Answers are grounded in this paper and cite their sources.
In order to solve the problem of high latency and low precision collaborative optimization in smart grid security prediction, this paper proposes a smart grid security prediction and reliability analysis method (EA-LSTM-GraphAgg, EALG) based on edge computing. In the initial prediction stage, EA-LSTM is deployed on edge servers in various regions. Through collaborative optimization of attention mechanism and competitive random search, the temporal traffic data of smart grid collection nodes is locally processed to generate local initial prediction results. In the aggregation stage, to address the local limitations of initial predictions, a GraphAgg aggregation network is designed. The EA-LSTM prediction results are transmitted through lightweight collaborative directional transmission between edge nodes, rather than the original data. At the edge, a graph structure is constructed with smart grid devices as nodes and communication connections as edges. Through localized information aggregation, node traffic associations are mined to improve prediction accuracy. Experiments and analysis show that EALG can output high-precision smart grid security prediction results. It can provide reliable support for smart grid reliability analysis and traffic anomaly warning in smart grid operation and maintenance.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: