Neural networks and non-destructive test/evaluation methods [manuscript]

Abstract

With today's reports of deteriorating highways and infrastructure as well as increased litigation arising from structural failures and the construction process, there is an increasing desire to employ non-destructive testing and evaluation (NDTE) methods for analyzing structural concrete members as well as other construction materials in a noninvasive manner. A major part of NDTE techniques is defect characterization, which is a typical pattern classification problem. The current state of the art for solving this problem is the application of a human expert's knowledge and experience for interpreting NDTE data. Artificial neural networks (ANNs) have shown a propensity for solving the pattern classification problem in the areas of speech and vision recognition, as well as problems in system modeling and simulation. As a result of these successful ANN applications, this paper explores the possibility of using ANNs for the NDTE defect characterization problem. Part of the solution of defect characterization entails the capability to filter what would otherwise be considered noisy data. Therefore, an ANN architecture is proposed and tested via computer simulation for the purpose of discerning between cracks and other surface defects found in photographs of defective reinforced concrete sections