基于火焰动态纹理的电熔镁炉工况识别

电熔镁炉制备电熔镁砂的工艺过程中，会交替出现正常熔炼、加料和欠烧等不同工况，针对不同的工况需要采取相应的处理方式来保证生产过程的正常进行。目前，工况的识别主要依靠人工完成，这种方式存在工人劳动强度大、容易漏检误检等问题。本文依据不同工况下炉口火焰具有不同的动态可视化特征，提出一种基于动态纹理的工况识别技术。首先，通过建立炉口火焰的线性动态系统模型来刻画图像的动态特性，然后，设计了基于子空间角度的核函数来度量火焰动态模型相似度。对比实验表明本文设计的基于子空间角度的工况分类器具有更好的分类精度及鲁棒性。

Conditions recognition of fused magnesia furnace based on flame dynamic texture

In the process of preparing fused magnesia in fused magnesium furnace, different working conditions such as Smelting condition, Feeding condition and Semi-fused condition alternately occur, and corresponding treatment methods are needed for these conditions to ensure the normal production process. At present, the identification of working conditions mainly relies on workers. This way has some problems such as high labor intensity of workers and misdetection. According to different dynamic visualization characteristics of furnace flame under different working conditions, a dynamic condition recognition technology based on dynamic texture is proposed. Firstly, the dynamic characteristics of the image are portrayed by establishing a linear dynamic system model of the furnace flame. Secondly, a kernel function based on subspace angle is designed to measure the similarity of flame dynamic model. The contrast experiments show that working condition classifier based on the subspace angle designed in this paper has better classification accuracy and robustness.