Backcalculation of full-depth asphalt pavement layer moduli considering nonlinear stress-dependent subgrade behavior

A condition assessment procedure for pavement layers that uses multiload-level falling weight deflectometer (FWD) deflections is presented. A dynamic finite element program that incorporates a stress-dependent soil model was used to generate the synthetic deflection database. On the basis of the data in this database, the relationships between surface deflections and critical pavement responses, such as the stresses and strains in each layer, have been established. The FWD deflection data, distress survey results, temperature, and laboratory testing results used to develop this procedure were collected from the Long-Term Pavement Performance project database. Research efforts also focused on the effect of the FWD load level on the condition assessment procedure. The results indicate that the proposed procedure can estimate the asphalt concrete (AC), base, and subgrade layer conditions. The AC layer modulus and the tensile strain at the bottom of the AC layer were found to be better indicators of the condition of the AC layer than the deflection basin parameter. It was also found that the structurally adjusted base damage index and base curvature index were good indicators for prediction of the stiffness characteristics of the aggregate base and subgrade, respectively. An FWD test with a load of 71.2 kN or less does not improve the accuracy of this procedure. The results from the study of the nonlinear behavior of a pavement structure indicate that the deflection ratio obtained from multiload-level deflections can predict the type and quality of the base and subgrade materials.

A new algorithm for predicting layer moduli using measurements from both falling weight deflectometer (FWD) and surface wave tests is presented. This algorithm employs numerical solutions of a multilayered half-space based on Hankel transforms as a forward model and an artificial neural network (ANN) for the inversion process. Phase velocities for frequencies ranging from 10 Hz to 10,000 Hz are calculated using the forward model for varying pavement structures with a range of layer moduli and thicknesses. These phase velocities, along with the layer moduli and thicknesses, are used to train an ANN to backcalculate layer moduli from dispersion curves (i.e., phase velocity versus frequency curves) constructed from the FWD and stress wave test data. To account for the effect of bedrock on the moduli prediction, another network is trained with layer thicknesses and phase velocities for predicting the depth to bedrock. Combining this network with the network for the moduli prediction results in a sequential dispersion analysis technique in which the depth to bedrock predicted from the first network becomes an input to the second network for predicting layer moduli. FWD and stress wave test measurements from an intact pavement and an asphalt overlay over cracked asphalt layer are processed using the sequential dispersion analysis technique and MODULUS 5.0 backcalculation program. Comparison of the results indicates that the dispersion analysis technique yields less variable subgrade moduli and is more sensitive to changes in the asphalt surface layer, because the high-frequency data from the stress wave test is incorporated.

The Falling Weight Deflectometer (FWD) is one of the most important non-destructive tests (NDT) used for pavement structural evaluation based on deflection measurements. Inverse analysis of FWD measurements—back-calculation—is a common process for obtaining the pavement layer moduli, including the subgrade soil, which are useful for estimating remaining pavement life for overlay design purposes. Based on a proficiency test (PT) involving testing devices from different manufacturers, the objective of this study is to analyze the influence of FWD uncertainty on the back-calculated flexible pavement and foundation layers moduli. A parametric study was performed considering the back-analysis of different asphalt pavements with regard to layer thickness and resilient modulus (unbound granular and asphalt concrete layers, including subgrade). It was concluded, in general, that FWD uncertainty could have an important influence on the layers modulus obtained from back-calculations. This influence was less significant in the case of asphalt concrete layers (±25%). The variation in resilient modulus was greater in the layers of the subgrade soil (−33 to +83%) and the unbound granular materials (−42 to +80%).

A pavement temperature profile is indispensable for analyzing and interpreting falling weight deflectometer (FWD) testing data. This paper presents a simple analytical approach to rapidly predicting the transient temperature profile within the asphalt layer of a flexible pavement on the basis of measured surface temperatures during FWD testing. The proposed analytical approach consists of the classical solution of a one-dimensional heat equation in a homogeneous half-space subjected to specified surface temperatures and a Gaussian-quadrature numerical scheme to approximate the classical solution. Model validation was conducted using a large temperature database. Comparisons between predicted and measured pavement subsurface temperatures suggest that the proposed simplified analytical method generated a reasonable asphalt temperature profile during the short period of FWD testing. The main advantage of the proposed method is that it can be easily applied to assist field engineers in rapidly estimating the temperature profile within the asphalt layer of a multilayered flexible pavement based on measured pavement surface temperatures during FWD testing.

