Find in Library

Collecting fine-grained data on electricity usage can reduce the user’s energy cost by accurate demand response and improve the services from utilities. However, non-intrusive power load monitoring (NILM) technology shows that these data will reveal the user’s privacy. In this paper, we first design a price-sensitive BLH scheme which achieves differential privacy. Then regarding such issues above, we propose a price-sensitive cost-friendly differential privacy (PCDP) scheme through a contextual multi-armed bandit framework. Last, realistic electricity consumption data is extracted to validate the proposed method. The results show that our method provides effective privacy protection and achieve considerable cost saving.