A power distribution network fault sample enhancement method and system based on a discrete event chain simulator

CN122154408APending Publication Date: 2026-06-05STATE GRID JIANGSU ELECTRIC POWER CO LIANYUNGANG POWER SUPPLY CO
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
STATE GRID JIANGSU ELECTRIC POWER CO LIANYUNGANG POWER SUPPLY CO
Filing Date
2026-02-05
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

Existing power distribution network fault sample generation technologies suffer from insufficient numbers of real fault samples, failing to meet the training requirements of data-driven methods. Traditional waveform simulation is costly and time-consuming, making it difficult to generate large-scale training sets. Simulators cannot automatically generate complete event chain structures, resulting in poor model generalization ability and insufficient consistency and reliability of data sources.

Method used

A method based on a discrete event chain simulator is adopted to generate fault samples containing complete event chains through topology modeling, discrete event generation, event chain scheduling and parameter perturbation. Combined with the structured labels of the event chains and multi-source observation sequences, rapid large-scale sample generation is achieved.

Benefits of technology

It enables the low-cost and efficient generation of a large number of fault samples with complete event chains, which is suitable for large-scale power distribution network applications. It improves the training performance and generalization ability of models such as Bayesian networks, simulates communication delays and real-world engineering problems, and provides more reliable data sources that are closer to engineering applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122154408A_ABST
    Figure CN122154408A_ABST
Patent Text Reader

Abstract

The application relates to a power distribution network fault sample enhancement method and system based on a discrete event chain simulator, and particularly relates to the following: a power distribution network discrete event chain simulator is constructed, including a topological modeling module, a discrete event generation module and an event chain scheduling module; before starting each simulation cycle, multi-dimensional disturbance parameters are dynamically injected into the simulator, the simulator generates a single event chain carrying a set disturbance scenario based on the parameter configuration after disturbance; according to the complete fault evolution process and device action logic recorded in the event chain, structured labels corresponding to the event chain are extracted and generated; guided by the structured labels corresponding to the event chain, mapping from logical events to standardized time sequence data is performed to obtain multi-source observation sequences aligned to a common time axis; a large-scale sample library is generated through multiple cycles. The application can quickly generate a large number of training samples with complete event chains, while retaining engineering characteristics such as electrical topology, protection logic and communication delay.
Need to check novelty before this filing date? Find Prior Art