Reservoir scheduling method based on GRU

Through the GRU-based reservoir scheduling method, the GRU model is constructed and adjusted using historical data, and combined with the support vector regression model to predict outbound flow, the problem of insufficient reservoir scheduling accuracy and real-time in the existing technology is solved, and high precision and High-responsive reservoir dispatch.

CN119398420AActive Publication Date: 2025-02-07CHINA INST OF WATER RESOURCES & HYDROPOWER RES
View PDF 3 Cites 1 Cited by

Patent Information

Application Number
CN202411471688.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-22
Publication Date
2025-02-07
Estimated Expiration
2044-10-22

AI Technical Summary

Technical Problem

Due to the uncertainty and nonlinear characteristics of the incoming flow, the existing reservoir scheduling technology makes it difficult to accurately simulate the outgoing flow, the scheduling accuracy is not high, the real-time performance is poor, and the risks are high.

Method used

Using the GRU-based reservoir scheduling method, the GRU model is constructed and error training is performed by obtaining reservoir historical data, combining the support vector regression model to predict the outbound flow, and adjust the GRU parameters through the Nash efficiency coefficient and generalization coefficient to improve the model's Accuracy and response speed.

Benefits of technology

Real-time prediction and automatic control of reservoir outflow flow is realized, scheduling accuracy and response speed are improved, and scheduling risks are reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure FT_1
    Figure FT_1
  • Figure FT_2
    Figure FT_2
  • Figure SMS_19
    Figure SMS_19
Patent Text Reader

Abstract

The invention relates to a GRU-based reservoir scheduling method, which is characterized in that a reservoir scheduling model is constructed based on a GRU, the output flow of a reservoir can be predicted in real time, so that a discharge gate of the reservoir can be automatically controlled in real time, and when the reservoir scheduling model is constructed, automatic adjustment of GRU parameters can be realized through set evaluation indexes, so that the working efficiency of the reservoir scheduling model is improved. Therefore, the response speed and the scheduling precision of the reservoir scheduling model for implementing scheduling meet the set requirements, and the scheduling risk is reduced.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Method and system for realizing water level early warning based on clustering and GRU

    CN112561203A

  • Drainage pipe network sewage flow prediction method based on space-time diagram convolutional network

    CN115526428A

  • Hybrid reservoir dispatching simulation method based on mechanism and data driving

    CN117634288A

Cited By

  • Water regimen forecasting and early warning method for pumped storage power station

    CN121075097A