A Recommendation Method and System for Feeder-Level Demand Response Mechanisms in Distribution Networks Based on Bayesian Classification

By generating feeder-level random scenario data through Bayesian classification and Monte Carlo simulation, and combining an improved Bayesian classifier to evaluate flexibility, load guidelines are constructed, and the scheduling of photovoltaic, energy storage, and electric vehicles is optimized. This solves the supply-demand imbalance and loss problems between distribution network feeders, and achieves precise demand response and cost reduction.

CN122136929APending Publication Date: 2026-06-02SOUTHEAST UNIV +1
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SOUTHEAST UNIV
Filing Date
2026-01-06
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Under conditions of high proportion of distributed photovoltaic and electric vehicle access, existing distribution network demand response methods have failed to effectively solve the problems of supply and demand imbalance between feeders, line congestion, localization of curtailment, and increased feeder losses, and lack feeder-level collaborative optimization and flexibility assessment.

Method used

A Bayesian classification-based approach is adopted to generate feeder-level random scenario data through Monte Carlo simulation, extract feature attributes, evaluate feeder flexibility using an improved Bayesian classifier, construct feeder-level load guidelines, optimize the coordinated scheduling of photovoltaic, energy storage and electric vehicles, and recommend personalized demand response mechanisms.

Benefits of technology

It achieves feeder power balance, reduces losses, enhances photovoltaic absorption capacity, reduces distribution network operating costs, and improves the accuracy and flexibility of the response mechanism.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122136929A_ABST
    Figure CN122136929A_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for recommending demand response mechanisms at the feeder level in distribution networks based on Bayesian classification. The method includes: Step 1, randomly acquiring hourly output data of distributed photovoltaic (PV) power and electric vehicle (EV) behavior data from each feeder in the distribution network using the Monte Carlo method, and selecting typical scenario data; Step 2, extracting feeder-level feature attribute data from the typical scenario data, using the feature attribute data as input conditions, and having an improved Bayesian classifier output feeder flexibility category labels, recommending feeder-level demand response mechanisms, including price-based demand response and quasi-linear demand response, to the distribution network operator based on the flexibility category labels; Step 3, constructing feeder-level load quasi-linear variables; Step 4, performing PV-energy storage-EV collaborative optimization scheduling. This invention achieves refined management of the distribution network and collaborative optimization of PV-energy storage-charging systems by intelligently assessing feeder flexibility and constructing feeder-level load quasi-linear variables.
Need to check novelty before this filing date? Find Prior Art