The primary objective of this research was to develop a model to accurately predict the modulus of flexible pavement layers from surface deflections measured using the falling weight deflectometer (FWD) device. To do so, a synthetic dataset consisting of 10,000 flexible pavements was created using the layered elastic theory. The developed dataset contained the moduli values for different pavement sections, and deflections at known distances from the load center. Next, the moduli of different asphalt pavement layers consisting of a surface course, base course, and subgrade were calculated using the Artificial Neural Network (ANN) methodology through backcalculation. The inputs for the neural network are thicknesses and deflection values at seven distances from the load center. The outputs are the moduli for different layers. The optimum neural network consists of two hidden layers and has a general architecture of 9-36-18-3. The transfer function is sigmoid for hidden layers and linear for the output layer. Results indicate that the ANN model can predict the modulus of different layers accurately with a coefficient of determination (R2) of more than 0.999 in all cases. For validation, the results from the developed model were compared with typical backcalculation software of ISSEM4, MODCOMP, MODULUS, WESDEF, and BAKFAA as well as 386 LTPP sections. Moreover, an analysis was conducted to assess the contribution of inputs in moduli prediction for each of the pavement layers using the Garson algorithm and the Connection Weight methods. It was concluded that the results from the latter method are in better agreement with the theories. Finally, the significance of different input variables was assessed using an indirect method by excluding each from the analysis and checking of the model predictability power. It was concluded that in order of significance, the base layer thickness, 1st geophone deflection, and the AC layer thickness are the variables that their exclusion causes large errors.

The generalization ability of artificial neural networks (ANNs) and M5 model tree (M5Tree) in modeling reference evapotranspiration (ET 0 ) is investigated in this study. Daily climatic data, average temperature, solar radiation, wind speed, and relative humidity from six different stations operated by California Irrigation Management Information System (CIMIS) located in two different regions of the USA were used in the applications. King-City Oasis Rd., Arroyo Seco, and Salinas North stations are located in San Joaquin region, and San Luis Obispo, Santa Monica, and Santa Barbara stations are located in the Southern region. In the first part of the study, the ANN and M5Tree models were used for estimating ET 0 of six stations and results were compared with the empirical methods. The ANN and M5Tree models were found to be better than the empirical models. In the second part of the study, the ANN and M5Tree models obtained from one station were tested using the data from the other two stations for each region. ANN models performed better than the CIMIS Penman, Hargreaves, Ritchie, and Turc models in two stations while the M5Tree models generally showed better accuracy than the corresponding empirical models in all stations. In the third part of the study, the ANN and M5Tree models were calibrated using three stations located in San Joaquin region and tested using the data from the other three stations located in the Southern region. Four-input ANN and M5Tree models performed better than the CIMIS Penman in only one station while the two-input ANN models were found to be better than the Hargreaves, Ritchie, and Turc models in two stations.

BackgroundThis study aimed to develop an artificial neural network (ANN) model for predicting synchronous organ-specific metastasis in lung cancer (LC) patients.MethodsA total of 62,151 patients who diagnosed as LC without data missing between 2010 and 2015 were identified from Surveillance, Epidemiology, and End Results (SEER) program. The ANN model was trained and tested on an 75/25 split of the dataset. The receiver operating characteristic (ROC) curves, area under the curve (AUC) and sensitivity were used to evaluate and compare the ANN model with the random forest model.ResultsFor distant metastasis in the whole cohort, the ANN model had metrics AUC = 0.759, accuracy = 0.669, sensitivity = 0.906, and specificity = 0.613, which was better than the random forest model. For organ-specific metastasis in the cohort with distant metastasis, the sensitivity in bone metastasis, brain metastasis and liver metastasis were 0.913, 0.906 and 0.925, respectively. The most important variable was separate tumor nodules with 100% importance. The second important variable was visceral pleural invasion for distant metastasis, while histology for organ-specific metastasis.ConclusionsOur study developed a “two-step” ANN model for predicting synchronous organ-specific metastasis in LC patients. This ANN model may provide clinicians with more personalized clinical decisions, contribute to rationalize metastasis screening, and reduce the burden on patients and the health care system.

