Input Material Properties Research Articles

Stability Evaluation of Diversion Tunnel Portal Slopes at Lau Simeme Dam Site, Indonesia, using Limit Equilibrium Method

Construction of the Lau Simeme Dam used a tunnel as a diversion channel. Slopes at the diversion tunnel portals were prone to failure due to the tunnel excavation and earthquake. Earthquake load was not considered in the designs of the inlet and outlet portal slopes. This research evaluated stability of the tunnel portal slopes under static and earthquake loads using limit equilibrium methods of the Bishop Simplified and Morgenstern-Price. Input material properties for the slope stability analyses were obtained from evaluations of soil and rock cores, including determination of lithology type and rock mass quality based on Geology Strength Index (GSI), and laboratory testing. Evaluations of soil and rock cores indicated that the inlet portal slope consisted of residual soil, good quality tuff breccia, and good quality sandstone. The outlet portal slope consisted of residual soil, poor quality sandstone, poor quality tuff breccia, fair quality sandstone, fair quality tuff breccia, and good quality tuff breccia. The earthquake load coefficient determined by the seismic analyses based on the peak ground acceleration map for 10% probability exceedance in 50 year was 0.125 g. The slope stability analyses showed that the designed inlet and outlet portal slopes were stable under static and earthquake loads. The Bishop Simplified and Morgenstern-Price resulted in relatively similar Fs values. The Fs values of the inlet and outlet portal slopes decreased with the earthquake load application. Although Fs values of the outlet slope under static and earthquake loads met the requirements specified by SNI 8460:2017, the rock mass conditions, particularly the poor rock masses of layers 2 and 3, required special attention. Application of slope reinforcement methods, such as shotcrete, is suggested to further increase the slope stability.

Ultra-efficient and parameter-free computation of submicron thermal transport with phonon Boltzmann transport equation

Understanding thermal transport at the submicron scale is crucial for engineering applications, especially in the thermal management of electronics and tailoring the thermal conductivity of thermoelectric materials. At the submicron scale, the macroscopic heat diffusion equation is no longer valid and the phonon Boltzmann transport equation (BTE) becomes the governing equation for thermal transport. However, previous thermal simulations based on the phonon BTE have two main limitations: relying on empirical parameters and prohibitive computational costs. Therefore, the phonon BTE is commonly used for qualitatively studying the non-Fourier thermal transport phenomena of toy problems. In this work, we demonstrate an ultra-efficient and parameter-free computational method of the phonon BTE to achieve quantitatively accurate thermal simulation for realistic materials and devices. By properly integrating the phonon properties from first-principles calculations, our method does not rely on empirical material properties input. It can be generally applicable for different materials and the predicted results can match well with experimental results. Moreover, by developing a suitable ensemble of advanced numerical algorithms, our method exhibits superior numerical efficiency. The full-scale (from ballistic to diffusive) thermal simulation of a 3-dimensional fin field-effect transistor with 13 million degrees of freedom, which is prohibitive for existing phonon BTE solvers even on supercomputers, can now be completed within two hours on a single personal computer. Our method makes it possible to achieve the predictive design of realistic nanostructures for the desired thermal conductivity. It also enables accurately resolving the temperature profiles at the transistor level, which helps in better understanding the self-heating effect of electronics.

Study of an acoustic energy harvester consisting of electro-spun polyvinylidene difluoride nanofibers.

Sound pollution has been capturing more and more attention around the world. Piezoelectric materials convert acoustic energy into electrical energy and actively attenuate the sound simultaneously. In this paper, an electro-spun nonwoven polyvinylidene difluoride nanofiber membrane as a high-performance piezoelectric material is found to have an ultra-high acoustoelectric conversion capability at the low sound frequency range. The novelty of the material in this paper is the proposed electro-spun piezoelectric nano-fiber web, which presents a strong acoustic-to-electric conversion performance. The piezoelectric acoustic energy harvester consists of the polyvinylidene difluoride nanofiber membrane that vibrates under the sound wave excitation. The piezoelectric acoustic energy harvester device can precisely detect the sound of 72.5 Hz with a sensitivity as high as 711.3 mV Pa-1 which is higher than the sensitivity of a commercial piezoelectric poly (vinylidene fluoride) membrane device. The energy harvesting performance of the piezoelectric acoustic energy harvester device is simulated by the comsol software and then validated with the experimental results to illustrate its excellent energy harvesting ability. Based on the validated simulation model, a regression parameter model is developed from the comsol software simulation results using the response surface method. The empirical regression parameter model is applied to predict the energy harvesting performance of the acoustic energy harvester from input design parameters or material property parameters where the sensitivity of the design parameters or material property parameters and their interactions can be analyzed. The design or material property parameters can be optimized for the best energy harvesting performance based on the regression parameter model. The optimization results show a significant improvement in the energy harvesting performance. The sensitivity of the parameters on the energy harvesting performance also indicates the potential of the large-scale application of this acoustic energy harvester.

