Feature Analysis of Generalized Load Patterns Considering Active Load Response to Real-Time Pricing

Abstract

In the future Smart Grid, it is necessary to study the generalized load (GL) patterns feature considering the impact of active load response to real-time pricing (RTP). There are two challenges in the study. Firstly, how to quantitatively calculate the impact of RTP on GL is the main difficulty of the research. Secondly, the conventional indexes cannot accurately reflect the evolution trend of GL pattern feature. To overcome the first challenge, this paper proposed a novel calculation method based on elasticity coefficient. Elasticity coefficient can effectively formulate the relationship between RTP fluctuation and load response. The GL data after demand response (DR) is generated based on the elasticity coefficient. For the second challenge, based on the generated GL data, the hierarchical clustering method is used to extract the typical GL patterns. Then, new indexes are added to measure the evolution trend of GL pattern feature, including the proportion of maximum load decrease (PMALD), the proportion of minimum load increase (PMILI), the proportion of peak-valley difference of load decrease (PPVDLD), the proportion of large peak-valley difference of load (PLPVDL), and the proportion of large hourly climbing load (PLHCL), etc. The feature of GL patterns is investigated by a case study using the data of ISO New England. This study will provide significant auxiliary information for load control, load forecasting, and electricity price setting of smart grid.

Keywords

• Active load
• clustering analysis
• demand response
• elasticity coefficient
• load pattern
• real-time pricing