Dynamic security misclassification constraint method for power system based on umbrella algorithm

Abstract

A dynamic security misclassification constraint method for a power system based on an umbrella algorithm comprises the following steps: 1, constructing a corresponding dynamic security classificationlabel based on a dynamic security classification rule; 2, performing feature selection on an initial sample set constructed based on a historical operation database of the power system to obtain a keyfeature set; 3, constructing an online dynamic security evaluation model based on a Nelmann Pearson comprehensive classifier of an umbrella algorithm, and sending the key feature set and the corresponding dynamic security classification label as input into the classifier for offline training; 4, considering various power system operation influence factors, performing time domain simulation to obtain an updated sample set, and sending the updated sample set into a comprehensive NP classifier for model updating; and 5, carrying out online DSA through synchronous phasor measurement unit data collected in real time. The dynamic security misclassification constraint method can reduce the impact on the safe operation of a power system from DSA wrong classification, and reduces the social and economic loss.

Description

technical field [0001] The invention belongs to the field of power system dynamic security evaluation, and in particular relates to a power system dynamic security misclassification constraint method based on an umbrella algorithm. Background technique [0002] In recent years, with the continuous development of modern power systems, it has become one of the most complex and largest artificial systems in the world. At the same time, the power system is also facing severe challenges in the new operating environment. On the one hand, due to the high degree of interconnection and coupling of modern power systems, when disturbances and faults reach a certain level, the power system may lose stability and cause serious catastrophic consequences. On the other hand, the continuous access of distributed power sources and various uncertain factors will increase the operational risk of the power system. If the operator cannot quickly and accurately capture the accident safety alarm ...

AI Technical Summary

Problems solved by technology

②The general power system DSA scheme adopts a classifier to construct the DSA model, the evaluation result is too monotonous and conservative and depends on the performance of the classifier to the system, and there is a problem that the applicability to different systems may be different

③For DSA output, the traditional DSA model outputs a single result, which cannot accurately reflect the security state of the system

[0007] ① It is impossible to carry out different constraints on the severity of misclassification of power system faults, and setting a misclassification threshold will easily lead to insufficient constraints on serious class I classification errors and increase the risk of safe operation of the system;

[0008] ② It is unavoidable to provide system operators with a variety of optimal classifiers with one class of classification error constraints to construct a comprehensive DSA model, and the evaluation performance depends on the RPTSVM model itself;

[0009] ③Unable to provide integrated classification results, and the reliability of the results cannot be ensured only based on the degree of membership of the evaluation

Images