Subsynchronous/supersynchronous oscillation intelligent identification method based on synchronous phasor

Abstract

The invention discloses a subsynchronous / supersynchronous oscillation intelligent identification method based on synchronous phasors, and belongs to the technical field of artificial intelligence. The identification method comprises the steps of sub / super-synchronous oscillation training data generation, offline training of a support vector machine (SVM) order determination model and a convolutional neural network (CNN) intelligent parameter identification model, and online identification. The method comprises the following steps: firstly, generating training data of different oscillation mode orders, frequencies, damping, amplitudes and phases, then training an SVM order determination model and a CNN intelligent parameter identification model offline by using the data, and finally applying to actual oscillation parameter identification: (1) data acquisition; (2) data processing; (3) modal order determination; (4) performing modal decomposition; and (5) parameter identification. The method can solve the problem that the sub / super synchronous oscillation parameters cannot be identified based on the synchronous phasor data of the master station, and has the characteristics of fast identification and high precision.

Description

Technical field [0001] The present invention relates to a power system wide-area monitoring technology, specifically a secondary / supersynchronous oscillation intelligent identification method based on synchrophasors, belonging to the field of artificial intelligence technology. Background [0002] With the "double carbon" goal proposed, the proportion of renewable energy and power electronic equipment connected to the grid will be greatly improved, and the "double high" characteristics of the new power system with new energy as the main body will be more significant in the future. In this context, the probability of sub-synchronous oscillation is also greatly improved, which seriously threatens the safe and stable operation of the system. Effective monitoring and analysis of sub-synchronous oscillations of power systems ranging from several Hz to tens of Hz is an important challenge in this field. [0003] At present, the methods of subsynchronous oscillation identification bas...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide an intelligent identification method of sub / supersynchronous oscillation based on synchronized phasors, by analyzing the collected sub / supersynchronous oscillation data at the master station, the accuracy of the sub / supersynchronous oscillation of the wide-area power system can be realized Identification, to solve the current problem that cannot be accurately identified based on the synchrophasor data of the master station

Images