Reconstruction of flame permittivity distribution based on ECT and L<sub>p</sub> regularized D-bar method
Abstract The measurement of flame permittivity is significant in obtaining the combustion state. A method of L p regularized D-bar is proposed in this article, which is used to reconstruct flame permittivity distribution by electrical capacitance tomography (ECT). Firstly, the distribution characteristic of flame permittivity is analyzed using the electric probe method. The simulation model of ECT flame measurement is set up based on the result of flame permittivity distribution. Then, the relationship model between the truncation radius of the D-bar algorithm and noise level is fitted based on the simulation reconstruction results of flame permittivity. The truncation radius of the D-bar algorithm for flame permittivity reconstruction is obtained by the relationship model. The L p regularization is introduced into the scattering transformation solving of the D-bar algorithm, and the simulation results of flame permittivity reconstruction under different p -values are compared. In the experiment, the reconstruction of flame permittivity by the D-bar algorithm with different truncation radii is compared. The experimental results verify the truncation radius strategy based on noise level. Moreover, compared with L 1 and L 2 regularization methods, L p regularization combined with the D-bar algorithm is more accurate in the experiment results of flame permittivity reconstruction.
Authors
- Shuo Jin
- Kexin Zhang (ORCID: https://orcid.org/0000-0001-6610-0045)
- Chao Wang (ORCID: https://orcid.org/0000-0002-4362-947X)
- Qi Guo (ORCID: https://orcid.org/0000-0002-3607-4808)
Institutions
- Tianjin University (CN)
- Tianjin Special Equipment Supervision and Inspection Technology Research Institute (CN)
Publication Details
- Journal
- Measurement Science and Technology
- Published
- 2024-10-22
- DOI
- https://doi.org/10.1088/1361-6501/ad89ea
- Citations
- 1
- Primary Topic
- High voltage insulation and dielectric phenomena
- Type
- article
- Field-Weighted Citation Impact
- 0.12
Funders
- National Natural Science Foundation of China
- Tianjin Research Innovation Project for Postgraduate Students