Generative adversarial network-based power system dynamic security assessment method

Abstract

The invention discloses a generative adversarial network-based power system dynamic security assessment method. The method comprises the following steps of 1, constructing an initial sample set containing a large number of operation variables and transient stability margin security indexes; 2, constructing an efficient sample set containing key feature data and TSMs labels; 3, obtaining a GAN model and a dynamic security evaluation model capable of accurately filling missing data; 4, detecting the integrity of PMUs data in real time through the intelligent system, and if the data is complete,directly inputting the data into the DSA model for DSA of the power system; and if there is data missing, inputting incomplete data collected by the PMUs into the GAN model to fill the missing data, and inputting the filled complete data set into the DSA model to perform DSA of the power system. According to the electric power system DSA method based on the GAN, missing data can be rapidly and accurately filled and DSA can be performed on the system for the problem of data missing of the electric power system so that electric power operation personnel are enabled to rapidly take control measures and occurrence of faults and economic loss can be avoided.

Description

technical field [0001] The invention relates to the field of dynamic security assessment of power systems, in particular to a method for dynamic security assessment of power systems based on generative confrontation networks. Background technique [0002] With the increasing popularity of renewable energy and the active response of the demand side, the power system introduces more uncertainties, which poses a great challenge to the dynamic security assessment (Dynamic Security Assessment, DSA) of the power system. In general, traditional DSA model inputs are complete and always available online. However, due to practical problems such as synchrophasor measurement units (PHAsor Measurement Units, PMUs) failure, communication congestion / failure, and even cyber attacks, the dynamic security assessment of power systems will be affected. While traditional methods are based on complete data entry, without data these methods fail. Therefore, there is an urgent need for fault-tole...

AI Technical Summary

Problems solved by technology

[0006] ① Missing data is generated using mathematical model-based methods, and the rate of data generation is relatively low compared with data-driven methods;

[0007] ② This method of filling missing data fails to directly give the predicted probability of generated data, and lacks sufficient experiments to prove the applicability of this method in power system security assessment, and there is still room for improvement in terms of feasibility and scalability

Images