A power distribution network load decomposition prediction method and system based on a physical information neural network

By constructing a physical information neural network in the distribution network and combining the constraints of power grid physical laws and meteorological data, the problem of low load restoration accuracy caused by the lack of distributed power source data was solved, and high-precision load decomposition and restoration were achieved.

CN122456479APending Publication Date: 2026-07-24GUANGDONG DIANWANG GONGSI YUNFU POWER SUPPLY BUREAU
0 Cites 0 Cited by

Patent Information

Application Number
CN202610903859.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-23
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

In the absence of distributed generation data, traditional methods cannot effectively utilize meteorological data to reasonably constrain and infer the output of distributed generation, resulting in low accuracy of distribution network load restoration.

Method used

A load decomposition prediction method based on physical information neural networks is constructed. By embedding power grid physical laws and meteorological data, physical constraints are constructed using power grid physical laws such as node power balance and meteorological data, generating electrical and meteorological causal consistency constraints, thereby improving the load decomposition accuracy in scenarios with missing data.

Benefits of technology

In the absence of distributed power generation output data, the accuracy of load decomposition was significantly improved, ensuring that the model output conforms to the physical laws of the power grid and meteorological causal relationships, and improving the accuracy of natural load reconstruction.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses a power distribution network load decomposition prediction method and system based on a physical information neural network, and relates to the technical field of load prediction. The method comprises the following steps: acquiring measurement data of each monitoring node of a target power distribution network, including total injected power time series data, part of known distributed power output time series data, meteorological time series data and power grid static topology parameters; constructing a physical information neural network model, including decomposing the total injected power into a data-driven subnetwork of natural load and equivalent contribution of distributed power, and a physical constraint subnetwork which is mapped to the physical space of the power grid and calculates a physical loss term; training by using part of the known distributed power data as labels and combining the physical loss term to construct a joint loss function; and inputting real-time measurement data into the trained model to output natural load time series data after stripping the influence of distributed power. The application ensures that the model output is physically interpretable and verifies, and significantly improves the load decomposition accuracy in the data missing scenario.
Need to check novelty before this filing date? Find Prior Art