Grouping Pavement Condition Variables for Performance Modeling Using Self-Organizing Maps

Different modeling techniques have been employed for the evaluation of pavement performance, determination of structural capacity, and performance predictions. The evaluation of performance involves the functional analysis of pavements based on the history of the riding quality. The riding comfort and pavement performance can be conveniently defined in terms of roughness and pavement distresses. Thus different models have been developed relating roughness with distresses to predict pavement performance. These models are too complex and require parsimonious equations involving fewer variables. Artificial neural networks have been used successfully in the development of performance-prediction models. This article demonstrates the use of an artificial intelligence neural networks self-organizing maps for the grouping of pavement condition variables in developing pavement performance models to evaluate pavement conditions on the basis of pavement distresses.     

Stiffness characterisation of full-scale airfield test pavements using computational intelligence techniques

The falling weight deflectometer (FWD) is a non-destructive test equipment used to assess the structural condition of highway and airfield pavement systems and to determine the moduli of pavement layers. The backcalculated moduli are not only good pavement layer condition indicators but are also necessary inputs for conducting mechanistic based pavement structural analysis. In this study, artificial neural networks (ANNs)-based backcalculation models were employed to rapidly and accurately predict flexible airport pavement layer moduli from realistic FWD deflection basins acquired at the U.S. Federal Aviation Administration's National Airport Pavement Test Facility (NAPTF). The uniformity characteristics of NAPTF flexible pavements were successfully mapped using the ANN predictions.     

Assessing artificial neural network performance for predicting interlayer conditions and layer modulus of multi-layered flexible pavement

The objective of this study is to evaluate the performance of the artificial neural network (ANN) approach for predicting interlayer conditions and layer modulus of a multi-layered flexible pavement structure. To achieve this goal, two ANN based back-calculation models were proposed to predict the interlayer conditions and layer modulus of the pavement structure. The corresponding database built with ANSYS based finite element method computations for four types of a structure subjected to falling weight deflectometer load. In addition, two proposed ANN models were verified by comparing the results of ANN models with the results of PADAL and double multiple regression models. The measured pavement deflection basin data was used for the verifications. The comparing results concluded that there are no significant differences between the results estimated by ANN and double multiple regression models. PADAL modeling results were not accurate due to the inability to reflect the real pavement structure because pavement structure was not completely continuous. The prediction and verification results concluded that the proposed back-calculation model developed with ANN could be used to accurately predict layer modulus and interlayer conditions. In addition, the back-calculation model avoided the back-calculation errors by considering the interlayer condition, which was barely considered by former models reported in the published studies.     

Prediction of Pavement Performance through Neuro-Fuzzy Reasoning

Abstract:  Government agencies and consulting companies in charge of pavement management face the challenge of maintaining pavements in serviceable conditions throughout their life from the functional and structural standpoints. For this, the assessment and prediction of the pavement conditions are crucial. This study proposes a neuro-fuzzy model to predict the performance of flexible pavements using the parameters routinely collected by agencies to characterize the condition of an existing pavement. These parameters are generally obtained by performing falling weight deflectometer tests and monitoring the development of distresses on the pavement surface. The proposed hybrid model for predicting pavement performance was characterized by multilayer, feedforward neural networks that led the reasoning process of the IF-THEN fuzzy rules. The results of the neuro-fuzzy model were superior to those of the linear regression model in terms of accuracy in the approximation. The proposed neuro-fuzzy model showed good generalization capability, and the evaluation of the model performance produced satisfactory results, demonstrating the efficiency and potential of these new mathematical modeling techniques .     

Application of artificial neural networks to fracture analysis at the Äspö HRL, Sweden: fracture sets classification

This study investigates the potential of artificial neural networks (ANNs) to recognize, classify and predict patterns of different fracture sets in the top 450 m in crystalline rocks at the Äspö Hard Rock Laboratory (HRL), Southeastern Sweden. ANNs are computer systems composed of a number of processing elements that are interconnected in a particular topology which is problem dependent. ANNs have the ability to learn from examples using different learning algorithms; these involve incremental adjustment of a set of parameters to minimize the error between the desired output and the actual network output. Six fracture-sets with particular ranges of strike and dip have been distinguished. A series of trials were carried out using backpropagation (BP) neural networks for supervised classification, and the BP networks recognized different fracture sets accurately. Self-organizing neural networks have been used for data clustering analysis with supervised learning algorithms; (competitive learning and learning vector quantization), and unsupervised learning algorithms; (self-organizing maps). The self-organizing networks adapted successfully to different fracture clusters (sets). A set of trials has been carried out to investigate the effect of changing the network's topologies on the performance of the BP networks. Using two hidden layers with tan-sigmoid and linear transfer functions was beneficial for the performance of BP classification. ANNs improved fracture sets classification that was based on Kamb contouring method with constraint on areas between fracture clusters.     

