Power system state space-time deduction method and system based on causal-data fusion

By using a causal-data fusion approach, physical mechanism models and real-time power grid data are fused into a spatiotemporal tensor, which is then input into a convolutional long short-term memory network. This solves the problems of slow extrapolation speed and low reliability in power systems under extreme weather conditions, and enables efficient and reliable risk warning and decision support.

CN122173783APending Publication Date: 2026-06-09SHANDONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANDONG UNIV
Filing Date
2026-03-25
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing methods for extrapolating power system scenarios under extreme weather conditions suffer from slow speed, poor physical model flexibility, and unreliable data-driven models in extreme scenarios, making it impossible to achieve efficient and reliable risk warning and decision support.

Method used

By employing a causal-data fusion approach, the probability of equipment-level failures is calculated through a physical mechanism model and fused with real-time power grid status data into a spatiotemporal tensor format. This data is then input into a specially designed convolutional long short-term memory network for inference, thus constructing a collaborative framework to achieve ultra-fast and highly reliable power system status inference.

Benefits of technology

It enables ultra-fast online simulation and early warning of power systems under extreme weather conditions at the minute or even second level, improving the resilience and decision-making intelligence of the power grid under extreme weather conditions, and providing clear risk trajectories and emergency resource allocation plans.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122173783A_ABST
    Figure CN122173783A_ABST
Patent Text Reader

Abstract

The application provides a power system state space-time deduction method and system based on causal-data fusion, relates to the technical field of power system safety analysis and emergency prevention and control, and comprises the following steps: acquiring meteorological forecast data, power grid real-time state data and topological data; based on the meteorological forecast data, the time-varying fault probability of equipment level in the future period is calculated by using a causal model method of physical mechanism; the time-varying fault probability is taken as risk prior features, and is fused with the power grid real-time state data and the topological data to be reconstructed into a tensor format retaining the space-time topological structure; the space-time tensor obtained after reconstruction is input into a trained space-time deduction model, and the power system state in the future period is recursively deduced; the application realizes minute-level or even second-level ultrafast, high-credible online deduction and early warning of the operation condition and risk trajectory of the power system under extreme weather, and significantly enhances the resilience, operation safety and decision-making intelligent level of the power grid under extreme weather.
Need to check novelty before this filing date? Find Prior Art