Identification of the state-space dynamics of oil flames through computer vision and modal techniques

作者：

Highlights：

• This paper presents a new approach to estimate unstable flame dynamics.

• Computer vision processed images from CCD cameras provide a discriminant characteristic vector.

• A four-degree-of-freedom second order model is identified through Operational Modal Analysis.

• Unstable and stable flame models are cross-validated through spectral analysis.

• The method is capable of predicting flame extinction.

摘要

•This paper presents a new approach to estimate unstable flame dynamics.•Computer vision processed images from CCD cameras provide a discriminant characteristic vector.•A four-degree-of-freedom second order model is identified through Operational Modal Analysis.•Unstable and stable flame models are cross-validated through spectral analysis.•The method is capable of predicting flame extinction.

论文关键词：Flame instabilization characterizing,Operational Modal Analysis,Discriminant characteristic vectors,Artificial intelligence image processing

论文评审过程：Available online 4 November 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.10.030