Modeling the pavement serviceability ratio of flexible highway pavements by artificial neural networks

The term “present serviceability” was adopted to represent the momentary ability of pavement to serve traffic, and the performance of the pavement was represented by its serviceability history in conjunction with its load application history. Serviceability was found to be influenced by longitudinal and transverse profile as well as the extent of cracking and patching. The amount of weight to assign to each element in the determination of the over-all serviceability is a matter of subjective opinion.In this study, artificial neural networks (ANN) is used in modeling the present serviceability index of the flexible pavements. Experimental data obtained from AASHTO include slope variance, rut depth, patches, cracking and longitudinal cracking. The developed ANN model has higher regression value than AASHO model. This approach can be easily and realistically performed to solve the problems which do not have a formulation or function about the solution.     

Prediction of elastic modulus of normal and high strength concrete by artificial neural networks

In the present paper, application of artificial neural networks (ANNs) to predict elastic modulus of both normal and high strength concrete is investigated. The paper aims to show a possible applicability of ANN to predict the elastic modulus of both high and normal strength concrete. An ANN model is built, trained and tested using the available test data gathered from the literature. The ANN model is found to predict elastic modulus of concrete well within the ranges of the input parameters considered. The average value of the experimental elastic modulus to the predicted elastic modulus ratio is found to be 1.00. The elastic modulus results predicted by ANN are also compared to those obtained using empirical results of the buildings codes and various models. These comparisons show that ANNs have strong potential as a feasible tool for predicting elastic modulus of both normal and high strength within the range of input parameters considered.     

Comparative analysis of artificial neural networks and evolutionary programming in roof deterioration modeling

Modeling of roof performance and deterioration has been done in the past years by means of regression and condition rating as part of the management of civil infrastructure. However, in recent years artificial neural networks (ANNs) have also been used to model the performance of civil infrastructure systems such as pavements. The evolutionary algorithm (EA) method is one the most current methods that can also be used to predict the performance of infrastructure systems. This paper presents the comparative analyses of ANNs and EA methods in predicting and modeling the performance of a roofing system.     

Prediction of engine performance and exhaust emissions with different proportions of ethanol–gasoline blends using artificial neural networks

The main purpose of this study is to experimentally investigate the use of ANNs (artificial neural networks) modelling to predict engine power, torque and exhaust emissions of a spark ignition engine which operates with gasoline and methanol blends. For the ANN modelling, the standard back-propagation algorithm was found to be the optimal choice for training the model. Afterwards, the performance of the ANN predictions was evaluated with the experimental results by comparing the predictions. Fuel type and engine speed have been used as the input layer, while engine torque, power, exhaust emissions, Tex and BSFC have also been used separately as the output layer. It was found that the ANN model is able to predict the engine performance, exhaust emissions, Tex and BSFC with a correlation coefficient of 0.9991887425, 0.9990868573, 0.9986749623, 0.9988624137, 0.9976761492, 0.9992943894 and 0.9978899033 for the Power, Torque, CO, CO₂, HC, Tex and BSFC for testing data, respectively.     

Daily suspended sediment concentration simulation using ANN and neuro-fuzzy models

In the present study, artificial neural networks (ANNs), neuro-fuzzy (NF), multi linear regression (MLR) and conventional sediment rating curve (SRC) models are considered for time series modeling of suspended sediment concentration (SSC) in rivers. As for the artificial intelligence systems, feed forward back propagation (FFBP) method and Sugeno inference system are used for ANNs and NF models, respectively. The models are trained using daily river discharge and SSC data belonging to Little Black River and Salt River gauging stations in the USA.Obtained results demonstrate that ANN and NF models are in good agreement with the observed SSC values; while they depict better results than MLR and SRC methods. For example, in Little Black River station, the determination coefficient is 0.697 for NF model, while it is 0.457, 0.257 and 0.225 for ANN, MLR and SRC models, respectively. The values of cumulative suspended sediment load estimated by ANN and NF models are closer to the observed data than the other models. In general, the results illustrate that NF model presents better performance in SSC prediction in compression to other models.     

Prediction of effective thermal conductivity of moist porous materials using artificial neural network approach

An artificial neural networks (ANNs) approach is presented for the prediction of effective thermal conductivity of porous systems filled with different liquids. ANN models are based on feedforward backpropagation network with training functions: Levenberg–Marquardt (LM), conjugate gradient with Fletcher–Reeves updates (CGF), one-step secant (OSS), conjugates gradient with Powell–Beale restarts (CGB), Broyden, Fletcher, Goldfrab and Shanno (BFGS) quasi-Newton (BFG), conjugates gradient with Polak–Ribiere updates (CGP). Training algorithm for neurons and hidden layers for different feedforward backpropagation networks at the uniform threshold function TANSIG-PURELIN are used and run for 1000 epochs. The complex structure encountered in moist porous materials, along with the differences in thermal conductivity of the constituents makes it difficult to predict the effective thermal conductivity accurately. For this reason, artificial neural networks (ANNs) have been utilized in this field. The resultant predictions of effective thermal conductivity (ETC) of moist porous materials by the different models of ANN agree well with the available experimental data.     

Burst detection using hydraulic data from water distribution systems with artificial neural networks

This paper presents research into the application of artificial neural networks (ANNs) for analysis of data from sensors measuring hydraulic parameters (flow and pressure) of the water flow in treated water distribution systems. Two neural architectures (static and time delay) are applied for time series pattern classification from the perspective of detecting leakage. Results are presented using data from an experimental site in a distribution system of a UK water company in which bursts were simulated by hydrant flushing. Field trials have shown how ANNs can be used effectively for a leakage detection task. Both static and time delay ANNs learned patterns of leaks/bursts. The time delay neural network showed improved performance over the static network. It is concluded that the effectiveness of an ANN in discovering relationships within the data is dependent upon two key factors: availability of sufficient exemplars and data quality.     

Discussion on “Predicting the shear strength of reinforced concrete beams using artificial neural networks” by M.Y. Mansour,M. Dicleli,J.Y. Lee,J. Zhang [Eng Struct 26 (2004) 781–799]

The paper reporting the study by the authors considers the use of artificial neural networks (ANNs) to predict the ultimate shear strengths of reinforced concrete (RC) beams with transverse reinforcements. Because of some paradoxes in the results, the proposed ANN model has almost no reliability. The aim of this discussion is to point out controversial points of the paper.     

基于神经网络的碳纤维布与混凝土极限粘结强度分析

碳纤维布与混凝土的极限粘结强度问题属于高度非线性问题,难以建立精确的数学表达式进行分析。对基于拉出试验的极限粘结强度数据进行分析,建立了人工神经网络,对极限粘结强度进行仿真预测。神经网络的建立考虑了碳纤维布的厚度、宽度、粘结长度、弹性模量、抗拉强度和混凝土试块抗压强度、抗拉强度、宽度这8个参数,运用了118组试验数据对网络进行训练,对15组数据进行了预测分析。将神经网络计算结果同4种经验公式计算结果进行比较,其精度明显高于其他4种模型。结果表明,运用人工神经网络对碳纤维布与混凝土的极限粘结强度进行预测是可行的。     

A neural-network-based classification scheme for sorting sources and ages of fecal contamination in water

Artificial neural networks (ANNs) were successfully applied to data observations from a small watershed consisting of commonly measured indicator bacteria, weather conditions, and turbidity to distinguish between human sewage and animal-impacted runoff, fresh runoff from aged, and agricultural land-use-associated fresh runoff from that of suburban land-use-associated-fresh runoff. The ANNs were applied in a cascading, or hierarchical scheme. ANN performance was measured in two ways: (1) training and (2) testing. An ANN was able to sort sewage from runoff with < 1% error. Turbidity was found to be relatively unimportant for sorting sewage from runoff, while gross measurements of gram-negative and gram-positive bacteria were required. Predictions clustered tightly around the known values. ANN classification of aged suburban runoff from fresh, and agricultural runoff from suburban was accomplished with > 90% accuracy.     