The present study was carried out using artificial neural network (ANN) model for predicting the sodium hazardness, i.e., sodium adsorption ratio (SAR), percent sodium (%Na) residual, Kelly’s ratio (KR), and residual sodium carbonate (RSC) in the groundwater of the Pratapgarh district of Southern Rajasthan, India. This study focuses on verifying the suitability of water for irrigational purpose, wherein more groundwater decline coupled with water quality problems compared to the other areas are observed. The southern part of the Rajasthan State is more populated as compared to the rest of the parts. The southern part of the Rajasthan is more populated as compared to the rest of the Rajasthan, which leads to the industrialization, urbanization, and evolutionary changes in the agricultural production in the southern region. Therefore, it is necessary to propose innovative methods for analyzing and predicting the water quality (WQ) for agricultural use. The study aims to develop an optimized artificial neural network (ANN) model to predict the sodium hazardness of groundwater for irrigation purposes. The ANN model was developed using ‘nntool’ in MATLAB software. The ANN model was trained and validated for ten years (2010–2020) of water quality data. An L-M 3-layer back propagation technique was adopted in ANN architecture to develop a reliable and accurate model for predicting the suitability of groundwater for irrigation. Furthermore, statistical performance indicators, such as RMSE, IA, R, and MBE, were used to check the consistency of ANN prediction results. The developed ANN model, i.e., ANN4 (3-12-1), ANN4 (4-15-1), ANN1 (4-5-1), and ANN4 (3-12-1), were found best suited for SAR, %Na, RSC, and KR water quality indicators for the Pratapgarh district. The performance analysis of the developed model (3-12-1) led to a correlation coefficient = 1, IA = 1, RMS = 0.14, and MBE = 0.0050. Hence, the proposed model provides a satisfactory match to the empirically generated datasets in the observed wells. This development of water quality modeling using an ANN model may help to useful for the planning of sustainable management and groundwater resources with crop suitability plans as per water quality.

ABSTRACTStatistical methods such as Poisson distribution, negative binomial regression (NB), and zero inflated negative binomial regression (ZINB) have generally been used in road safety studies to establish the complex relationships between variables. Over the last few years, the artificial neural networks (ANN) model has also been used. The ANN model does not have any prior limitations such as the equality condition of mean and variance observed in Poisson regression. However, though the ANN model has been used in the analysis of different accident types, to the best of our knowledge, no study has used the ANN model for establishing this relationship with truck accident data on divided multilane interurban roads. In this study, the road sections D750/07–D750/15 in Ankara–Aksaray–Eregli, Turkey, were considered and truck accident data from 2008 to 2011 were analyzed using NB and ANN. The analysis show that the ANN model has lower errors and higher R2 values than NB and performs slightly better than NB for predicting the number of trucks involved in accidents. Based on a comparison of performances the study concludes, that ANN could be used as an alternative model for analyzing truck accident data on divided multilane interurban roads in Turkey.

Purpose – This paper aims to predict the needle penetration force (NPF) in denim fabrics using the artificial neural network (ANN) and multiple linear regression (MLR) models based on the effects of various sewing parameters. Design/methodology/approach – In order to design the ANN and MLR models, four parameters including fabric weight, number of fabric layers, weave pattern, and sewing needle size are taken into account as the input parameters and NPF as the output parameter. According to these parameters, 140 samples of data were resulted. Each sample was tested five times. From these 140 data (input-output data pairs), 112 were used for training the ANN and MLR models and 28 samples were used to test the performance of ANN and MLR. Also, the NPF was measured on the Instron tensile tester to simulate sewing process. Findings – The results indicated that the NPF in denim fabrics can be well predicted in terms of sewing parameters by using ANN and MLR models, in which the ANN model exhibits greater performance than MLR (RANN=0.989> RMLR=0.901). Research limitations/implications – The NPF measurement method is limited at low speed. Originality/value – Using the ANN model for forecasting NPF in denim fabrics can help the garment manufactures to produce high-quality denim products and improve the sewing process through reducing seam damage. The NPF could be also measured in the cycle loading conditions similar to sewing machine process by using a special designed tools mounted on the Instron tensile tester.

