazole dissolving in blends led to the identification of an enthalpy and/or entropy-driven mechanism as well as an endothermic process. Moreover, the average local ionization energy and electrostatic potential of molecular surface were utilized as effective instruments to demonstrate the microscopic electrostatic properties of acidity and basicity, respectively. The fenbendazole molecule's groups of –N in the five-membered ring and –NH in the straight chain serve as the main targets for electrophilic and nucleophilic assault. The weak interactions between fenbendazole and several solvents were shown using an independent gradient model based on Hirshfeld partition analysis.

Application of nonlocal strain gradient theory for the analysis of bandgap formation in metamaterial nanobeams

The aim of this paper is to investigate bandgap formation in an Euler-Bernoulli nanobeam using the nonlocal strain gradient theory. Such a model enables the capture of both softening and stiffening size-dependent behavior of the nanobeams using two length-scale parameters, namely, a nonlocal parameter and a strain gradient parameter. Hamilton's principle is used to develop the normalized equation of motion yielding a sixth order differential equation. The nanobeam is assumed to be infinitely long and the dispersion of an elastic wave in a single representative periodic unit cell of the structure is used to estimate the bandgap edge frequencies. To this end, two unit cell approaches are employed to obtain the dispersion relations, namely, the homogenization and transfer matrix approaches. Both methods yield similar dispersion curves and showed good agreement with classical cases from the literature. A parametric study was carried out to investigate the effect of several parameters on the bandgap formation in the metamaterial nanobeams. These parameters include, the nonlocal parameter, the strain gradient parameter, the length of the unit cell and the resonator mass to unit cell mass ratio. The study conducted using a purely strain gradient model showed a comparable response to that of the classical macro problem with the usual hardening when increasing the value of the strain gradient length scale. Nonlocal and nonlocal strain gradient models introduced coupling between parameters of the study with the occasional introduction of non-monotonic dispersion curves. Such non-monotonic behavior is believed to be characteristic of large values of the nonlocal length-scale.

A 3D pantographic metamaterial behaving as a mechanical shield: Experimental and numerical evidence

Multilayer pantographic metamaterials, in short, pantographic blocks, have shown peculiar mechanical behavior, especially when their constitutive hinges are revolving (i.e., perfect) joints. The pantographic block, which is the subject of the present paper, has been printed using a Powder Bed Fusion technology and its hinges may be modeled as perfect ones. In the reported in situ 3-point flexural test, the predictions obtained by second gradient models for its mechanical response are shown to be experimentally consistent thanks to measurements via Digital Volume Correlation. The deformation applied by the upper central support is almost entirely shielded by the pantographic block, namely, the specimen barely crosses through the reference bottom plane defined by the lower lateral supports, even when subjected to very large deformations. The mathematical model employed herein captures this observation in terms of a nonlinear ‘arching’ effect activated in the beams of the pantographic structure, provided elastic locking is introduced to prevent pantographic zero-energy modes.

DFT Investigation on the Complexation of β-Cyclodextrin and Hydroxypropyl-β-Cyclodextrin as Recognition Hosts with Trichloroethylene

In this investigation, the potential use of native β-cyclodextrin (β-CD) and hydroxypropyl-β-cyclodextrin (HP-β-CD) as encapsulating agents for trichloroethylene (TCE) was assessed. Various quantum chemical parameters, including HOMO, LUMO, and HOMO–LUMO gap, were calculated. The docking process was examined by considering different initial configurations. The complexation energies were calculated at the molecular level using DFT/BLYP-D4 and PBEh-3c calculations to gain insight into TCE encapsulation within the β-CD and HP-β-CD cavities. We used the independent gradient model (IGM) and extended charge decomposition analysis (ECDA) approaches to examine non-covalent interactions and charge transfer within TCE@β-CD and TCE@HP-β-CD complexes. The calculated thermodynamic data and complexation energies exhibited negative values for both considered complexes, indicating a favorable complexation process. Weak Van der Waals intermolecular interactions were the main driving forces in stabilizing the formed complex. Additionally, Monte Carlo simulations were conducted for a better understanding of the inclusion process. Our results provide evidence for the use of β-CD and HP-β-CD as suitable macrocyclic hosts for complexing trichloroethylene.

