Demand Response Strategy for Carbon Reduction Considering the Impact of Environmental Protection Advocacy on Customer Behavior

Residential loads have a great potential to reduce the carbon emission of electricity systems via demand response (DR) programs. One major challenge in DR for carbon reduction is to handle unknown and uncertain customer behaviors. Therefore, this paper designs a DR strategy for carbon reduction based on online learning to solve this challenge. Firstly, considering the carbon emission corresponding to the customer's electricity consumption behavior, the carbon emission evaluation level evaluation system is designed and sent to the customer. Secondly, the overall architecture of demand response is constructed, and the demand response problem affected by customer behavior is constructed as an online learning model. Finally, the Linear Upper Confidence Bound (LinUCB) algorithm is proposed to solve the proposed model. The results show that the reliability of the control strategy is enhanced, and the carbon reduction of the system is increased under the strategy mentioned in this paper.