Intelligent power grid line state error data injection attack detector and detection method

Abstract

The invention relates to an intelligent power grid line state error data injection attack detector and detection method, and belongs to the field of intelligent power grid safety detection application. The detector is composed of four parts of a detector data collector, a line identifier, a data aggregator and a classifier. Compared with the conventional detector, the detector can be installed oneach sub-controller of the intelligent power grid. The computing resource, the storage resource and the communication resource of the sub-controllers are fully utilized to design a distributed detector framework. Each sub-controller is provided with an independent detector for judging error data injection. The characteristic of limited computing capacity of the sub-controllers is considered, the training process of the classifier is transferred to the central controller and each sub-controller only requires to classify the data. Meanwhile, the related six-dimensional arrays are selected to generate the sample points to act as the input of abnormity detection according to the inherent attribute of the line, i.e. the characteristic of constant resistance.

Description

technical field [0001] The invention relates to a detector and a detection method aiming at the attack of the wrong data injection of the circuit state of the smart grid, and belongs to the application field of smart grid security detection. Background technique [0002] Smart grid is a large-scale and complex system that is closely integrated with power grid and information system, including control system, detection system and physical system, such as figure 1 shown. Physical systems generate, transmit, distribute and use electricity. The detection system verifies whether the perceived physical system data is complete, and the verified data is transmitted to the control system. The control system sends out new control signals based on the perceived data, and the physical system further executes according to the control signals. The physical system consists of many grid physical components, such as transmission lines and consumers. The detection system consists of sensor...

AI Technical Summary

Problems solved by technology

[0006] Although the above methods can effectively detect the injected erroneous data, there are still the following problems: (1) Most of the existing detection methods are not aimed at the erroneous data injection attack of smart grid transmission line sensors, if directly applied to the line state information Malicious modification detection has low detection accuracy; (2) The existing detection mechanism placed in the central controller cannot meet the needs of the sub-controller to quickly operate the subsystem. When the sub-controller makes a decision based on the sensory data, it needs to wait for the central detection tor detection results, causing a delay

The delay will cause the execution of system control to be delayed, which may cause physical system failure or even cascading failure

Images