Grid disturbance prediction method and grid disturbance prediction device

Abstract

The invention provides a grid disturbance prediction method and a grid disturbance prediction device. The method comprises the following steps: acquiring monitoring data of nodes in a grid; judging whether the nodes are disturbed according to the monitoring data; when one or more nodes are disturbed, inputting the disturbance state values of the nodes at the current moment to an LSTM-RTRBM deep learning model; predicting the disturbance state values of the nodes within one or more time steps after the current moment through the LSTM-RTRBM deep learning model; and predicting the propagation condition of disturbance between the nodes along the one or more time steps according to the disturbance state values of the nodes within one or more time steps after the current moment. There is no need for operation according to various actual physical parameters in the grid. By using the LSTM-RTRBM deep learning model, the propagation condition of disturbance in space and time dimensions can be predicted through full data drive. The efficiency and accuracy of disturbance propagation prediction are increased.

Description

technical field [0001] The invention relates to the technical field of power system operation and control, in particular to a power grid disturbance prediction method and device. Background technique [0002] With the rapid development of my country's power grid, the connection between power networks is becoming increasingly close, and the phenomenon of disturbances in the power grid spreading between various nodes in the power grid is becoming more and more common. Once a disturbance occurs in the power grid, it may lead to chain events such as transmission line overload, generator oscillation, island effect in the power network and even a blackout event. These incidents reflect the vulnerability of the connections between power transmission systems and the lack of effective fault protection measures. Therefore, diagnosing power faults in time and studying the disturbance propagation characteristics between power grids can accurately locate vulnerable nodes and take proact...

AI Technical Summary

Problems solved by technology

Due to the increasingly complex structure of the power grid, the calculation becomes more and more complex, the efficiency is too low, and effective results cannot be obtained in time, so it is difficult to apply it to the actual power network security situation assessment

Images