基于燃烧特征量和模糊C均值聚类的燃烧诊断

Combustion diagnosis based on combustion feature and fuzzy C-means clustering

以某电厂350 MW燃油锅炉燃烧数据为基础,提出了一种基于燃烧特征量和模糊C均值聚类的燃烧诊断方法.采用时、频域统计分析和信息熵分析技术提取锅炉火检信号中反映燃烧状况的统计特征量和信息熵特征量,再利用模糊C均值聚类算法对这些燃烧特征量进行聚类分析,得到的聚类中心可以作为燃烧状态判别的标准模式,通过计算待诊断样本对标准模式的隶属度实现燃烧诊断.在燃烧诊断的基础上,根据模糊隶属度的思想提出了一种模糊燃烧指数,可以实现对燃烧状态的定量监测.研究表明,该方法能够对燃烧状态进行有效监测和定量表征,为进一步实现燃烧调整和控制提供依据.

On the basis of the combustion data of a 350 MW oil fired boiler in a certain power plant, a method for combustion diagnosis based on combustion feature and FCM （fuzzy C-means） cluste- ring is proposed. Feature extraction using statistical analysis and information entropy is employed on the signal of flame detector to acquire the statistical and information entropy features which can re- flect the combustion state. Then FCM clustering algorithm is used to cluster the combustion features, and the clustering centers are obtained to be the standard modes for combustion identification. Com- bustion diagnosis is realized by calculating the membership between the unknown samples and the standard modes of combustion. Furthermore, a fuzzy combustion index is presented to quantifica- tionally monitor the combustion state according to the theory of fuzzy membership. The results show that the method can be used to monitor and evaluate the combustion state effectively, providing foun- dation for regulate and control of combustion.