A Stochastic Approach to “Dynamic-Demand” Refrigerator Control

Dynamic demand management is a very promising research direction for improving power system resilience. This paper considers the problem of managing power consumption by means of “smart” thermostatic control of domestic refrigerators. In this approach, the operating temperature of these appliances and thus their energy consumption, is modified dynamically, within a safe range, in response to mains frequency fluctuations. Previous research has highlighted the potential of this idea for responding to sudden power plant outages. However, deterministic control schemes have proved inadequate as individual appliances tend to “synchronize” with each other, leading to unacceptable levels of overshoot in energy demand, when they “recover” their steady-state operating cycles. In this paper we design decentralized random controllers that are able to respond to sudden plant outages and which avoid the instability phenomena associated with other feedback strategies. Stochasticity is used to achieve desynchronization of individual refrigerators while keeping overall power consumption tightly regulated.