Predictive modelling and non-destructive testing (NDT) of oil and gas storage tank

E Odiowei, S Thomas, P Nzerem, Y Babayev and V Anye
Department of Electrical and Electronics/Computer Engineering, Nile University of Nigeria

Oil and gas storage tanks play a vital role in the conservation and storage of petroleum product. A high premium is placed by multinationals on the maintenance and repair of oil and gas storage tanks. However, oil and gas storage tanks are subject to deterioration and degradation over time. Evaluation and monitoring of the conditions of oil and gas storage tanks will help to optimise their operation and reduce failures to an acceptable safety limit. This research studies the electrical behaviour of ultrasonic guided wave propagation in steel metal with defects. The ultrasound wave properties and characteristics are used to develop an artificial neural network (ANN) model that will predict the condition of oil and gas storage tank. Input parameters for the network will be derived from ultrasonic non-destructive testing (NDT) methods. The model is expected to help operators to assess and predict the condition of existing oil and gas storage tanks and hence prioritise the planning of their inspection and rehabilitation.

Keywords: ultrasonic, artificial neural network, EMAT, finite element analysis, modelling.