Transient stability serious fault screening method combining deep learning and simulation calculation

Abstract

The invention relates to a transient stability serious fault screening method combining deep learning and simulation calculation, and belongs to the technical field of power system stability analysis.The method comprises the following steps: firstly, acquiring the data of a power system before and after fault removal to obtain a training set and a verification set; training the data through utilizing a deep belief network training algorithm before the fault occurs and after the fault is removed to obtain two transient stability evaluation models; calculating the judgment accuracy of the modelunder different thresholds by utilizing the verification set to obtain trustworthy thresholds of output results of the two models; and in combination with the two deep belief network models and simulation calculation, performing the layer-by-layer screening of transient stability serious faults to obtain a final serious fault list. According to the method, the deep learning model and the simulation calculation are combined, a part of transient stability serious faults are screened out by utilizing the deep learning model, and the simulation calculation is adopted to ensure the accuracy of results except a trusted threshold, so that the rapidity and the accuracy of transient stability serious fault screening are both considered.

Description

technical field [0001] The invention relates to a screening method for severe transient stability faults combined with deep learning and simulation calculation, and belongs to the technical field of power system stability analysis. Background technique [0002] The destruction of transient stability is an important reason for large-scale power outages in power systems. How to quickly and accurately judge the transient stability of power systems is one of the important issues to be considered in the safety prevention and control of power systems. Transient stability assessment can be performed at different times, mainly including three types of application scenarios: offline analysis, online pre-decision and real-time calculation. Among them, the online transient stability assessment mode adopts a short-term periodic refresh method, and each refresh cycle will The evaluation is carried out according to the current operating conditions and preset faults of the system. Because ...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to propose a method for screening serious faults in transient stability combined with deep learning and simulation calculations. Aiming at the problem that it is difficult to take into account the rapidity and accuracy of transient stability calculations with a single simulation calculation and machine learning method, the present invention uses simulation The combination of calculation and two deep learning models screens the transient stability preset fault set layer by layer to obtain the final list of serious faults, taking into account the respective advantages of deep learning and simulation calculations, and improving the accuracy of critical fault screening for transient stability evaluation. speed and accuracy

Images