Passive acoustic leak detection for sodium cooled fast reactors using hidden Markov models

• Acoustic leak detection for steam generators of sodium fast reactors have been an active research topic since the early 1970s and several methods have been tested over the years. Inspired by its success in the field of automatic speech recognition, we here apply hidden Markov models (HMM) in combination with Gaussian mixture models (GMM) to the problem. To achieve this, we propose a new feature calculation scheme, based on the temporal evolution of the power spectral density (PSD) of the signal. Using acoustic signals recorded during steam/water injection experiments done at the Indira Gandhi Centre for Atomic Research (IGCAR), the proposed method is tested. We perform parametric studies on the HMM+GMM model size and demonstrate that the proposed method a) performs well without a priori knowledge of injection noise, b) can incorporate several noise models and c) has an output distribution that simplifies false alarm rate control.

Ämnesord

NATURVETENSKAP  -- Fysik (hsv//swe)

NATURAL SCIENCES  -- Physical Sciences (hsv//eng)

Nyckelord

Acoustic noise

Fast reactors

Leak detection

Markov processes

Nuclear instrumentation

Power spectral density

Signal detection

Sodium

Spectral density

Speech recognition

Steam generators

Trellis codes

Acoustic leak detections

Automatic speech recognition

Calculation scheme

Gaussian Mixture Model

Output distribution

Power spectral densities (PSD)

Sodium cooled fast reactor

Sodium fast reactors

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)