Search

Due to the problem that dynamic vulnerability assessment takes too long when blind attack occurs in the range of 1-10 transmission lines when the original method is used for vulnerability analysis and quantitative assessment of power system lines, a method of power system line vulnerability analysis and quantitative assessment under blind attack is proposed. The information physical power system is abstracted into an undirected network, and the corresponding adjacency matrix of the information physical power system is represented by the topology structure of the undirected network, so as to realize the construction of the power system model. Under the blind attack, the operation status of the power system model is initialized, and the functional failure of line components is determined, and the vulnerability analysis of power system lines is implemented. A set model of power system line vulnerability is constructed, and the quantitative assessment of power system line vulnerability is carried out according to the model. In order to prove that the dynamic vulnerability assessment of this method takes less time, the method is compared with the original method. The experimental results show that the dynamic vulnerability assessment time of this method is lower than that of the original method, and the evaluation performance is improved.

Aiming at the problem of low detection accuracy in traditional power system network intrusion detection methods, in order to improve the performance of power system network intrusion detection, a power system network intrusion detection method based on random forest algorithm is proposed. Firstly, the power system network intrusion sub sample is selected to construct the random forest decision tree. The random forest model is optimized by using the edge function. The accuracy of the vector is judged by the minimum state vector of the power system network, and the measurement residual of the power system network attack is calculated. Finally, the power system network intrusion data set is clustered by Gaussian mixture clustering Through the design of power system network intrusion detection process, the power system network intrusion detection is realized. The experimental results show that the power system network intrusion detection method based on random forest algorithm has high network intrusion detection performance.