Robust Metal Additive Manufacturing Process Selection and Development for Aerospace Components

Metal additive manufacturing (AM) encapsulates the myriad of manufacturing processes available to meet industrial needs. Determining which of these AM processes is best for a specific aerospace application can be overwhelming. Based on the application, each of these AM processes has advantages and challenges. The most common metal AM methods in use include Powder Bed Fusion, Directed Energy Deposition, and various solid-state processes. Within each of these processes, there are different energy sources and feedstock requirements. Component requirements heavily affect the process determination, despite existing literature on these AM processes (often inclusive of input parameters and material properties). This article provides an overview of the considerations taken for metal AM process selection for aerospace components based on various attributes. These attributes include geometric considerations, metallurgical characteristics and properties, cost basis, post-processing, and industrialization supply chain maturity. To provide information for trade studies and selection, data on these attributes were compiled through literature reviews, internal NASA studies, as well as academic and industry partner studies and data. These studies include multiple AM components and sample build experiments to evaluate (1) material and geometric variations and constraints within the processes, (2) alloy characterization and mechanical testing, (3) pathfinder component development and hot-fire evaluations, and (4) qualification approaches. This article summarizes these results and is meant to introduce various considerations when designing a metal AM component.

Coupling machine learning and stochastic finite element to evaluate heterogeneous concrete infrastructure

Probabilistic analyses facilitate the incorporation of inherent uncertainties in dam safety evaluation. Their application is limited due to the high computational burden, especially for the complex transient analyses and the cases in which the random field theory is applied to account for spatial uncertainty in material properties. In this work, uncertainty in the heterogeneous concrete is modeled by a series of probabilistic seismic finite element simulations. The outcome of stochastic models is fed into a random forests model for further post-processing and sensitivity analysis. The challenging aspect of this research lies in the fact that the initial number of finite element realizations is relatively small (due to computational time) regarding the random variables. Despite such an unfavorable setting, the meta-model showed reasonable accuracy for predicting the maximum displacement at the crest and the maximum stress at the base. A comprehensive discussion is provided on the variable importance and the partial effect of the inputs: the areas in the dam body with the highest impact in the response are identified, and the spatial distribution of input material properties is appropriately captured. These results show the possibilities of random forests in high-dimensional problems for predicting and interpreting dam behavior.

Magnetic Resonance Elastography Reconstruction for Anisotropic Tissues.

Elastography has become widely used clinically for characterising changes in soft tissue mechanics that are associated with altered tissue structure and composition. However, some soft tissues, such as muscle, are not isotropic as is assumed in clinical elastography implementations. This limits the ability of these methods to capture changes in anisotropic tissues associated with disease. The objective of this study was to develop and validate a novel elastography reconstruction technique suitable for estimating the linear viscoelastic mechanical properties of transversely isotropic soft tissues. We derived a divergence-free formulation of the governing equations for acoustic wave propagation through a linearly transversely isotropic viscoelastic material, and transformed this into a weak form. This was then implemented into a finite element framework, enabling the analysis of wave input data and tissue structural fibre orientations, in this case based on diffusion tensor imaging. To validate the material constants obtained with this method, numerous in silico phantom experiments were run which encompassed a range of variations in wave input directions, material properties, fibre structure and noise. The method was also tested on ex vivo muscle and in vivo human volunteer calf muscles, and compared with a previous curl-based inversion method. The new method robustly extracted the transversely isotropic shear moduli (G⊥', G∥', G″) from the in silico phantom tests with minimal bias, including in the presence of experimentally realistic levels of noise in either fibre orientation or wave data. This new method performed better than the previous method in the presence of noise. Anisotropy estimates from the ex vivo muscle phantom agreed well with rheological tests. In vivo experiments on human calf muscles were able to detect increases in muscle shear moduli with passive muscle stretch. This new reconstruction method can be applied to quantify tissue mechanical properties of anisotropic soft tissues, such as muscle, in health and disease.

