Construction of power network security risk assessment model based on LSA-SVM algorithm in the background of smart grid

Abstract

AbstractDue to theintricate and interdependent nature of the smart grid, it has encountered an increasing number of security threats in recent years. Currently, conventional security measures such as firewalls, intrusion detection, and malicious detection technologies offer specific protection based on their unique perspectives. However, as the types and concealment of attacksincrease, these measures struggle to detect them promptly and respond accordingly. In order to meet the social demand for the accuracy and computation speed of the power network security risk evaluation model, the study develops a fusion power network security risk evaluation algorithm by fusing the flash search algorithm with the support vector machine. This algorithm is then used as the foundation for building an improved power network security risk evaluation model based on the fusion algorithm.The study's improved algorithm's accuracy is 96.2%, which is higher than the accuracy of the other comparative algorithms; its error rate is 3.8%, which is lower than the error rate of the other comparative algorithms; and its loss function curve convergence is quicker than that of the other algorithms.The risk evaluation model's accuracy is 97.8%, which is higher than the accuracy of other comparative models; the error rate is 1.9%, which is lower than the error rate of other comparative models; the computing time of the improved power network security risk evaluation model is 4.4 s, which is lower than the computing time of other comparative models; and its expert score is high. These findings are supported by empirical analysis of the improved power network security risk evaluation model proposed in the study. According to the study's findings, the fusion algorithm and the upgraded power network security risk evaluation model outperform other approaches in terms of accuracy and processing speed. This allows the study's maintenance staff to better meet the needs of the community by assisting them in identifying potential security hazards early on and taking the necessary preventative and remedial action to ensure the power system's continued safe operation.