火电机组在运行过程中产生大量的历史数据，而目前所使用数据分析方法仅仅对这些历史数据进行简单的分类和统计，并不能对这些数据所隐含的规律进行挖掘。利用相关性分析对某电厂的实时数据进行研究，从大量的机组运行参数中筛选出对机组能耗影响较大的重要参数：负荷、循环水入口温度、主蒸汽温度、再热蒸汽温度、主蒸汽压力、循环水流量。然后，介绍了模糊C-均值聚类算法的相关理论及其应用，利用此方法对以上6个参数进行工况划分。实际应用结果表明，在对电厂大量实时进行数据聚类和合理工况划分过程中，模糊C-均值聚类算法起到一定作用，并且对优化运行和机组节能优化有重大的意义。%Thermal power unit produces a large number of historical data during the operation process, and the currently used methods for data analysis classify these historical data and carry out statistics in a rather simple way, which cannot reveal the hidden rules beneath these data. The correlation analysis is applied for the study of real-time data for a power plant. Some parameters, such as the load, circulating water entrance temperature, main steam temperature, reheat steam temperature, steam pressure, and circulating water flow, are selected and consid-ered as important ones who have great influence on the energy consumption of the unit. Then, the related theory of fuzzy C-mean clustering algorithm and its application are introduced, and by using this method, six parameters mentioned above are divided according to the working condition. The results obtained during practical application show that during the reasonable working condition division and data clustering process, fuzzy C- means clustering algorithm works and is of great significance to the optimization of the operation and energy saving of the group.
展开▼
