Expert System Development for Acoustic Analysis in Concrete Harbor NDT

Port and Maritime Organization of Iran (PMO), in connection with a research project at Information Technology Mechatronic Offshore research and development cooperative society (ITMO), has added another dimension to its subsea inspection activities by introducing new methods of NDT and expert system for condition monitoring and assessment of concrete structures. ITOM provided a wide range of special and advanced techniques for most aspects of subsea and underwater. The repair of concrete structures under water presents many complex problems. The harsh environmental conditions and specific problems associated with working underwater or in the splash zone area causes many differences. Proper evaluation of the present condition of the structure is the first essential step for designing long-term repairs. To be most effective, evaluation of the existing structure requires historical information on the structure and its environment, including any changes made to the structure over time, and the records of periodic on-site inspections or repairs. Reduction of the human experts involvement in the diagnosis process has gradually taken place due to the recent developments in the modern Artificial Intelligence (AI) tools. AI is a research field between psychology, cognitive science and computer science with the overal goal to improve reasoning capabilities of computers. Artificial Neural Networks (ANNs), fuzzy and adaptive fuzzy systems, and expert systems are good candidates for the automation of the diagnostic procedures and e-maintenance application (Filippetti, et al., 1992 & Hedayati 2009). It is often necessary to test concrete structures after the concrete has hardened to determine whether the structure is suitable for its designed use. Ideally such testing should be done without damaging the concrete. The tests available for testing concrete range from the completely non-destructive, where there is no damage to the concrete, through those where the concrete surface is slightly damaged, to partially destructive tests, such as core tests and pullout and pull off tests, where the surface has to be repaired after the test. The present work surveys the principles and a criterion of the diagnosis signal processing and introduces these achievements to an expert system technique. In this paper adoption of a new sensor is discussed and experimental results are presented for an expert system application, based on the concept of spectrum and cepstrum analysis of detected signals and the method of measuring defected parts of subsea concrete without disturbing their structures for a