Distributed Privacy-Preserving Estimation for Multi-Area Power Systems With Multichannel Event-Triggered Mechanisms

In power systems with multi-area power exchanges, distributed state estimation not only depends on sensor measurements but also requires area-interaction. Plentiful transmitted information puts pressure on data privacy and communication resources. To preserve the privacy of sensor and interaction data, a switching encryption, whose minimum dwell-time is less than the decryption time of eavesdroppers, is proposed. Based on encrypted data, multichannel event-triggered mechanisms are established to save the communication resources of sensor-to-estimator and area-interaction channels, simultaneously. By designing the appropriate estimator gains and encryption-compensating mechanisms, the asymptotic stability of estimation error dynamics is guaranteed. Compared with the existing privacy-preserving technologies, the switching encryption improves the estimation accuracy of power system states while avoiding the decryption of eavesdroppers. The effectiveness of distributed privacy-preserving estimation is illustrated by a three-area power system. Note to Practitioners—Distributed estimation of multi-area power systems produces lots of communication data, which challenges data privacy and communication resources. Since the estimation of power system states generally requires high accuracy, the existing privacy-preserving technologies, which sacrifice estimation accuracy, are inappropriate for power systems. Therefore, a switching encryption scheme combined with the encryption-compensating mechanisms is proposed to preserve data privacy while the asymptotic stability of estimation error dynamics is guaranteed. Furthermore, the simultaneous transmissions of sensor measurement and area-interaction data put pressure on the communication resource of distributed estimation. Therefore, the multichannel event-triggered mechanisms are proposed to reduce the transmission frequencies of encrypted data.