采用扩展的贝叶斯人工神经网络选择结构损伤监测模型

对采用规则的动态数据进行结构损伤监测时,模式识别是一个有效的方法,人工神经网络作为匹配模式特征的系统方式广泛应用于模式识别研究中。人工神经网络设计是影响模型识别性能和效率的最基本因素。由Lam等人提出的贝叶斯人工神经网络设计法则为单隐层前馈人工神经网络确定大量隐性神经单元提供了严格的数学手段。本文的第一个目标是对贝叶斯人工神经网络设计法则进行拓展,包括选择隐层中神经单元的传递函数。所提出的法则具有高效的特点,适用于实时人工神经网络设计。目前,许多人工神经网络设计技术需要在训练前已知人工神经网络模型的类型,因此,最基本的问题是自动选择优化的人工神经网络模型类型的技术。由于模型参数和Ritz向量一般用于描述模式的特征,本文的第二个目标是采用模式识别对结构损伤监测中这两个模式特征进行比较。为了清楚判断这两个特征,研究中采用了IASC-ASCE准则。研究结果显示:采用模型参数进行训练的人工神经网络性能略优于采用Ritz向量进行训练的人工神经网络性能。     

Multivariate adaptive regression (MARS) and hinged hyperplanes (HHP) for doweled pavement performance modeling

Standard neural networks in infrastructure performance modeling cannot handle discontinuities in the input training data set, and the performance can in some cases be an issue in the presence of higher frequency and higher order non linearity in pavement condition, traffic and other environmental data. This makes the traditional neural network more of a “black box” with limited physical explanation of the results. This paper is a comparative analysis between multivariate adaptive regression and hinged hyperplanes for doweled pavement performance modeling.     

Disinfection by-product formation following chlorination of drinking water: Artificial neural network models and changes in speciation with treatment

Artificial neural network (ANN) models were developed to predict disinfection by-product (DBP) formation during municipal drinking water treatment using the Information Collection Rule Treatment Studies database complied by the United States Environmental Protection Agency. The formation of trihalomethanes (THMs), haloacetic acids (HAAs), and total organic halide (TOX) upon chlorination of untreated water, and after conventional treatment, granular activated carbon treatment, and nanofiltration were quantified using ANNs. Highly accurate predictions of DBP concentrations were possible using physically meaningful water quality parameters as ANN inputs including dissolved organic carbon (DOC) concentration, ultraviolet absorbance at 254 nm and one cm path length (UV₂₅₄), bromide ion concentration (Br⁻), chlorine dose, chlorination pH, contact time, and reaction temperature. This highlights the ability of ANNs to closely capture the highly complex and non-linear relationships underlying DBP formation. Accurate simulations suggest the potential use of ANNs for process control and optimization, comparison of treatment alternatives for DBP control prior to piloting, and even to reduce the number of experiments to evaluate water quality variations when operating conditions are changed. Changes in THM and HAA speciation and bromine substitution patterns following treatment are also discussed.     

Neural networks for modelling ultimate pure bending of steel circular tubes

The behaviour of steel circular tubes under pure bending is complex and highly nonlinear. The literature has a number of solutions to predict the response of steel circular tubes under pure bending; however, most of these solutions are complicated and difficult to use in routine design practice. In this paper, the feasibility of using artificial neural networks (ANNs) for developing more accurate and simple-to-use models for predicting the ultimate pure bending of steel circular tubes is investigated. The data used to calibrate and validate the ANN models are obtained from the literature and comprise a series of 49 pure bending tests conducted on fabricated steel circular tubes and 55 tests carried out on cold-formed tubes. Multilayer feed-forward neural networks that are trained with the back-propagation algorithm are constructed using four design parameters (i.e. tube thickness, tube diameter, yield strength of steel and modulus of elasticity of steel) as network inputs and the ultimate pure bending as the only output. A sensitivity analysis is conducted on the ANN models to investigate the generalization ability (robustness) of the developed models, and predictions from the ANN models are compared with those obtained from most available codes and standards. To facilitate the use of the developed ANN models, they are translated into design equations suitable for spreadsheet programming or hand calculations. The results indicate that ANNs are capable of predicting the ultimate bending capacity of steel circular tubes with a high degree of accuracy, and outperform most available codes and standards.