Transient stability prevention-emergency coordinated control embedded with assessment model of light gradient boosting machine
Author:
Affiliation:

Clc Number:

TM712

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    To bring the improvement of the transient stability of the power grid by the operation mode control and generator tripping control into full play,a power system transient stability prevention-emergency coordinated control decision-making method embedded in the light gradient boosting machine (LightGBM) assessment model is proposed. In order to quickly evaluate the degree of improvement of the system stability margin by the control measures,firstly,the hybrid control sample generation method and the LightGBM algorithm is used to construct a assessment model of the prevention-emergency control on the system stability margin. Considering that the unreasonable generator tripping and load shedding control may deteriorate the system stability. The numerical sensitivity of the LightGBM surrogate model is used to identify effective control sites and reduce the decision space. The LightGBM model is further embedded in the transient stability two-layer optimal control model,which replaces the transient stability time-domain simulation,combines the improved non-dominated sorting genetic algorithm-Ⅱ (NSGA-Ⅱ) to realize the rapid solution of the coordinated control strategy. Through the IEEE 39-node test example,it is verified that the proposed method can realize the coordination and cooperation between the preventive control before the occurrence of faults and the emergency control after the occurrence of faults of different severity,improve the security and stability of the power grid, reduce the cost of optimal dispatching.

    Reference
    Related
    Cited by
Get Citation
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:March 16,2023
  • Revised:May 23,2023
  • Adopted:November 21,2022
  • Online: October 10,2023
  • Published: September 28,2023