Research on Security Situation Awareness and Testing in Power Dispatching Based on Machine Learning Bayesian Algorithm

To address the problem that power dispatch communication networks are prone to abnormal behaviors such as anomalous flows and at the same time reduce the reliability of power dispatch, we have investigated a machine learning Bayesian algorithm-based security situational awareness method for power dispatch. Specifically, we use the Hadoop platform to collect power dispatch data corresponding to the security situational awareness impact indicators. At the same time, we use the subject model algorithm to project the collected power dispatch data into the optimal identification vector space to determine the power dispatch security situational awareness feature vector. Then we use the Bayesian algorithm to calculate the a posteriori probability of the security situational awareness feature vector to determine whether the power dispatch data belongs to the anomaly category and realize the power dispatch security situational awareness. The test results show that the method can effectively sense abnormal behaviors such as traffic anomalies and protocol anomalies in power scheduling, and issue timely warnings in response to the results of security situational awareness, so as to improve the reliability of the operation of the power system.