亚临界燃煤机组协调控制模型辨识方法及系统
By combining recurrent neural networks with big data optimization, the problem of inaccurate model parameter acquisition in AGC mode for subcritical coal-fired units was solved, achieving accurate identification of the coordinated control model, reducing on-site commissioning time, and improving work efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING GUODIAN NANZI WEIMEIDE AUTOMATION CO LTD
- Filing Date
- 2025-07-21
- Publication Date
- 2026-07-17
AI Technical Summary
Under dual-carbon conditions, subcritical coal-fired units cannot switch to manual mode for open-loop testing in AGC mode, resulting in inaccurate engineering calculations and increasing the debugging time for on-site optimization control functions.
By combining recurrent neural networks with big data optimization, a recurrent neural network model is established by collecting target operating parameter data of subcritical coal-fired power units. The optimization algorithm is then used to calculate the dynamic relationships between different variables in the controlled object model of the coordinated control system, thereby achieving accurate model identification.
It reduced the debugging time for engineers on site, improved work efficiency, solved the problem of inaccurate model parameter acquisition in AGC mode, and realized precise coordinated control of subcritical coal-fired units.
Smart Images
Figure CN120871612B_ABST
Abstract
Citation Information
Patent Citations
Method for identifying unit main steam temperature controlled process transfer function model
CN106054609A
Distributed coordination and control system for thermoelectric generating set based on multi-parameter dynamic matrix control
WO2019047561A1