Following a significant disturbance in the power grid

the urgency of power demand varies across different stages of frequency change. To ensure frequency security of the power grid

a multiple time interval scheduling decision model for the air conditioning load aggregator is established to optimize the allocation of limited controllable power for air conditioning loads. Firstly

this paper comprehensively considers users’ willingness to respond

the temperature range of user acceptability

and the uncertainty of user behaviors to design a diversified incentive compensation model and a reward-penalty mechanism. Secondly

a risk assessment model that considers the risk of user withdrawal from response is proposed to assess the risk of users withdrawing during each response period. Based on the conditional value at risk theory

a multiple time interval scheduling decision model for air conditioning load aggregators that consider frequency security of the power grid and the risk of user withdrawal is established. Finally

the effectiveness of the proposed model is validated through calculation examples.