【摘要】 在锅炉燃烧效率优化问题的研究中,针对燃烧稳定性问题,提高燃烧检测准确度。利用锅炉中火焰燃烧图像的分析来检测锅炉内火焰燃烧稳定性,当燃烧受煤粉或其它杂质的影响火焰出现短时脉动时,传统的灰度方差的检测方法不能避免火焰脉动的影响,对锅炉内燃烧稳定性检测偏差大。为此提出了一种采用神经网络的火焰燃烧稳定性检测方法,选取与火焰稳定性直接相关的多个特征作为神经网络的输入向量,通过对样本的多次训练能够去除杂质燃烧引起的微小脉动的影响,克服了传统方法检测准确度不高的问题。实验证明,改进方法能够有效避免杂质对燃烧的影响,准确对锅炉内的燃烧稳定性作出检测,并取得了满意的结果。更多还原

【Abstract】 Research flame stability to improving the detection accuracy of the combustion.This paper presented a neural network-based detection methods of flame stability.The mothod selects characteristics related directly to flame stability as neural network input vector,through training samples to remove the impurities caused by the burning effects of small pulse.Experiments show that this method can effectively avoid the effects of impurities on the combustion,accurately detect the combustion stability,and achieved satisfactory results.更多还原