This paper presents a modeling study based on finite element (FE) analysis to mechanistically evaluate flexible pavements constructed with quarry by-products (QB). Twelve pavement test sections, a control section, and 11 others incorporating QB as unbound subgrade replacement and chemically stabilized base/subbase applications were evaluated for field performance using accelerated pavement testing (APT). First, falling weight deflectometer (FWD) deflection basin parameters were calculated and compared with critical pavement responses to evaluate the structural adequacies of the QB pavement sections. The moduli of the constructed pavement layers were then backcalculated from the FWD deflections using the GT-PAVE FE analysis program with nonlinear and cross-anisotropic layer characterizations. For stabilized QB applications, the layer properties calculated from this analysis for the base/subbase layers were used to calculate critical pavement responses. The sections were compared using a response benefit parameter, defined as the ratio of maximum resilient surface deflection in a conventional pavement, as the control section, to that obtained for each section having a certain QB application in consideration. According to the results obtained from the APT sections and the mechanistic FE analyses, the measured and calculated FWD deflection basins were successfully matched with individual sensor errors not exceeding 5% for all 12 test sections. The calculated response benefits indicated significant advantages for using cement-stabilized QB applications over fly ash-stabilized QB applications and conventional flexible pavement sections. Considering the pavement structural response benefits and good performance trends observed, major cost benefits can be realized by routine use of these sustainable QB applications.

Monthly river flow forecasting using artificial neural network and support vector regression models coupled with wavelet transform

The falling weight deflectometer (FWD) test is a common non-destructive testing method for evaluating the structural capacity of pavements. At present, data processing of the FWD test mainly focuses on the deflection data, while paying less attention to the deflection-time history. Because a FWD is equipped with impulse loads and geophones, which allow for the generation and capture of surface wave signal propagation, it is hypothesized that Rayleigh wave dispersion theory can be applied to calculate the modulus profile along the pavement depth by analysing the dispersive properties of the deflection signal measured during FWD tests. To test this hypothesis, we develop a new methodology for the FWD test and data analysis, referred to as the FWD dispersion curve method. We first introduce the concept of the new method, followed by an illustration of the procedure and the experimental set-up. Case studies on three concrete pavement segments are then presented to evaluate the effectiveness of the FWD dispersion curve method. Modifications to the existing FWD device are further recommended for the impact loading sources and signal collection process so that the modulus of a much shallower layer, such as the concrete slab and upper asphalt layers, can be obtained. This article is part of the theme issue 'Artificial intelligence in failure analysis of transportation infrastructure and materials'.

The power capacity of solar photovoltaics (PVs) in Korea has grown dramatically in recent years, and an accurate estimation of solar resources is crucial for the efficient management of these solar PV systems. Since the number of solar irradiance measurement sites is insufficient for Korea, satellite images can be useful sources for estimating solar irradiance over a wide area of Korea. In this study, an artificial neural network (ANN) model was constructed to calculate hourly global horizontal solar irradiance (GHI) from Korea Communication, Ocean and Meteorological Satellite (COMS) Meteorological Imager (MI) images. Solar position variables and five COMS MI channels were used as inputs for the ANN model. The basic ANN model was determined to have a window size of five for the input satellite images and two hidden layers, with 30 nodes on each hidden layer. After these ANN parameters were determined, the temporal and spatial applicability of the ANN model for solar irradiance mapping was validated. The final ANN ensemble model, which calculated the hourly GHI from 10 independent ANN models, exhibited a correlation coefficient (R) of 0.975 and root mean square error (RMSE) of 54.44 W/m² (12.93%), which were better results than for other remote-sensing based works for Korea. Finally, GHI maps for Korea were generated using the final ANN ensemble model. This COMS-based ANN model can contribute to the efficient estimation of solar resources and the improvement of the operational efficiency of solar PV systems for Korea.

Falling Weight Deflectometer (FWD) testing is used for nondestructive structural evaluation of existing pavements. The deflection data from FWD can be used for estimation of layer moduli by an iterative technique. In the calculation process, commonly known as backcalculation, comparison is done between calculated deflections from a suitable theory and those measured by the FWD. The estimated moduli values correspond to the best fit of the measured deflection basin. In the present study, a nonlinear optimization technique has been used to estimate the layer moduli in the backcalculation process. The backcalculation scheme developed is applicable to 3-layer to 6-layer pavements. The method was verified by simulated deflection data from elastic-layer theory and FWD-deflection measurements on 13 in-service pavements in Arizona. They were found to approximate closely the manual approach of matching deflection basins. The algorithm has been coded in a computer program called the Arizona Deflection Analysis Method (ADAM).