Model updating of wind turbine blade cross sections with invertible neural networks

AbstractFabricated wind turbine blades have unavoidable deviations from their designs due to imperfections in the manufacturing processes. Model updating is a common approach to enhance model predictions and therefore improve the numerical blade design accuracy compared to the built blade. An updated model can provide a basis for a digital twin of the rotor blade including the manufacturing deviations. Classical optimization algorithms, most often combined with reduced order or surrogate models, represent the state of the art in structural model updating. However, these deterministic methods suffer from high computational costs and a missing probabilistic evaluation. This feasibility study approaches the model updating task by inverting the model through the application of invertible neural networks, which allow for inferring a posterior distribution of the input parameters from given output parameters, without costly optimization or sampling algorithms. In our use case, rotor blade cross sections are updated to match given cross‐sectional parameters. To this end, a sensitivity analysis of the input (material properties or layup locations) and output parameters (such as stiffness and mass matrix entries) first selects relevant features in advance to then set up and train the invertible neural network. The trained network predicts with outstanding accuracy most of the selected cross‐sectional input parameters for different radial positions; that is, the posterior distribution of these parameters shows a narrow width. At the same time, it identifies some parameters that are hard to recover accurately or contain intrinsic ambiguities. Hence, we demonstrate that invertible neural networks are highly capable for structural model updating.

Prediction of fire resistance of concrete encased steel composite columns using artificial neural network

Concrete encased steel (CES) columns, also known as steel reinforced concrete (SRC) composite columns, exhibit superior fire resistance due to concrete acting as a protection layer for the embedded steel section. While modern design codes have provided design guides for the fire resistance of CES columns, they are only applicable to those made of normal strength concrete. For high strength CES columns, advanced analysis is needed to capture the brittleness of high strength concrete at elevated temperature. In this paper, two methods, namely the artificial neural network (ANN) and the analytical equations, are proposed to predict the fire resistance of axially-loaded CES columns made of high strength concrete. To train the ANN, a finite difference model is developed to compute the temperature field in CES columns and it is used to establish a database containing 15,200 specimens. The cross-sectional dimensions and materials grades of the specimens are carefully selected to cover a wide range of values including those commonly adopted in real-life applications. The inputs of the ANN are identified through an extensive parametric analysis. The selected ANN consists of 7 inputs, 3 outputs and 2 hidden layers and achieves a high determination coefficient R2 value of 0.999. For practical implementation, analytical equations are also derived and achieve high R2 values above 0.953. The predictive power of the ANN and the analytical equations are examined against the observations obtained from actual fire tests, showing reasonable accuracy of prediction. Both methods are simple, of high accuracy and have implicitly accounted for temperature-dependent material degradation, and hence do not require input of temperature-dependent material properties and advanced analysis software.

A Bayesian machine learning approach to rapidly quantifying the fatigue probability of failure for steel catenary risers

Quantifying the risk of system failure is a key input to a risk-based decision-making process for the in-service integrity management of aging steel catenary risers (SCRs), which are prone to fatigue failure within the touchdown zone (TDZ). However, in practice the computational cost of assessing the probability of failure (POF) for SCRs is prohibitive. In this paper, we propose an efficient framework for quantifying the fatigue POF within the SCR TDZ by using the Bayesian machine learning technique adopting Gaussian Processes (GP) for regression. GP-based predictive models perform stochastic response prediction rapidly and are therefore attractive for rapid evaluation of the SCR fatigue POF. In this paper, we introduce innovative techniques for predicting fatigue damage profiles within the SCR TDZ generated by an irregular sea-state applied to the host vessel at its mean offset position, and for dimensionality reduction in the regression analysis. We utilize a realistic case study to demonstrate the framework, which consists of a representative 20″ SCR connected to a semi-submersible host vessel located in 950 m water depth. In addition, we use a site-specific 32-year hindcast wave data and 1-year measured current data, with associated statistical distributions, for the long-term fatigue loading conditions. We also employ numerical simulation and a random sampling technique to create simulation-based input and output datasets for training and testing of the derived probabilistic simulation-based surrogate. The results show that the proposed method is able to predict the fatigue life of the representative riser efficiently and accurately. Finally, we demonstrate the usefulness of the proposed method by rapidly generating the fatigue POF for the 20″ SCR considering the uncertainties associated with input loads, material properties, geometric parameters, and soil stiffness. • A practical machine-learning frameworks is proposed to enable adopting quantified risk-based process for in-service integrity management of SCRs. • A systematic approach is introduced for developing and verifying probabilistic simulation-based surrogates using the Gaussian Process concept for rapidly predicting the damage profile at SCR touchdown zone (TDZ) due to irregular sea-states. • An efficient methodology is proposed for rapid and accurate quantification of the fatigue probability of failure (POF) at SCRs TDZ. • The accuracy and usefulness of the framework is demonstrated using a realistic case study where the probability distribution of the fatigue life of a representative 20″ SCR is quantified via Monte Carlo Simulation.

