Combining electric vehicle and rechargeable battery for household load hiding

The transition from electromechanical to advanced smart meters offers great opportunities in load management, energy saving, and resource optimization. However, the fine-grained usage data collected by the smart meters also raises privacy concerns. The household load profile can be analyzed by non-intrusive load monitoring techniques to infer customer activity routines and behavioral preference. Several data obfuscation techniques based on controlling a local rechargeable battery or a controllable load have been proposed to mitigate the privacy leakage of the customers. In this paper, we introduce the use of electric vehicles (EVs) for load hiding. In particular, we propose an EV-assisted battery load hiding algorithm, which combines the use of an EV with a local rechargeable battery to achieve the dual purpose of optimal charging and measurement obfuscation. Since EVs are becoming increasingly popular and part of many households, their use renders the implementation of load hiding less costly compared to methods only using dedicated rechargeable batteries. Furthermore, EVs provide more flexibility than other household loads in terms of energy storage. We evaluate our proposed algorithm using real data for electricity price, household consumption, and EV parameters. Adopting mutual information as a measure for information leakage, numerical results show that the proposed algorithm can reduce the customer's cost while maintaining the expected privacy level.