Modeling asphalt pavement overlay transverse cracks using the genetic operation tree and Levenberg–Marquardt Method

Hsie, Machine; Ho, Yueh-Feng; Lin, Chih-tsang; Yeh, I-Cheng

doi:10.1016/j.eswa.2011.10.005

Cited by 9 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Levenberg–Marquardt (Levenberg 1944; Marquardt 1963) algorithm is a blend of gradient descent and Gauss–Newton iterations, aimed to solve nonlinear least square problems. The method works as a steepest descent method when the parameters are far away from their optimal values and acts like a Gauss–Newton method when the parameters are closer to their optimal values (Hsie et al 2012). The parameter set is updated iteratively by adding negative of the scaled gradient at each step, with the error between measurements written as:

f_{i} (p_{1}, p_{2}, \dots p_{n}) = h_{normalobs} – h_{normalsim},

where, f i is the error estimated based on observed ( h obs ) and simulated ( h sim ) measurements.…”

Section: Materials and Methodologymentioning

confidence: 99%

Performance of Gradient and Gradient‐Free Optimizers in Transient Hydraulic Tomography

Chintala,

Kambhammettu,

Harmya

2023

Groundwater

View full text Add to dashboard Cite

Sub‐surface characterization in fractured aquifers is challenging due to the co‐existence of contrasting materials namely matrix and fractures. Transient hydraulic tomography (THT) is proved to be an efficient and robust technique to estimate hydraulic (Km, Kf) and storage (Sm, Sf) properties in such complex hydrogeologic settings. However, performance of THT is governed by data quality and optimization technique used in inversion. We assessed the performance of gradient and gradient‐free optimizers with THT inversion. Laboratory experiments were performed on a two‐dimensional, granite rock (80 cm x 45 cm x 5 cm) with known fracture pattern. Cross‐hole pumping experiments were conducted at 10 ports (located on fractures), and time‐drawdown responses were monitored at 25 ports (located on matrix and fractures). Pumping ports were ranked based on weighted signal‐to‐noise ratio (SNR) computed at each observation port. Noise‐free, good quality (SNR > 100) datasets were inverted using Levenberg Marquardt: LM (gradient) and Nelder Mead: NM (gradient‐free) methods. All simulations were performed using a coupled simulation‐optimization model. Performance of the two optimizers is evaluated by comparing model predictions with observations made at two validation ports that were not used in simulation. Both LM and NM algorithms have broadly captured the preferential flow paths (fracture network) via K and S tomograms, however LM has outperformed NM during validation () Our results conclude that, while method of optimization has a trivial effect on model predictions, exclusion of low quality (SNR ≤ 100) datasets can significantly improve the model performance.This article is protected by copyright. All rights reserved.

show abstract

f_{i} (p_{1}, p_{2}, \dots p_{n}) = h_{normalobs} – h_{normalsim},

where, f i is the error estimated based on observed ( h obs ) and simulated ( h sim ) measurements.…”

Section: Materials and Methodologymentioning

confidence: 99%

Performance of Gradient and Gradient‐Free Optimizers in Transient Hydraulic Tomography

Chintala,

Kambhammettu,

Harmya

2023

Groundwater

View full text Add to dashboard Cite

show abstract

“…Aminian et al 18 presented a new empirical model to estimate the base shear of plane steel structures subjected to earthquake load using a hybrid method integrating genetic programming GP and simulated annealing SA , called GP/SA. Hsie et al 19 proposed a novel approach, called "LMGOT," that integrates two optimization techniques: the Levenberg Marquardt LM Method and the genetic operation tree GOT . The GOT borrows the concept from the genetic algorithm, a famous algorithm for solving discrete optimization problems, to generate operation trees OTs , which represent the structures of the formulas.…”

