Soft Computing Approach to Crack Detection and FPGA Implementation

For many years, there has been considerable interest in applications of nondestructive testing (NDT) techniques of real-time crack detection. Various authors have suggested different approaches to find a solution to solve the problem of real-time crack detection (Meitzler et al., 2008; Zheng et al., 2008; Berthelot et al.,2000; Tong and Wu, 2001). This paper suggests a soft computing approach using fuzzy logic as an effective technique to solve the problem of real-time crack detection. It is also desirable to develop an integrated-circuit microprocessor chip, which can be used for the detection of cracks. The objectives of this paper are to report on efforts to develop an automated procedure for crack detection and to devise a technique so that the proposed technique can be easily implemented on a microprocessor chip. The developed fuzzy inference system is supported by the rule base which defines the behavior of a crack detection system. A very large-scale integration (VLSI) circuit for crack detection is developed on the basis of the fuzzy logic model and using a hardware description language (HDL). Field-programmable gate array (FPGA) implementation and circuit testing were also performed. KEYWORDS: fuzzy logic, nondestructive testing techniques, crack detection, neuro-fuzzy logic, Mamdani technique, FPGA.