Propagation of Input Uncertainties in Numerical Simulations of Laser Powder Bed Fusion

Laser powder bed fusion has the potential of redefining state-of-the-art processing and production methods, but defect formation and inconsistent build quality have limited the implementation of this process. Numerical models are widely used to study this process and predict the formation of these defects. Presently, the uncertainties of model input parameters and thermophysical properties used by these numerical simulations have not been investigated. In the present study, the uncertainty in these input parameters and material properties are quantified for laser powder bed fusion, with and without a simulated powder bed, to determine their influence on the predictive accuracy of an experimentally validated numerical model. Accounting for all possible sources of uncertainty quickly becomes computationally expensive on account of the curse of dimensionality. Uncertainty in laser absorption, solid, and liquid specific heat of the metal were found to have the largest effect on model prediction reliability with or without the use of a powder bed. Results also illustrate that accounting for these three uncertain parameters still captures the majority of model prediction uncertainty. Furthermore, the methodology of this study may be used to understand the uncertainty in as-built microstructure through propagation to microstructure prediction models, or applied under processing conditions where high Péclet numbers are observed and the thermal convection and fluid flow within the molten pool are substantial.

Comparison on two ways of phonon spectral property reconstruction from thermal spectroscopy experiment

We examine the reliability of the cumulative thermal conductivity as a function of the free path, F(Λ), in the context of reconstruction of phonon transport properties from thermal spectroscopy experiments. We specifically show that a given F(Λ)does not correspond to a unique relaxation-times function, in the sense that more than one distribution of relaxation-times can result in the same F(Λ). Since different relaxation-time distributions will, in general, lead to different thermal responses, F(Λ) does not uniquely predict the material thermal response in all transport regimes. This implies that in the context of thermal transport at the nanoscale within the Boltzmann relaxation-time approximation framework, the “fundamental” property that should be reconstructed from the thermal spectroscopy experiments is the frequency-dependent relaxation-times function (provided group velocities are known), since it explicitly appears in the governing equation as the input material property. Extensive global optimization studies show that a previously proposed formulation for reconstruction based on the frequency-dependent relaxation-times function [Physical Review B 94, 155439 (2016)] provides numerically unique solutions.

Validation of a multi-scale simulation strategy based on the Pointwise Strain Superposition Method

This paper details the experimental validation of a multi-scale simulation strategy that we developed for predicting the stresses and distortions induced by Powder Bed Fusion processes. The strategy comprises a meso-scale model, a macro-scale model, and a scaling method named Pointwise Strain Superposition. The first model evaluates the temperature, stress, and strain fields produced by a single scan line. The scaling method transfers the meso-scale results to the macro-scale model, which is then able to simulate the entire manufacturing process with a reasonable computational cost. The simulation strategy was validated by comparing its results with the stresses and distortions measured on several specimens made of selective laser melted Inconel 718. Stresses were measured through the blind hole drilling method on a cylindrical specimen printed with two different scanning strategies, while distortions were measured on a hollow cylinder and on a cantilever-shaped specimen after removing its supports. In both cases the simulation showed first- or higher-order accuracy despite the significant uncertainties regarding the input parameters and material properties. This robustness, coupled with its computational efficiency, leads us to believe that our simulation strategy could enhance the process optimization and provide a better understanding of the underlying physical phenomena along with their effects on the manufactured parts.

