Abstract
Titanium (Ti) is an ideal structural material, whose use is gradually emerging in civil engineering. Regular defect evaluation is indispensable during the long-term use of Ti-sheets, which can be performed effectively using eddy current (EC) imaging, a method of visualizing defects that is convenient for inspectors. However, as EC scan images contain abundant information and have discrepancies in terms of their quality, this makes it difficult to extract effective features, thus affecting the evaluation results. In this paper, we propose a method that combines EC imaging technology with a quantitative evaluation method for Ti-sheet defects based on Deep Belief Network (DBN) and Least Squares Support Vector Machine (LSSVM). A multilayer DBN is constructed to extract the effective features from EC scan images for Ti-sheet defects. Based on the extracted feature vectors, a multi-objective regression model of defect dimensions is established using the LSSVM algorithm. Then, the dimensions of Ti-sheet defect such as length, diameter, and depth were quantitatively evaluated by the effective features and the efficient regression model. The experimental results have shown that the evaluation errors for the defect lengths and depths tested were less than 3% and 5% respectively, confirming the validity of the proposed method.
Highlights
Titanium (Ti) is an ideal structural material with excellent all-round properties, such as low density, high specific strength, and excellent corrosion resistance (Cui et al, 2011)
Considering the high information content and discrepancy in quality of eddy current (EC) scan images, we propose a method for the quantitative evaluation of Ti sheet defects based on the deep belief network (DBN) and least squares support vector machine (LSSVM) combined with EC scan imaging
In terms of the regression algorithm used in these methods, the LSSVM is much better than multiple linear regression (MLR), and the average error is one order of magnitude smaller, which indicates that the LSSVM is more suitable for dealing with the nonlinear relationship between high-dimensional features and defect dimensions
Summary
Titanium (Ti) is an ideal structural material with excellent all-round properties, such as low density, high specific strength, and excellent corrosion resistance (Cui et al, 2011). The dimensions of a defect have been estimated by visual judgment of ECT detection signals by an inspector This method usually requires highly trained personnel, and the results are usually influenced by the subjectivity of the inspector (Fan et al, 2016). In this article, we propose a method that combines the EC scan imaging technology with a quantitative evaluation method for Ti sheet defect evaluation based on DBNs and least squares support vector machine (LSSVM). Based on the extracted feature vectors, a multi-objective regression model of defect dimensions was established using the LSSVM algorithm This combination of effective features and an efficient regression model was used to perform the quantitative evaluation of Ti sheet defects.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have