A method for predicting AGC regulation performance of a coal-fired unit

Abstract

This invention discloses a method for predicting the AGC (Automatic Generative Control) performance of a coal-fired power unit. The method includes: selecting historical operating data of the coal-fired power unit under stable operating conditions within a preset time period based on historical operating parameters; identifying various AGC adjustment processes without frequency regulation actions based on the historical operating data, and obtaining the AGC performance index corresponding to each AGC adjustment process; inputting the AGC performance index and the corresponding historical operating parameters of each AGC adjustment process as sample data into a preset BP (Backpropagation) neural network for training; and combining the current operating parameters under stable operating conditions with the BP neural network to obtain the predicted result of the AGC performance of the coal-fired power unit. Using this invention, the AGC performance of the unit under different operating conditions can be dynamically monitored, and the AGC performance can be accurately predicted in advance using a neural network.

Images