Producing a POD curve with emulated signal response data

Tuomas Koskinen, Mikko Virkkunen, Suvi Papula, Teemu Sarikka, Jonne Haapalainen

Research output: Contribution to journalArticleScientificpeer-review

194 Downloads (Pure)


Ultrasonic testing is an important tool for assessing the structural integrity of pressurised components in nuclear power plants during in-service inspection. Consequently, the reliability of inspection is of high signi cance. In particular, it is important to determine the largest crack that could conceivably be missed during in-service inspection. This information is utilised in order to choose the most e ective method for di erent situations. Probability of detection (POD) curves are used to quantify the e ectiveness of the inspection. However, these POD curves require a lot of data points in order to provide reliable estimates of the lower limit performance. Traditionally, obtaining these curves has been relatively expensive and this is why di erent simulation tools have been used to reduce costs and the number of physical test-pieces needed. In the current study, a novel method based on limited test-pieces is explored to produce a POD curve from the measured data. The present data contains just three arti cial cracks made with thermal fatigue. This
is insu cient to produce a POD curve. However, the data was su cient to evaluate the novel approach and to gain preliminary estimates of the expected POD and the most important in uential parameters. In this study, the idea is to emulate the amplitude response from the measured crack in a way that represents an amplitude response from a certain crack size. This amplitude data is converted to a B-scan image from which the inspector will evaluate whether there is a crack or not. Then, a POD curve is generated from the achieved hit/miss data.
Original languageEnglish
Pages (from-to)42-48
Number of pages7
Issue number1
Publication statusPublished - 1 Jan 2018
MoE publication typeA1 Journal article-refereed


  • NDT reliability
  • Ultrasonic inspection

Fingerprint Dive into the research topics of 'Producing a POD curve with emulated signal response data'. Together they form a unique fingerprint.

Cite this