A Dynamic Adaptive Method for Power System Terminal Security Threat Assessment

Abstract

The invention relates to a dynamic self-adaptive power system terminal security threat evaluation method. The dynamic self-adaptive power system terminal security threat evaluation method comprises the following steps: acquiring data and cleaning the data; extracting characteristics, namely aiming at independent data generated by a power system terminal, extracting the characteristics to establish a characteristic vector of the power system terminal; constructing an evaluation model; constructing an evaluation set; carrying out fuzzy evaluation: constructing a judgment matrix according to a judgment sample total body and each judgment score; and quantifying an evaluation result. According to the dynamic self-adaptive power system terminal security threat evaluation method, starting from the aspect of a hierarchical model, a related matrix is arranged and a threat effect can be transmitted to the whole model when one party is influenced between layers, so that the risk evaluation of environment dynamic changes is realized. A strategy set is set, and a strategy is selected from the strategy set firstly when dynamic adjustment is needed, so that the evaluation effect can be more rapidly given by applying the model, and the efficiency is greatly improved.

Description

technical field [0001] The invention relates to an assessment method in the technical field of power system security threat assessment, in particular to a dynamic self-adaptive power system terminal security threat assessment method. Background technique [0002] As the power system interacts more frequently with the actual physical world, the security of power terminals is facing more and more challenges. Fully excavating monitoring data information has become one of the important directions of power system for terminal risk assessment and security protection, especially with the development of cyber-physical integration, the attention to the assessment of power system terminal security is more urgent. [0003] At present, there are many security evaluation models. The hierarchical analysis model establishes the security evaluation hierarchy of the terminal from the perspective of security layering and grading, and achieves security transmission through the mutual influence...

AI Technical Summary

Problems solved by technology

However, the lack of dynamic evaluation considerations makes the adaptability of the security evaluation model poor when the physical environment changes, which is not conducive to actual deployment and application

The dynamic risk assessment based on the neural network algorithm needs to use a large amount of historical data to train the network, and the calculation complexity is high. Usually, the network needs to be retrained after the data changes, which has great limitations on the dynamic adaptability of the environment. In addition, data training There are also problems of overtraining and network instability

The algorithm based on the hidden Markov model has strong theoretical guidance in theory, but in actual use, it has to face the problem of large deviation of the model algorithm caused by data anomalies

Images