1.Alxa Power Supply Company of Inner Mongolia Electric Power (Group) Co., Ltd., Alxa 750300, China;2.School of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan 430074, China

The peak shaving resource at demand side is an important part of peak shaving resources in new power systems. However, due to the internal and external factors such as the uncertainty of user behaviors and abnormal equipment status, some loads often fail to respond reliably. Aiming at the low response reliability of demand-side resources in industrial parks during the actual peak shaving, a precise peak shaving strategy at demand side is proposed considering load response reliability and aggregator scheduling economy. Firstly, a regulation model is established for the direct control of central air conditioning load and industrial load, and the relevant factors that affect its response reliability are analyzed by combining diversified characteristic loads. Secondly, the load response reliability index is defined to quantitatively characterize the influence degree of internal and external factors on load response reliability. Together with load regulation cost, it is used as a decision variable for load aggregation, and the peak shaving response optimization scheme at demand side is solved through iterative algorithms. Finally, based on the central air conditioning load data from the Open Energy Data Initiative (OEDI) dataset of America and the load data from an industrial park in Hubei Province, China, the proposed precise peak shaving strategy is compared and simulated with the economic-only peak shaving strategy by using economic incentives as the only means, and the total scheduling cost and actual response ratio of load aggregators are analyzed as evaluation indicators to verify the effectiveness of the proposed strategy.