Power system event sensing method based on compressed PMU data and local outlier factor

Abstract

The invention relates to a power system event monitoring method based on compressed PMU data and a local outlier factor, and the method comprises the steps: providing an improved bottom-up (BU) compression algorithm, so as to reduce the scale of the PMU data in a transformer substation, thereby reducing the burden of a communication system; defining local anomaly factors (LOFs) with similarity between any two nodes searched for by an area distance function and principal component analysis (PCA) suitable for metric unequal interval data intervals and monitoring anomalous events in the power system using respective theories, the substantial area of the event may be determined by the LOF value of each node. The method can be applied to on-line monitoring and enhance the situation awareness capability of power system operation personnel.

Description

technical field [0001] The invention relates to the field of power systems, in particular to a power system event perception method based on compressed PMU data and local outlier factors. Background technique [0002] With the continuous growth of renewable energy and clean energy, the dynamic response of the power system is becoming more and more complex, and the power system has become more and more complex and invisible. Therefore, new event monitoring algorithms and recognition algorithms must be proposed to improve the situational awareness of power system operators. [0003] In recent years, Wide-Area Measurement System (WAMS) has been widely used in power systems, which can provide data support for online power system situation awareness. Some real-time operating state identification and development trend prediction methods have been proposed, such as data-based and physics-based methods. In recent years, the methods widely discussed by scholars include discrete wav...

AI Technical Summary

Problems solved by technology

For DWT, KT, and VMD, they require complex mathematical transformations and are also very time-consuming

For PCA, the dynamic characteristics of the power system are linearized, which will lead to inaccurate event monitoring results

For KPCA and RQA, although the nonlinear characteristics of the system are considered, it is difficult to choose a suitable nonlinear mapping function

For the MVEE algorithm, it is first necessary to solve a complex mathematical optimization problem, which limits its real-time application in large power systems

Moreover, with the popularity of PMUs and WAMS, sampling a large amount of data is a heavy task for communication systems, which is not considered by all the above methods

Images