Predicting length of fatigue cracks by means of machine learning algorithms in the small-data regime

Abstract

In this paper several statistical learning algorithms are used to predict the maximal length of fatigue cracks based on a sample composed of 31 observations. The small-data regime is still a problem for many professionals, especially in the areas where failures occur rarely. The analyzed object is a high-pressure Nozzle of a heavy-duty gas turbine. Operating parameters of the engines are used for the regression analysis. The following algorithms are used in this work: multiple linear and polynomial regression, random forest, kernel-based methods, AdaBoost and extreme gradient boosting and artificial neural networks. A substantial part of the paper provides advice on the effective selection of features. The paper explains how to process the dataset in order to reduce uncertainty; thus, simplifying the analysis of the results. The proposed loss and cost functions are custom and promote solutions accurately predicting the longest cracks. The obtained results confirm that some of the algorithms can accurately predict maximal lengths of the fatigue cracks, even if the sample is small.

Identification of crashworthiness indicators of column energy absorbers with triggers in the form of cylindrical embossing on the lateral edges using artificial neural networks

“…There are many publications in which artificial intelligence has been used to predict specific parameters. In paper [8] the authors illustrated this using fatigue crack length as an example, while paper [3] applied the network to aircraft engine failure prediction. An interesting look at energy efficiency issues on another technical facility is shown in publication [22].…”

A comprehensive study on the application of firefly algorithm in prediction of energy dissipation on block ramps

“…In recent years, the successful applications of machine learning methods in modeling engineering problems have been reported in numerous studies [4,18,19,26,33]. Despite the empirical formulas are easy to use, the outcome of research conducted in recent decades, including studies conducted in hydraulic engineering, indicate that the machine learning methods are more accurate [14,16,22,29,39].…”