Section: Genetic Programmingmentioning

confidence: 99%

Artificial Intelligence in Civil Engineering

Chen

Zheng

2012

Mathematical Problems in Engineering

163

View full text Add to dashboard Cite

Artificial intelligence is a branch of computer science, involved in the research, design, and application of intelligent computer. Traditional methods for modeling and optimizing complex structure systems require huge amounts of computing resources, and artificial-intelligence-based solutions can often provide valuable alternatives for efficiently solving problems in the civil engineering. This paper summarizes recently developed methods and theories in the developing direction for applications of artificial intelligence in civil engineering, including evolutionary computation, neural networks, fuzzy systems, expert system, reasoning, classification, and learning, as well as others like chaos theory, cuckoo search, firefly algorithm, knowledge-based engineering, and simulated annealing. The main research trends are also pointed out in the end. The paper provides an overview of the advances of artificial intelligence applied in civil engineering. of different forms in civil engineering. Bassuoni and Nehdi 6 developed neuro-fuzzy based prediction of the durability of self-consolidating concrete to various sodium sulfate exposure regimes. Prasad et al. 7 presented an artificial neural network ANN to predict a 28day compressive strength of a normal and high strength self-compacting concrete SCC and high performance concrete HPC with high volume fly ash. Lee et al. 8 used an artificial intelligence technique of back-propagation neural networks to assess the slope failure. The numerical results demonstrate the effectiveness of artificial neural networks in the evaluation of slope failure potential. Shaheen et al. 9 presented a proposed methodology for extracting the information from experts to develop the fuzzy expert system rules, and a tunneling case study was used to illustrate the features of the integrated system. Das et al. 10 described two artificial intelligence techniques for prediction of maximum dry density MDD and unconfined compressive strength UCS of cement stabilized soil. Forcael et al. 11 presented the results of a study that incorporates computer simulations in teaching linear scheduling concepts and techniques, in a civil engineering course "Construction Planning and Scheduling." To assess the effect of incorporating computer simulation in teaching linear scheduling, the students' evaluations and answers to the questionnaire were statistically compared. Krcaronemen and Kouba 12 proposed a methodology for designing ontologybacked software applications that make the ontology possible to evolve while being exploited by one or more applications at the same time. The methodology relies on a contract between the ontology and the application that is formally expressed in terms of integrity constraints. In addition, a reference Java implementation of the methodology and the proof-of-concept application in the civil engineering domain was introduced.Due to a lot of uncertain factors, complicated influence factors in civil engineering, each project has its individual character and generality; function of expe...

show abstract

“…Artificial neural network models are considered by many researchers as "black-boxes" (Attoh-Okine et al 2009;Hsie et al 2012;Benítez et al 1997;Prechelt 1998). With a complex network structure, it is difficult to explicitly describe the learned relationship between the input and the output variables.…”

Section: Forward Calculationsmentioning

confidence: 99%

Development of an artificial neural network model to predict subgrade resilient modulus from continuous deflection testing

et al. 2017

View full text Add to dashboard Cite

Abstract:The subgrade resilient modulus is an important parameter in pavement analysis and design. However, available non-destructive testing devices such as the falling weight deflectometer (FWD) have limitations that prevent their widespread use at the network level. This study describes the development of a model that utilizes the rolling wheel deflectometer (RWD) measurements to predict the subgrade resilient modulus at the network level for flexible pavements. Measurements of RWD and FWD obtained from a testing program conducted in Louisiana were used to train an artificial neural network (ANN) based model. The ANN model was validated using data from a testing program independently conducted in Minnesota. The ANN model showed acceptable accuracy in both the development and validation phases with coefficients of determination of 0.73 and 0.72, respectively. Furthermore, the limits of agreement methodology showed that 95% of the differences between the subgrade resilient modulus calculated based on FWD and RWD measurements will not exceed the range of ±21 MPa (±3 ksi).Key words: subgrade resilient modulus, pavement deflection, rolling wheel deflectometer (RWD), network level. Mots-clés : module de résilience du sol de fondation, déflexion du revêtement, déflectomètre de roulement (« RWD »), niveau du réseau.

show abstract

Modeling asphalt pavement overlay transverse cracks using the genetic operation tree and Levenberg–Marquardt Method

Cited by 9 publications

References 14 publications

Performance of Gradient and Gradient‐Free Optimizers in Transient Hydraulic Tomography

Performance of Gradient and Gradient‐Free Optimizers in Transient Hydraulic Tomography

Artificial Intelligence in Civil Engineering

Development of an artificial neural network model to predict subgrade resilient modulus from continuous deflection testing

Contact Info

Product

Resources

About