Mila Nedić and Tea Tušar

Abstract
This paper presents a method for changing the energy use of consumers participating in Demand Response (DR) programs, focusing on peak balancing to improve grid stability. Multiple objectives including costs and risks are considered, and a weighted sum is used to transform them into a single objective. This results in an optimization problem that can be optimally solved. To calculate the costs, the load consumption baseline needs to be established. Since this is challenging and can be exploited,
we conduct initial experiments to test whether our method to adjust the baseline can be easily manipulated. We explore an original scenario and three of its variants to examine the effects
of various parameters on the optimization outcome. Our results indicate that 1) an excessive emphasis on risk results in no energy change, 2) enforcing a net zero energy change minimizes energy use while still securing the rebate, and 3) without an adjustment period, the consumer is less inclined to increase the load just before the demand period. In future work, we will reformulate some objectives to avoid exploitation and better reflect the real-world needs of DR.