论文部分内容阅读
锅炉高温受热面运行状况是极其复杂的,与很多因素相关。为了找出诸多变量之间隐藏着的关系,尝试着将基于属性的模糊聚类方法应用到受热面超温分析过程中,该方法可在无任何专家经验的基础上,通过处理DCS存储的大量数据,建立模糊聚类动态图,进而可全面直观地了解各个变量之间的关系,并可实现属性约减,减少进行分析的决策因素,提高分析诊断效果。利用该方法对某电厂再热器超温问题进行实例分析,结果证明属性模糊聚类方法用于锅炉高温受热面的运行分析中是可行有效的,它可以建立锅炉各运行参数对类别的不确定性描述,准确客观地反映运行参数之间的关系,对研究分析锅炉高温受热面事故、指导现场运行调节以及目前正在兴起的电厂数据挖掘技术有重要的意义。
Operating condition of boiler heating surface is extremely complex, with many factors related. In order to find out the hidden relationship among many variables, this paper attempts to apply the fuzzy clustering method based on attribute to the over-temperature analysis of heated surface. This method can process large quantities of DCS stored without any expert’s experience Data to establish fuzzy clustering dynamic graph, and then can comprehensively and intuitively understand the relationship between the various variables, and can reduce the attributes, reduce the analysis of the decision-making factors to improve the diagnostic results. The method is used to analyze the overheating problem of a reheater in a power plant. The results show that the fuzzy clustering method is feasible and effective in the operation analysis of the high temperature heating surface of the boiler. It can establish the uncertainty of the boiler operating parameters Therefore, it is of great significance to study and analyze the boiler heating surface accident and guide the on-site operation regulation as well as the power plant data mining technology that is currently emerging.