A Hierarchical Localization Method for Forced Oscillations Based on Multi-Stage Transfer Learning

Abstract

The invention discloses a forced oscillation hierarchical positioning method based on multi-stage transfer learning. The method includes two parts: offline training and online positioning. First, the power system is partitioned according to the generator correlation, and the smooth pseudo Wigner-Ville distribution of the oscillation principal components of each partition is visualized to form an interval WVD image. Then, the first stage of migration learning is performed on the pre-trained convolutional neural network to obtain the first layer partition localization model. Input the WVD image in the area of ​​the positioning partition, and then perform the second stage transfer learning on the partition positioning model to obtain the second-level unit positioning model. Finally, the offline positioning accuracy of this method is verified. The on-line location of the disturbance source is realized by successively inputting the WVD image of the section and the WVD image in the area where the forced power oscillation actually occurs into the partition location model and the unit location model. The present invention not only has high positioning accuracy, but also has the characteristics of fast positioning speed, high adaptability, strong robustness and the like.

Description

technical field [0001] The invention relates to a forced oscillation hierarchical positioning method, in particular to a forced oscillation hierarchical positioning method based on multi-stage transfer learning. Background technique [0002] In the context of the interconnection of large regional power grids, the risk of low-frequency oscillations in power systems is increasing. As a kind of low-frequency oscillation, forced power oscillation may seriously threaten the safe and stable operation of the power system once it occurs and is not dealt with in time due to its zero-attenuation characteristics. It occurs because there are continuous periodic disturbance sources in the power system. The most direct suppression method is to find the location of the disturbance source in the power grid and remove it quickly. However, when the disturbance frequency of the disturbance source is equal to or close to the natural frequency of the system, the power system will have low-frequ...

AI Technical Summary

Problems solved by technology

However, as the scale of the power grid expands, the system operation mode becomes more changeable, which will lead to some important assumptions of the above method no longer valid.

For example, when using the traveling wave detection method for positioning, the technology is not mature enough for the occurrence of multi-mode oscillation, which may lead to misjudgment; when using the transient energy function to locate the disturbance source, for the case of large network loss, it may Two or more positioning results will be obtained

At present, studies have pointed out that the transient energy function method sets too strict model assumptions for the power system network and loads, which leads to great limitations in the case of forced power oscillations in the actual power system.

In addition, with the development of large-area interconnection of the power grid, the characteristics of the power grid are becoming more and more complex. It is increasingly difficult to solve the problem of disturbance source location only from the physical mechanism of forced oscillation itself and manual experience, and there are problems such as harsh operating conditions and low accuracy.

Moreover, analyzing the location of the disturbance source from the perspective of the entire network, due to the need to transmit and process the data of the entire network, there will be problems of excessive data volume and information redundancy

Images