Demand response ability evaluation based on seasonal and trend decomposition using LOESS and S–G filtering algorithms

• DOI: 10.1016/j.egyr.2022.02.139
• Publication Date: 2022-03-02T20:30:39Z
• Authors: Di Wu; Yunchu Wang; Lei Li; Pengfei Lu; Shengyuan Liu; Chang Dai; Yizhou Pan; Zhi Zhang; Zhenzhi Lin; Li Yang
• Keywords: Demand Response; Dynamic demand

With the more and frequent short-term peak loads appear in power systems recent years, it is challenging to keep load frequency steady for systems. Demand response launched shave fill valley. Daily regularity of electricity consumption behavior special transformer users are studied this paper. ability evaluation method based on seasonal trend decomposition using locally weighted regression (LOESS) Savitzky–Golay (S–G) filtering algorithms proposed, which can optimize effectiveness demand response, make total as close expectations possible. First, according arrangements startup shutdown various devices actual production business, curve platforms that reflect habit put into a day obtained by LOESS (STL). Second, S–G algorithm used determine each platform evaluate users. To show proposed method, peak-shaving Zhejiang province utilized case studies. The simulation results with different abilities accurately provide scientific guidance companies implement demand-side management.