An Improved Abnormal Power Consumption Detection System Based on Federated Learning
Zhencong Cai, Xinyuan Jiao, Liu Wei, Ying Wang · 2023
In the power business scenario, the existing abnormal power consumption detection system based on federated learning generally uses a single encryption algorithm to encrypt information during data transmission, which has certain limitations in efficiency and security. In this paper, we propose an improved abnormal power consumption detection system based on federated learning, which optimizes the encryption algorithm for the actual power business scenario, and adds the mixed use of multiple algorithms. The results show that in the actual power business scenario, compared with the traditional abnormal power consumption detection system based on federated learning, the system can better improve the efficiency of model training.