作 者: (魏祎璇);
机构地区: 宁波诺丁汉大学,315100
出 处: 《暖通空调》 2017年第9期27-32,54,共7页
摘 要: 简要介绍了4种常用的数据挖掘技术,结合建筑负荷分类和预测相关研究进行了评估比较。使用K均值聚类算法对上海某居民小区的用电数据进行了分析,结果表明,利用K均值算法的聚类分析和统计学方法结合所挖掘出的数据与实际情况表现出高度吻合性,并且比逻辑推理分析的结果更加精确可靠。 Briefly presents four common data mining techniques and evaluates them based on related studies of building load classification and prediction. Analyses the electricity data of a residential community in Shanghai by means of K mean clustering algorithm, and the result shows a high degree of consistency between the clustering analysis and statistical method using the K mean algorithm combining with the data mined and the actual situation, and more accurate and reliable than the results of logical reasoning.