基于FDA的居民用户空调用电行为分类分析方法

针对居民空调用电行为分类中存在事件型数据，导致分类分析耗时长、结果不准确等问题，提出一种基于函数型数据分析（FDA）模型的居民空调用电行为分类分析方法。该方法采用多重分形理论提取居民用电行为特征，使用函数型数据分析算法对居民空调用电行为进行聚类后获取居民空调用电行为类别，采用改进动态时间规整算法对居民空调用电行为实施分类处理，得到居民空调用电行为。根据非介入式设备采集到的实际居民用电行为信息检验该方法的有效性，实验结果表明:该方法可以较好地提取居民用电行为特征，且可有效提高用户空调用电行为分类精度以及缩短分类耗时，可充分描述居民空调开启情况以及消耗电量，具备较好的应用效果。

Classification Analysis Method of Residential Air Conditioning Electricity Consumption Behavior Based on Functional Data Analysis Model

Targeting the problems of long time consuming and inaccurate results due to the existence of event-based data in residential air conditioning electricity consumption behavior classification， this paper proposes a classification analysis method of the residential air conditioning electricity consumption behavior based on functional data analysis model. In this method， multifractal theory is used to extract the characteristics of the electricity consumption behavior， functional data analysis algorithm is used to cluster the electricity consumption behavior to obtain its behavior category， and improved dynamic time warping algorithm is used to classify the electricity consumption behavior to obtain the electricity consumption behavior. The effectiveness of this method is verified by the actual residential air conditioning electricity consumption data collected by non-intrusive equipment. The experimental results show that this method has good ability to extract the characteristics of the residential electricity behavior， and can effectively improve the classification accuracy of the residential air conditioning electricity consumption behavior and shorten the classification time consuming. It can fully describe the situation of the residential air conditioning and electricity consumption， and has a good application effect.