Research On Market Based Decision-making For Load Aggregators Under Smart Grid Environment

The development of smart grid technology has greatly promoted the implementation of demand side response programs. As a professional integrator of demand response resources, the Load Aggregator, LA, has arisen recently to provide opportunities for adjustable customers to participate in wholesale electricity market, which is unutilized due to their small and medium-sized capacity. At present, the research on the bidding and dispatching strategies of LA is unsystematic. So it’s meaningful and worthy to do decision-making research for LA participating in market transactions. A series of decision problems from resource aggregating, market bidding to load dispatching are studied in this article. Besides, the uncertainty factors resulted from the usage habits and subjectivity of electricity users are considered, and elusion measures are given.Firstly, the functions and trading manners of LA are summarized based on overseas research status, and the framework of the wholesale electricity market with LA is established. Then, considering the information of wholesale market and the availability of adjustable customers, a bidirectional decision model for LA is proposed under this framework. The market layer makes the market bidding strategies for LA, which seeks to maximize LA’s payoff for participation in electricity market. The customer layer arranges the load-shedding plans for customers, to minimize LA’s scheduling cost. The proposed model is examined based on PJM historical metered load data at a certain day. And the benefits analysis with LA participating in the markets shows the necessity of promoting LA in domestic electricity market.However, the uncertainty factors resulted from the electricity users will have an effect on its economic efficiency. To compensate for the short power, an energy storage capacity optimization model is established based on grading compensation rules. The bounded uncertain load, caused by customers’ usage habits and subjectivity, is represented by truncated normal distribution model in this article. And according to the percentage of unresponsive load, LAs are classified into four grades, facing different compensation standards and market admission rules. Through examples with PJM historical load data, it is proved that the model can solve the energy storage capacity optimization problems for LAs to reduce the effects of unresponsive load. Besides, another conclusion can be drawn that reasonable market rules can propel LAs to upgrade their resources, which will definitely contribute to the stable and reliable running of the market.