Multi-party secure computing algorithm for power grid data transmission based on federated learning

Power grid data usually contains sensitive information, such as user electricity habits, power grid operation status, etc. Direct transmission of such information over the network may result in interception by unauthorized parties, causing privacy breaches. To this end, a multi-party secure computing algorithm for power grid data transmission based on federated learning was studied. By obtaining power grid data, removing duplicate information, using interpolation to supplement missing values in power grid data, and correcting outliers and errors, the standardization of power grid data is completed. Constructing a federated learning model for power grid data and training it, only transmitting model parameters rather than raw data, effectively protecting data privacy. Based on this, secure transmission of power grid data is carried out. By constructing a multi-party secure computing protocol for power grid data transmission, and analyzing the correctness and security of the protocol, we can achieve multi-party secure computing for power grid data transmission. Experimental results have shown that the method proposed in this article can efficiently obtain data and standardize it, improving the accuracy and security of transmission.