A Fast-Acting Method for Simulating Precipitation During Heat Treatment of Superalloy 718

A fast-acting, mean-field method for simulating precipitation of the γ′′ and γ′ phases during aging of superalloy 718 following super-delta-solvus solution treatment was formulated and validated using observations in the literature. The approach assumed classical (homogeneous) nucleation and diffusion-controlled growth (N&G) of disk/ellipsoidal-shaped-γ′′ and spherical-γ′ particles. For the γ′′ precipitates in particular, the evolution equations for both nucleation and growth incorporated corrections for the non-spherical shape, assuming a fixed aspect ratio. In addition, special attention was paid to the choice of input material properties for simulations. These parameters included the bulk free energies of transformation, particle-matrix (misfit) elastic strain energy (for γ′′), effective diffusivities, and the γ′′–γ and γ′–γ interface energies. The applicability of the diffusivities and interface energies chosen for the N&G simulations was established by their consistency in replicating previously measured rate constants for the diffusion-controlled coarsening of both γ′′ and γ′. The N&G formulation was discretized to obtain numerical (spreadsheet) solutions via the Kampmann–Wagner approach. Simulation results for the temporal evolution of volume fraction and average size of the precipitates showed good agreement with experimental measurements. The sensitivity of model predictions to various input parameters was also quantified.

Wood sheathing durability from moisture sorption isotherm variability due to age and temperature

Hygrothermal modelling can be superior to experimental work in exploring durability of building components and assemblies, however, these models require accurate material property inputs. The results of hygrothermal models can be used to calculate mould growth risk, an indicator of building durability. Mould growth is particularly important in organic building materials such as wood. Wood framed buildings are extremely common in North American low-rise residential construction, and part of these systems is sheathing board. OSB and plywood are the most commonly used types of sheathing board. Material properties of OSB and plywood sheathing include moisture storage, represented by the sorption isotherm – however, some common hygrothermal modelling programs do not account for varying sorption isotherms. This study examines the variability of sorption isotherms in OSB and plywood sheathing after accelerated ageing and under different temperature conditions. The measured data was then input into hygrothermal models, and the results were post-processed to determine mould growth risk via the mould index, a scale for mould growth. Accounting for age and temperature resulted in differences of up to 60% in mould index, with the adjusted models generally displaying lower mould index values. Therefore, the use of a single static isotherm instead of one that accounts for age and temperature can result in the over-prediction of mould growth risk of up to 60% in low-rise residential wood framed walls using OSB or plywood sheathing board.

Factors affecting the solid particle erosion of environment pollutant and natural particulate filled polymer composites—A review

Solid particle erosion of polymer matrix composites filled with naturally available and environment pollutant fillers have not been studied to the same level as for metals or ceramics and is focus of the present study. In this article, review of the research associated with the erosion response of polymer composites is presented. Particulate polymer composites are employed extensively owing to their enhanced specific properties and tribological response. Particulate filler particles such as environmental pollutants and naturally available ones need to be effectively incorporated in utilitarian applications so as to reduce land fill burden issues and other specific problems. Nevertheless, adequate data is not available in review articles on the erosion of fillers that are environment pollutants and thereby an ample amount of research can be carried out in this regard. Erosion behavior of polymer composites in particular has gained a lot of attention among researches in the recent decade. Viability of incorporating various fillers in polymer matrix for erosion resistive applications needs to be assessed so that the potential of these composites can be well understood. Therefore in this study, erosion response of polymer composites reinforced with fillers is reviewed with a focus on input parameters (impact velocity, impingement angle and erodent properties) and material properties (density).

Compost Physical Properties Study on Degradation of Poultry Manure Composting in Closed-Aerated Composter

A variety of parameters including physical, chemical, and biological properties of different input materials contribute to different composting performance. This study aimed to investigate the compost physical properties (bulk density, porosity, specific surface area and water holding capacity) on the composting process at different initial moisture content (MC). The degradation of total organic carbon (TOC) for the compost inoculated with Bacillus coagulans (BC) and effective microorganism (EM) was determined. The composting materials consisted of 50 % sawdust, 12 % chicken dung and 38 % rice husk with a fixed initial C/N ratio of 30. A closed-aerated composter was fabricated with an optimum air flow rate of 0.3 L/min.kg compost to avoid O2 limitation for 7 d of composting. The compost temperature was recorded to exhibit the active reaction between microorganisms and compost materials will generate a considerable amount of heat. The effect of the initial MC of the compost bed has been intensively investigated with regards to compaction analysis and compost particle for the composting inoculated with BC or EM in an aerated closed-system composter. The results showed that composting using the single strain of BC provides comparable results to that degraded by the commercial mixed culture EM.

