论文部分内容阅读
锅炉各监控参数基准值的确定是分析锅炉运行能耗偏差的基础。该文充分利用锅炉运行数据的关联特性,提出了一种基于模糊C-均值聚类算法实现多参量同步聚类以确定锅炉监控参数基准值的方法。该方法可以在实际运行数据中同步挖掘出某典型负荷邻域区间对应的排烟氧量、排烟温度和飞灰含碳量等监控参数基准值,从而达到改善锅炉运行性能的目标。在多参量同步聚类算法中,利用有效性函数优化模糊聚类数,提出运行模式支持度的相关概念及其样本支持判定的规则,并对类中心点处较小ε区域内样本进行无偏估计。实例分析结果表明:该方法能够在兼顾参数之间耦合关系的基础上,得到高效工况下对应的各基准值样本点,并建立相应的基准值模型。
Boiler monitoring parameters to determine the baseline value is the basis for analysis of boiler operation energy consumption deviation. This paper makes full use of the correlation characteristics of boiler operation data, and proposes a method based on fuzzy C-means clustering algorithm to realize multi-parameter synchronous clustering to determine the boiler monitoring parameter reference value. The method can synchronously mine the reference values of the monitoring parameters such as the exhaust smoke oxygen content, the exhaust gas temperature and the fly ash carbon content in the neighborhood of a typical load in the actual operation data so as to achieve the objective of improving the boiler operation performance. In the multi-parameter synchronous clustering algorithm, the fuzzy clustering number is optimized by the validity function, and the related concepts of operation mode support degree and the rules of sample support decision are proposed. The samples in the smaller ε region at the class centroid are unbiased estimate. The results of the example analysis show that this method can obtain the corresponding sample points of reference value under the high efficiency condition and the corresponding reference value model based on the coupling relationship between the parameters.