Влияние режимов копчения на формирование цвета колбасных плавленых сыров

The research established the effect of temperature and smoking time on the color of dairy products with a milk fat substitute, i.e., cheese sausages. It resulted in a color matrix that determines the relationship between the smoking mode and the dynamics of the color spectrum. The article introduces a gradient model of color changes that makes it possible to program the surface color of the finished product in accordance with consumer preferences

Study on the interaction of tannins and salivary proteins affecting wine aroma volatility: Static HS-SPME and molecular dynamics simulation approaches

The interaction between tannins and salivary proteins might affect intraoral aroma release during wine consumption. In this study, the influence and underlying mechanism of interactions between EGCG and IB5 (salivary proline-rich protein) on wine aroma compounds was analysed by static HS-SPME in vitro and molecular dynamics (MD) simulation. The interaction between IB5 and EGCG could significantly reduce the volatility of most aroma compounds in the model wine by 20 %-70 % (p < 0.05). MD simulations indicated that the energy received by aroma compounds in the mixed system was more pronounced. In addition, the decline rate of rational correlation functions (RCF) of aroma compounds in the mixed system was obviously slower. The analysis of the independent gradient model (IGM) indicated that aroma compounds combined with aggregates of IB5 and EGCG through hydrogen bonds and van der Waals forces. The effect of the interaction between EGCG and IB5 on aroma compounds was confirmed by the volatility and molecular computational simulation. Overall, the results enhance the understanding of the mechanisms affecting retronasal aroma release during wine consumption.

Studies of oil product pollution in the aeration zone by mathematical modelling

Formulation of the problem. The article presents the results of a study to assess the pollution of the aeration zone due to the leakage of oil products as a result of the destruction of the «Amik» oil depot during the military operations in the town of Borodyanka. The destruction of the tanks resulted in an immediate leakage of light oil products onto the soil of the surrounding area. The result was a technogenic environmental situation that required immediate determination of the mechanisms of accidental penetration of pollutants into the soil and groundwater, determination of the area of contamination and scientifically sound methods of soil remediation of oil products. The article highlights the issues of organising mathematical modeling of various data for solving applied problems using Google Earth data. Materials and methods. In order to assess the contamination area and determine the mechanism of accidental penetration of oil products into the soil and groundwater, the existing models of pollutant penetration in case of accidental oil spill were analysed and mathematical spatial models of geomorphological, hydrogeological, engineering-geological and experimental filtration data were created using interpolation, mathematical approximation and gradient analysis. The latter made it possible to identify the directions of surface and groundwater flow both in the local area of the oil depot and in adjacent areas. Using the investigation and filtration parameters obtained during the field work, hydraulic gradients and soil filtration coefficients were calculated, and a spatial mathematical model of the local subterranean gradient of groundwater flow was created. Results. It was shown that the direction of the underground flow is controlled by a zone of local depression, which is expressed by a decrease in the absolute height of the natural lithological layer with a height difference of about 0.4 m or the maximum values of the height of the contamination lens. The direction of extension of this zone has been determined to be from north to south. This zone can be considered the zone of maximum accumulation of liquid with oil products. Based on the survey results, the thickness of the contamination lens was calculated to be 0.3 m on average. Using this indicator, the area of contamination is 4169 m2, the volume of contamination is 1250.7 m3, which corresponds to the claimed damage. However, taking into account the anisotropy of the filtration properties of the lithological types in different directions within the aeration zone, the presence of dissolved water due to dispersion and the presence of local accumulation depressions in the lithological strata, options with other indicators (0.5 m / 0.1 m) are also acceptable. The calculation of the contamination areas was carried out taking into account the power variations. The approximate total areas of oil product distribution in the aeration zone were plotted with reference to Google Earth imagery. Conclusions. The research enabled practical recommendations to be made for cleaning up the area of oil products and preventing the spread of the pollutant by pumping oil products out of the modelled and actually confirmed pollution lens. In accordance with the recommendations, 140 wells were drilled to a depth of 5.2 m, with a diameter of 0.11‑0.168 m and a total length of 710 m, which made it possible to prevent the spread of the contaminant by pumping out the oil products and reusing them after clean-up.

Fabrication of zwitterionic magnetic microporous organic network for efficient extraction of fluoroquinolone antibiotics from meat samples.

