Applicability of machine learning to a crack model in concrete bridges

Abstract

The growing demand for a more efficient maintenance of concrete bridges requires a model that tracks the deterioration of each bridge based on inspection data. Although it has been expected that machine learning could be applied to this problem, inspection data sparsely distributed over time are not suitable for machine learning in contrast to the continuous big data usually targeted. This study applies machine learning to a regression model of crack formation and propagation using inspection data to confirm the applicability. It includes the selection of the optimal algorithm, development of the model based on a novel methodology, and factor analysis using the model. Accordingly, the model was constructed by Gaussian process regression and it could appropriately extract the differences in the progress of crack damage due to multiple influential factors. The results demonstrate the excellent applicability of machine learning even to sparse data.

Undersampling Strategy for Machine-learned Deterioration Regression Model in Concrete Bridges

“…It is noted that feature selection, that is, the selection of explanatory variables, is conducted prior to the undersampling. The details of this are provided in our previous paper (Okazaki et al 2020) and are not included here. The explanatory variables by feature selection are listed in Table 2.…”

“…Many machine learning algorithms have been proposed and are freely available. The present study adopts the Gaussian process regression (GPR) (Rasmussen 2006) as the learning algorithm, as it can fit the trend even for a local characteristic of the target data based on a performance comparison of the models constructed with different common algorithms, namely, multiple linear regression, support vector machine, decision trees, artificial neural network, and GPR (Okazaki et al 2020).…”

“…In our previous study, crack damage in the main girders of concrete bridges was investigated (Okazaki et al 2020). The applicability of machine learning to the regression problem of crack damage formation and propagation using inspection data was validated by comparing the performance of different learning algorithms and tuning hyper-parameters.…”

Crack Detection from a Concrete Surface Image Based on Semantic Segmentation Using Deep Learning

“…The artificial intelligence technique including deep learning has gained popularity in recent years owing to its high performance to deal with the large amount of data and was used for civil engineering applications, for example, by Rafiei et al (2017), Adeli (2018a, 2018b), Chun et al (2019Chun et al ( , 2020cChun et al ( , 2020d, Moon et al (2019Moon et al ( , 2020 and Okazaki et al (2020). It is also used for the detection of the crack.…”

Development of a Machine Learning-Based Damage Identification Method Using Multi-Point Simultaneous Acceleration Measurement Results

“…Then, four algorithms—auto-associative neural network, factor analysis, Mahalanobis distance, and singular value decomposition—were applied and their accuracy was compared. Other studies include research using 1D Convolutional Neural Networks [ 16 ], support vector machines [ 17 , 18 , 19 , 20 ], Gaussian process regression [ 21 ], and Genetic algorithms [ 18 , 22 ]. Although these studies propose machine learning methods based on interesting experimental data, they basically remain at the level of damage detection, because it is very difficult to prepare test specimens with various damage geometries in the order of 10 2 or more in experiments.…”