Risk-averse transactions optimization strategy for building users participating in incentive-based demand response programs

Abstract

By participating in demand response programs, building users can provide energy flexibility for the power grid while earning economic benefits for themselves. However, the win-win situation described above is often challenged by the presence of ineffective demand response, over-target demand response and under-target demand response, which pose risks to the economic returns of building users and the stable operation of the grid. To this end, this study proposes a risk-averse transaction optimization strategy to offer bidding schemes and operational strategies that effectively balance the effectiveness of demand response with operational economic benefits. First, a novel model for calculating incentive subsidy is developed, which takes into account the power grid's requirements for demand response effectiveness, as well as differentiated subsidy and penalty mechanisms. Then, a risk-averse optimal dispatch model for building integrated energy systems is established. By collaboratively optimizing the declared response quantity and operation strategies, this method reduces the risk of incentive subsidy loss faced by building users and enhances demand response effectiveness. Lastly, the proposed models are applied in the power market of Shenzhen, China, and the effectiveness of the method is assessed based on three key performance indicators: incentive subsidy loss rate, operating cost reduction rate, the actual load response rate. The results indicate that the total incentive subsidy loss rate is reduced from 37.16 % to 17.75 % through the application of the risk-averse optimal dispatch model proposed in this paper, with operating cost savings ranging from 3 % to 8.09 %. More importantly, the effectiveness of demand response is significantly improved, with the proportion of effective response duration increasing from 18.60 % to 65.93 %. The proposed method is further validated by adjusting the penalty coefficient. Findings show that the risk-averse method provides more robust results and provides more reliable bidding schemes for declared response quantity.