Condition of the pipelines is extremely important in many industries producing gas, fertilizers, chemicals etc. They often transport hazardous chemicals, gases that can endanger life and property in case of any mishap. Therefore, the detection and characterization of defects is very critical for pipeline integrity assessment. In this work, we look to classify and characterize corrosion pits. A numerical 2D FEM model is developed to simulate the two fundamental wave modes i.e. the A0 and S0 mode of an ultrasonic guided wave. Damage models are established using both axial and circumferential guided waves. The corrosion pits are modelled as localized hemispherical cavities of certain depth and length. The simulated wave is then used to characterize the effect of pit depth and lengths for both axial and circumferential corrosion pits. Also, the interaction of the guided wave with an interacting colony of pits and cracks is also studied. A large simulation based dataset is collected which consists of responses from healthy and faulty pipelines, which is then used to train a dense neural network to classify responses accordingly from healthy and faulty pipes. On the responses classified as faulty, a Cross Sampling Entropy Analysis is conducted to define a related Damage Index (DI). Using this damage index, the corrosion pits are characterized based on their defect geometries. This paper aims to leverage the advantages of both axial and circumferential guided waves, along with the power of Neural Networks in analysing the simulation data more efficiently and accurately and thereby automating the defect classification and characterization process.    


Characterization and Classification of Corrosion Pits using Cross Sampling Entropy Analysis and Dense Neural Network

No proceedings papers in 2020. Video only.

QNDE2020 - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.