Ultrasonic temperature distribution reconstruction for circular area based on Markov radial basis approximation and singular value decomposition

Abstract

Temperature distribution reconstruction is of critical importance for circular area, and an ultrasonic technique is investigated to meet this demand in this paper. Considering the particularity of circular area, algorithm based on Markov radial basis approximation and singular value decomposition is proposed, while ultrasonic transducers layout and division of measured area are properly designed. The reconstruction performance is validated via numerical experiments using different temperature distribution models, and is compared with algorithm based on least square method. To study the anti-interference, various noises are adding to the theoretical value of time-of-flight. Experiment results indicate that the proposed algorithm can reconstruct temperature distribution with higher accuracy and stronger anti-interference, while without the problem of algorithm based on least square method that its reconstructions will lose much temperature information near the edge of measured area.

Keywords: Reconstruction algorithm; Temperature distribution; Temperature measurement.