Power System Transient Stability Judgment Method Based on Support Vector Machine

Abstract

The invention relates to a power system transient stability determination method based on a support vector machine and belongs to the technical field of power system automatic analysis. The method is characterized by determining an input characteristic set of the SVM (Support Vector Machine) using optimization searching step by step, and further establishing transient stability assessment rules through the SVM. The method comprises a first step of determining an input vector alternative set, an input vector element number, a kernel function of the SVM and a training parameter; a step of generating a training sample and a testing sampling; a step of adding all alternative characteristics into the input characteristic set one by one, training the SVM, and determining a characteristic which enables the SVM classification accuracy to be highest; a step of determining whether the characteristic selection computation is completed or not and output of the input characteristic set; and a final step of training the SVM to obtain the stability rules. According to the invention, the algorithm is easy to implement and the computational complexity is low; the method can be applied to on-line safety and stability assessment of power system at different levels such as regional power systems and provincial power systems in the country; and the system safe and stable operation is effectively ensured under the background that the power grid characteristic is gradually complex.

Description

technical field [0001] The invention belongs to the technical field of power system scheduling automation, and in particular relates to a power system transient stability judgment method based on a support vector machine. Background technique [0002] Power system transient stability analysis is used to discover potential hidden dangers in power system operation and guide power system security and stability control, which is an important link in the normal operation of power systems. At present, the transient stability analysis method used in the power system is the time-domain simulation method, which is complex and time-consuming, and it is difficult to realize online analysis and calculation. [0003] Mainly used for classification, support vector machines are an application of supervised learning methods, that is, training with samples to establish input-output mapping rules. The form of the training sample is (X, y), where X is the input vector, the elements of which n...

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to overcome the problems that the existing transient stability judgment method of the power system is complex in calculation, time-consuming, and difficult to realize online analysis, and provides a transient stability evaluation and judgment method of the power system based on a support vector machine. The present invention is based on data The excavation method generates power system transient stability evaluation rules off-line, which can be directly applied to online transient stability analysis, overcomes the problem of large amount of calculation in time domain simulation method, and realizes the judgment of online transient stability of power system

Images