Computer modelling for the prediction of the probability of detection of ultrasonic corrosion mapping : NDT in oil & gas industries

The reliability of NDT inspection has traditionally been determined by experimental trials; for example PISC I-III, NORDTEST and RACH. Such trials provide practical measures of the capability of inspection equipment. However, they can be expensive, may have poor statistics and the results obtained are specific to the equipment used as well as the specimens and defects included in the trials. An alternative approach is the use of computer modelling which, if sufficiently realistic, can provide a cost-effective alternative to experimental trials, and can be applied to a broad range of inspection equipment, geometries and defect characteristics. This paper describes a computer model for the prediction of the probability of detection (POD) of the zero degree ultrasonic corrosion mapping technique, which is widely used in the oil and gas industry for the in-service detection and characterisation of corrosion in pipes and vessels. The model has a relatively simple physical/mathematical basis but is believed to be a sufficiently good approximation to take into account the main factors which affect the POD of this inspection technique. The results from the model are compared with those from available experimental POD trials. Results are also given from a sensitivity study to identify and quantify the main parameters which influence POD under different inspection conditions.