Anti-collapse multi-time scale generative adversarial network generator parameter correction method

Abstract

The invention provides an anti-collapse multi-time scale generative adversarial network generator parameter correction method, and the method comprises the steps: carrying out the sampling of a random noise signal and a real signal of a generator at different time scales, and inputting the sampled generator signal data into a generator; inputting the sampled generator random noise signal, the generator real signal and the generated signal generated by the generator into a discriminator; and setting an anti-collapse threshold value, judging whether the probability that the judgment signal obtained in the discriminator is a real signal is greater than the anti-collapse threshold value, storing the data greater than the anti-collapse threshold value into the model database, and learning the data less than the anti-collapse threshold value through the conditional construction generative adversarial network again. According to the method, the problem of generator parameter correction can be effectively solved, the obtained result can be continuously optimized through multi-time-scale sampling, the generation and discrimination functions of the conditional construction generative adversarial network are achieved, and the anti-collapse capacity in the generator parameter correction process is improved.

Description

technical field [0001] The invention belongs to the field of electric power system and artificial intelligence application, relates to an artificial intelligence method, and is suitable for correcting parameters of electric power system generators. Background technique [0002] Power system dispatching operation, safety assessment, and stability analysis all rely on dynamic simulation results. The accuracy of the simulation results will directly affect the safe operation and economic dispatch of the power system. Therefore, it is particularly important to ensure that the simulation system can correctly reflect the dynamic characteristics of the actual system. . A large number of calculation and analysis results show that there is a big difference between the simulation results and the measured data, and the inaccuracy of the generator model parameters is considered to be the main cause of the error. In summary, the generator model parameter calibration is an indispensable p...

AI Technical Summary

Problems solved by technology

However, regular on-site testing of generator model parameters greatly increases the workload and expense of the power plant, and the normal operation of the generator is also affected to a certain extent. At the same time, the parameters measured on site will also be affected by various errors. influences

[0004] Therefore, the present invention proposes an anti-collapse multi-time scale generative confrontation network generator parameter correction method, which can solve the problems of workload and cost burden of the power plant

Images