A zwitterionic magnetic microporous organic network (MMON-SO3H-NH2) with numerous amino and sulfonic acid ion-pare binding sites was designed and synthesized for efficient magnetic solid-phase extraction (MSPE) of fluoroquinolones (FQs) from meat samples. The core-shell MMON-SO3H-NH2 offered large specific surface area, rapid magnetic responsiveness, good stability, and multiple binding sites for FQs. The density functional theory and independent gradient model evaluations confirmed hydrogen bonding, π-π and ion-pair interactions between MMON-SO3H-NH2 and FQs. Under the optimal conditions, the established MMON-SO3H-NH2-MSPE-HPLC-UV method gave wide linear range (0.15-1000μg L-1), low limits of detection (0.05-4.5μg L-1) and limits of quantitation (0.15-13μg L-1), and high enrichment factors (82.1-99.6) using 3mg of adsorbent. This work demonstrates that the preparation of zwitterionic MONs is an efficient way to promote the extraction performance of MONs for zwitterionic targets and provides an effective sample pretreatment method for enriching and monitoring FQs in complex food matrices.

Cross-linking gradient model in disperse systems

The article presents the research results of the disperse systems cross-linking process containing solid and liquid phase bases on the continuum mechanics apparatus use within the incompressible fluid model framework. The system formation schemes of interfacial layers in fine clearances between solid particles with wall-adjacent zones formation, in which conditions of liquid and solid phase interphase contact are provided, and the cores with constant interphase pressure are considered. The authors have obtained expressions for the values of wedging pressure depending on the thickness of the interfacial layers overlapping area, which determine the interfacial contacts strengthening degree. The dependences given in the paper are a theoretical basis for a new approach to modelling surface phenomena from the position of continuum mechanics equilibrium. The facilitating mechanism of the wedging pressure long-range action outside the interfacial layers system area is given. The authors present the modelling results of a single liquid-solid interphase layer bounded, on the one side, by a solid surface and, on the other side, by a volume liquid phase and the results of modelling the effects of structuring disperse systems based on the liquid interphase layer study formed in the clearance between solid surfaces. Under certain conditions in such systems the effect of structuring, leading to a sharp increase in the strength properties of disperse-filled composite materials, which is the basis of modern nanotechnology concepts, is manifested. The dualism of the liquid phase, consisting in the ability to exhibit the properties of both liquid and solid body, is shown. It is proved that the main disperse systems structure formation factor is the formation of interphase stresses powerful gradients at the thickness of interphase layers.

Predicting fetal alcohol spectrum disorders in preschool-aged children from early life factors.

Early life factors, including parental sociodemographic characteristics, pregnancy exposures, and physical and neurodevelopmental features measured in infancy are associated with fetal alcohol spectrum disorders (FASD). The objective of this study was to evaluate the performance of a classifier model for diagnosing FASD in preschool-aged children from pregnancy and infancy-related characteristics. We analyzed a prospective pregnancy cohort in Western Ukraine enrolled between 2008 and 2014. Maternal and paternal sociodemographic factors, maternal prenatal alcohol use and smoking behaviors, reproductive characteristics, birth outcomes, infant alcohol-related dysmorphic and physical features, and infant neurodevelopmental outcomes were used to predict FASD. Data were split into separate training (80%: n = 245) and test (20%: n = 58; 11 FASD, 47 no FASD) datasets. Training data were balanced using data augmentation through a synthetic minority oversampling technique. Four classifier models (random forest, extreme gradient boosting [XGBoost], logistic regression [full model] and backward stepwise logistic regression) were evaluated for accuracy, sensitivity, and specificity in the hold-out sample. Of 306 children evaluated for FASD, 61 had a diagnosis. Random forest models had the highest sensitivity (0.54), with accuracy of 0.86 (95% CI: 0.74, 0.94) in hold-out data. Boosted gradient models performed similarly, however, sensitivity was less than 50%. The full logistic regression model performed poorly (sensitivity = 0.18 and accuracy = 0.65), while stepwise logistic regression performed similarly to the boosted gradient model but with lower specificity. In a hold-out sample, the best performing algorithm correctly classified six of 11 children with FASD, and 44 of 47 children without FASD. As early identification and treatment optimize outcomes of children with FASD, classifier models from early life characteristics show promise in predicting FASD. Models may be improved through the inclusion of physiologic markers of prenatal alcohol exposure and should be tested in different samples.