Simulation study of temperature field reconstruction based on acoustic tomography

Abstract: To improve the understanding of the combustion process and evaluating the combustion efficiency, temperature of flame should be monitored. The conventional temperature measurements methods, like thermocouple and thermometer, can only provide pointwise estimation from their invasive measurement. Vision-based monitoring techniques heavily rely on the optical access into the process, and it is only suitable for some specific applications which allow opening an observation window at the boundary. A study in this paper aims to reconstruct the 2D temperature field in the furnace using acoustic tomography. Compared with other methods, acoustic tomography gives the global estimation of the temperature field with a good accuracy. In this tomographic modality, acoustic transducer and receivers are placed around the temperature field to estimate sound speed within the flame region using time-of-flight method, from which the temperature field can be derived. In this paper, the entire process of obtaining temperature filed is introduced. A simulation study is conducted by coupling two physical fields, temperature and acoustic pressure fields based on COMSOL Multiphysics. As the reconstruction quality suffers from the lack of measurements, and the inverse problem is always ill-posed, the temperature field is modelled by a sum of radial basis functions and then a truncated singular value decomposition method is used to solve the inverse problem. The possible error sources are analysed. Three test scenarios demonstrated acoustic tomography is able to reconstruct temperature field with reasonable accuracy.

Biography: Bao Yong received his BEng Hons degree from Dalian University of Technology, China in 2014. He achieved his MSc degree with distinction from the University of Edinburgh in 2015. Currently, he is a PhD candidate at the student of the Agile Tomography Group at IDCOM, University of Edinburgh. His research is focused on measuring temperature field and flow velocity using acoustic travel-time tomography.