大型集中供热系统热力站耗热量聚类分析

Cluster Analysis of Heat Consumption in Thermal Stations of Large-scale Central Heating System

合理掌握集中供热系统热力站耗热量现状,对指导集中供热系统节能运行、促进节能减排具有重要意义。对集中供热系统热力站耗热量及影响因素数据进行处理和转化,剔除数据中的异常值及离群值,并将处理后的文本型数据转化成数值型;运用皮尔森相关系数确定了影响热力站耗热量的主要因素为建筑保温性能和二次侧供回水平均温度;将耗热量及主要影响因素作为输入参数,运用SPSS软件分别进行层次聚类算法、划分聚类算法分析,同时利用Python语言进行密度聚类算法分析,通过3种聚类分析结果对比,最终采用K-means聚类分析算法建立耗热量的最佳聚类模型,得出各类热力站的耗热量及影响因素的聚类中心值,为进一步集中供热系统热力站节能潜力分析提供理论依据。

A reasonable grasp of the current status of heat consumption in the thermal power station of the central heating system is of great significance for guiding the energy-saving operation of the central heating system and promoting energy saving and emission reduction.The thesis first processes and transforms the data of heat consumption and influencing factors of the thermal power station of the central heating system,eliminates outliers in the data,and converts the processed text data into numerical data;uses Pearson correlation coefficient to determine.The main factors that affect the heat consumption of the thermal power station are the building insulation performance and the average temperature of the secondary side supply and return water;the heat consumption and the main influencing factors are used as input parameters,and the SPSS software is used to analyze the hierarchical clustering algorithm and partition clustering algorithm.Meanwhile,the Python language is used to analyze the density clustering algorithm.Through the comparison of the three clustering analysis results,the K-means clustering analysis algorithm is finally used to establish the best clustering model for heat consumption,and the heat consumption and influence of various thermal stations are obtained.The clustering center value of the factors provides a theoretical basis for further analysis of the energy-saving potential of the thermal station of the central heating system.