Power system bad data identification method based on improved Wasserstein GAN

Abstract

The invention discloses a bad data identification method for a power system based on an improved Wasserstein GAN. The bad data identification method comprises the following steps: screening historical measurement data only containing Gaussian white noise in a historical database and preprocessing the historical measurement data; using the preprocessed historical measurement data as target data to train a WGAN-GP model; respectively establishing loss functions of the generator and the discriminator, carrying out game training, and obtaining a WGAN-GP model after the training is completed; performing data preprocessing on the collected real-time measurement data of the current section, and inputting the trained WGAN-GP model to obtain measurement reconstruction data of the current section; and obtaining a reconstruction error of the current section based on the measurement reconstruction data of the current section and the real-time measurement data, and inputting the reconstruction error into the trained C4.5 decision tree model to carry out bad data identification on the measurement data of the current section. According to the method, on the basis of real-time measurement information reconstruction, the measurement reconstruction error of the current section is obtained, bad data can be rapidly and accurately identified, and the identification efficiency is considered while the identification performance is ensured.

Description

technical field [0001] The invention relates to a method for identifying bad data in a power system based on an improved Wasserstein GAN, and belongs to the technical field of power systems. Background technique [0002] With the proposal of the "30-60 double carbon" goal, a large number of new energy grids have led to an exponential increase in the amount of data that the power system needs to process, and at the same time make the data structure of the power system more and more complex, so it is very important for the reliability of the system operation. Higher requirements are put forward for performance, security and stability. As one of the core functions of the power system energy management system, state awareness is of great significance to the planning and operation of the power system, and provides a reliable database for real-time scheduling and a series of subsequent advanced applications and analysis of the power system. In addition to the normal data noise du...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems of low identification accuracy and poor identification efficiency of large-scale power system bad data, the present invention provides a power system bad data identification method based on improved Wasserstein GAN, which uses the improved Wasserstein GAN to reconstruct the real-time measurement information of a certain section , so as to obtain the measurement reconstruction error of the current section, so as to accurately detect the position of bad data in a set of real-time measurement information, and then quickly and accurately identify the bad data

Images