Power quality disturbance identification method based on online weighted SVM training

Abstract

The invention discloses a power quality disturbance identification method based on online weighted SVM training. The method comprises: according to different characteristics of common power quality disturbance signals, a mathematical model of the disturbance signals is established; on the basis of the mathematical morphology, a generalized morphological filter of an adaptive LMS algorithm is constructed to carry out denoising processing on a power quality disturbance signal; wavelet transform is carried out to extract energy distribution of different frequency bands of the power quality disturbance signal; a fractal dimension of the power quality disturbance signal is extracted based on a fractal theory and the energy distribution and the fractal dimension form a quantification characteristic index of the disturbance signal; and according to time of power quality disturbance occurrence, disturbance data are assigned with different weights, online training and testing are realized, and identification classification is carried out on power quality disturbance.

Description

technical field [0001] The invention relates to the field of power quality identification, in particular to a power quality disturbance identification method based on online training weighted SVM. Background technique [0002] my country's power system has developed into a time-varying system with multiple variables and multiple objectives. In recent years, due to the profound changes in the power supply, grid and load of the power system, more and more nonlinear, impact and unbalanced loads have been put into use in the power system, such as steelmaking electric arc furnaces, electrified railways, power electronics Devices, etc., lead to a series of power quality disturbance problems such as voltage fluctuations and flicker, harmonics, and system frequency fluctuations. Therefore, the power quality problem has attracted extensive attention. In order to control the occurrence of disturbances and improve power quality, accurate detection and identification of power quality ...

AI Technical Summary

Problems solved by technology

[0018] ①Nonlinear problems in power system components

[0026] ① Failure to establish a timely automatic control mechanism for bus voltage in substations

Untimely regulation of the busbar is the main cause of voltage over-limit

[0027] ② Lack of small hydropower grid connection management system

The non-standard connection of small hydropower will lead to the voltage limit of the 35KV and 10KV busbars in the substation

[0028] ③ Unreasonable grid configuration

For example, urbanization has developed rapidly in recent years, and the original suburbs have rapidly developed into new urban centers, but the grid power supply facilities cannot keep up, resulting in an excessively large power supply radius, which directly affects the reliability and stability of power supply

[0029] ④ Low level of power grid grassroots management

[0030] ⑤ Lack of power grid pollution control

There are a large number of nonlinear loads, impact loads, and unbalanced loads connected to the system in the power grid itself, which pose a serious threat to power quality, and the control of these pollutions has not yet received sufficient attention at this stage

Images