In industrial NDE it is increasingly common for data acquisition to be automated, driving a recent multiple increase in the availability of data. The data analysis remains a mostly manual task, performed by a skilled operator - a rather painstaking task given that most of the data contain no indications. Partial automation, using software to prioritise regions of interest, could simultaneously increase inspection reliability and decrease data analysis time, by optimising the use of the operator's time. The project output is a software system general enough to fit a wide array of NDE applications. Nonetheless, the work so far has focused on two specific examples of automatic NDE: the ultrasonic inspection of power station rotor bores and the ultrasonic immersion inspection of aerospace titanium turbine discs. The paper will outline the key stages of the data processing scheme and show recent results from the program.
展开▼