PREDICTION OF THE PERFORMANCE OF BITUMINOUS MIXES USING ADAPTIVE NEURO-FUZZY INFERENCE SYSTEMS

Accurately forecast performance and durability is a critical issue for improving the design of new and existing pavements. The poor pavement performance increases traffic congestion, compromises safety, and raises maintenance costs due to frequent repairs. The resilient modulus is one of the most critical unbound material property inputs in several current pavement design procedures. Recent studies have addressed the problem of resilient modulus prediction using different methods, including computational intelligence approaches. In this paper, a hybrid intelligent system called ANFIS (Adaptive Neuro-Fuzzy Inference System) is used for predicting the resilient modulus from an experimental database of 270 distinct compositions. ANFIS achieved superior performance when estimating the resilient modulus of bituminous mixes, which can potentially save laboratory resources.

Contact resistance and current crowding in tunneling type circular nano-contacts

Current transport and contact resistance in nanoscale electrical contacts are important to the overall device properties, especially for devices based on novel one-dimensional and two-dimensional materials or nanostructures. In this paper, we present a self-consistent method to model tunneling type circular thin film contacts. We solve the lumped circuit circular transmission line model (CTLM) with tunneling-induced specific contact resistivity which varies along the radial direction. The contacting members are separated by a thin insulating layer, where the radially dependent is calculated from local voltage dependent tunneling current density. The current and voltage distributions in such contacts and their overall contact resistance are studied in detail, for various input voltages, contact dimensions, and material properties (i.e. work function, sheet resistance of the contact members, and permittivity of the insulating layer). Our study shows that the contact resistance is voltage dependent, and the radial current distribution is strongly nonhomogeneous. The contact resistance and the current distribution can be controlled by engineering the contact layer properties and geometry radially. Although focused on Schrödinger tunneling type contacts in this work, our modified CTLM equations with radially varying are general, and may be readily used for other types of electrical contacts, such as ohmic and Schottky contacts.

Improvement of the Mechanical Properties of Thermosetting-Binding-System-Based Composites by Means of Kneading Procedure Modification and Composite Formulation

By understanding the effects of the physical properties of individual input materials (e.g., binding system) on the physical and thermal properties of a composite material, the latter can be engineered in advance according to the desired properties and application. Often, a need to replace a specific component in a composite material arises, due to various reasons such as high raw material prices, product price reduction, environmental issues, improvement of properties, and others. In this study, we focused on the substitution of a phenolic novolac resin binding system and the reduction of compounding process temperature in combination with material throughput and screw speed variation of a phenolic-novolac-resin-based composite material, manufactured by kneading process using a co-kneader single screw extruder. Modifications were carried out in the interest of reducing production process cost and positive environmental effect due to reduction of energy consumption in the compounding process. We achieved great success in improvement of mechanical properties with all four substituted phenolic molding compounds (PMCs), while the decrease in thermal stability was the lowest for PMCs prepared at higher screw speeds and material throughput. The results indicated that higher screw speeds produce the best combination of mechanical and thermal properties of PMCs.

Integrated finite element and artificial neural network methods for constructing asphalt concrete dynamic modulus master curve using deflection time-history data

Dynamic modulus |E*| is one of the essential material properties input in the American Association of State Highway and Transportation Officials (AASHTO) Mechanistic-Empirical Pavement Design Guide (MEPDG). Asphalt concrete (AC) dynamic modulus master curve is used to determine the modulus of asphalt concrete over a wide range of temperature and frequency. However, the standard laboratory test procedures for establishing asphalt concrete |E*| and plotting the AC |E*| master curve are time consuming and require considerable resources. Therefore, this study aims to predict AC |E*| master curve by using data from a falling weight deflectometer (FWD) deflection time-history. Prior to developing the model, a simple performance testing (SPT) dynamic modulus test was conducted in the laboratory on five core specimens to obtain the dynamic modulus data at several test temperatures and load frequencies. Results of SPT dynamic modulus show that the |E*| of all specimens is influenced by both loading rate and test temperature. The specimens are stiffer at low temperature and high frequency, and the |E*| values are the lowest at the highest temperature and lowest frequency. Artificial neural network (ANN) models are designed using the FWD deflection-time history data obtained by the finite element method (FEM) to predict the AC |E*| master curve. This study uses two types of ANN models, namely multilayer feed-forward neural network (MLFN) and radial basis function network (RBFN). ANN results show that both MLFN and RBFN models have a promising potential in the construction of AC |E*| master curve. A comparison of the two types of ANNs revealed that RBFN has a lower percentage of error, and is therefore more accurate than MLFN.