An Acoustic Emission Technique for Monitoring the Liquefied Natural Gas Cargo Tank

Increase in the market of supersized LNG (Liquefied Natural Gas) vessel, with doubled walled cargo tanks, has led to concerns regarding their safe operations. If both the primary and secondary wall of the cargo tank fails simultaneously, the hull of the vessel can be exposed to the LNG’s. This has the potential to cause brittle failure of the hull structure. This research presents a new Acoustic Emission (AE) technique that can be implemented to monitor the structural condition of the primary wall in the LNG cargo tank. The presented technique is able to provide information regarding critical damage so that appropriate maintenance can be carried out to avoid catastrophic failure. Acoustic Emission (AE) is a passive Non-Destructive Testing (NDT) technique, employed to identify critical damage in structures before failure can occur. Currently, AE monitoring is carried out by calculating the features of the waveform received by the AE sensor. User defined settings (i.e. timing and threshold) in the AE data acquisition system significantly affects many traditional AE features such as count, energy, centroid frequency, rise-time and duration. In AE monitoring, AE features are strongly related to the damage sources. Therefore, AE features, calculated due to inaccurate user defined acquisition settings can result in inaccurately classified damage sources. The new AE technique presented in this study is based on an AE feature of the waveform, which is independent of some user defined parameter (i.e. timing and threshold) used in the AE data acquisition system, unlike many traditional AE features. The presented AE feature is referred to as AE entropy in this research and is a measure of randomness in the waveform calculated using quadratic Renyi’s entropy. The effectiveness of AE entropy is evaluated by comparing it and traditional AE features under ideal conditions for a range of varying acquisition settings. Unlike the traditional feature, the AE entropy showed no variance with the acquisition settings and was effective in identifying different waveform shapes. The AE entropy was validated through fatigue and tensile tests on coupon specimens of austenitic stainless steel (material of the primary wall). The result suggested that AE entropy is effective in identifying the critical damages in austenitic stainless steel, irrespective of the data acquisition settings. Since AE entropy reduces the human involvement with the data acquisition system and can identify damages, it has the potential to be implemented in the commercial AE data acquisition system.