Details

Autor(en) / Beteiligte

• Hu, Xuekai
• Yang, Shaobo
• Hu, Wenping
• Wang, Lei
• Zhou, Wen
• Su, Can
• Wang, Hongli
• Ligao, Junjie
• Xie, Kaigui
• Hu, Jianlin
• Yang, Qingxin
• Li, Jian

Titel

Schedulable Potential Aggregation Model and LSTM-Based Time-Series Prediction Method of Flexible Load

Ist Teil von

• The Proceedings of the 17th Annual Conference of China Electrotechnical Society, 2023, Vol.1012, p.566-574

Ort / Verlag

Singapore: Springer

Erscheinungsjahr

2023

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• As a large number of load-side flexible resources are explored, more and more flexible loads are involved in the power dispatching process. However, the schedulable potential of individual flexible loads is small. It is difficult to analyze them accurately. The dispersed flexible load resources need to be aggregated to better participate in system dispatching. Therefore, how to accurately evaluate and predict the schedulable potential of flexible loads becomes a problem to be solved in demand-side management. In this paper, the behavioral characteristics of three typical flexible loads are modeled first. Then, in order to aggregate the typical flexible load models, a load schedulable potential aggregation model based on Minkowski sum is established. On this basis, an LSTM-based time series prediction method is proposed to predict the schedulable potential of flexible loads. Finally, the proposed model and prediction method are validated and analyzed by means of a case study.