Control of Residential Air-conditioning Loads to Provide Regulation Services under Uncertainties

This paper presents a model-based approach for the collective control of residential air-conditioning loads to deliver robust and optimal demand management. The proposed approach performs an optimal trade-off between accurate tracking of system operator specified load-set points and minimisation of consumer discomfort, while ensuring robustness to parametric uncertainties and fluctuations in outdoor temperature. Benefiting from robustness to uncertainties, the proposed approach is reliant on minimal household specific information. The mathematical model of the population of residential air-conditioning loads is obtained through the aggregation of individual household specific thermal models. This is followed by the development of a robust model predictive control approach for aggregate demand management to deliver optimum regulation services to account for uncertainties in model mismatch and the prediction errors associated with outdoor temperature. The approach is consistent with the existing demand response standards and is validated using a reference signal from PJM. The results demonstrate that the developed control scheme is capable of precisely following the system-operator specified load set-points even under the worst-case uncertainties of thermal parameters. While achieving the target set-point, it is further observed that customer comfort is always preserved along with minimum compressor control action on air conditioners.