A method for predicting NOx emission concentration in SCR system based on time delay estimation

Abstract

The invention discloses a method for predicting NOx emission concentration in a SCR system based on time delay prediction, which comprises the following steps: determining the input variables of a NOxemission concentration prediction model by analyzing the flue gas generation of a coal-fired unit and the mechanism of the SCR system; collecting and preprocessing the running data of correlation variables, estimating the time delay and reconstructing the sample phase space by using correlation coefficient iterative method; on the reconstructed samples, using the kernel partial least square method used to establish the dynamic model. The NOx concentration value of the dynamic model correction output is fed back to the controller in advance to improve the existing ammonia injection control system. The invention has the advantages that the prediction model comprehensively learns the relevant information of the NOx concentration at the outlet; the kernel partial least square method is used to improve the prediction ability. The NOx concentration at the outlet can be predicted in advance by reconstructing the phase space of the model sample. If there is a big difference between the modelsample and the set value, the model sample can be adjusted by ammonia injection in time, which has a guiding significance for reducing the pollutant emission and cost of coal-fired units.

Description

technical field [0001] The invention relates to the technical field of flue gas denitrification of coal-fired units, in particular, to an SCR system NO x Emission Concentration Prediction Methods. Background technique [0002] With the increasingly prominent problem of air pollution, the emission of air pollutants from thermal power units in my country has been strictly regulated. traditional low NO x Combustion control is difficult to meet the emission standards, and it needs to cooperate with the tail flue gas denitrification device. SCR flue gas denitrification technology has been widely used at home and abroad, and the most important thing is the control of the ammonia injection system. The amount of ammonia injection can usually achieve better results through control under stable working conditions, but the control effect can be achieved when the unit operating conditions change, such as coal type replacement, unit AGC (Automatic Generation Control, automatic generati...

AI Technical Summary

Problems solved by technology

[0004] The object of the present invention is to provide a kind of SCR system NO based on delay estimation x Emission concentration prediction method to solve the complex reaction process and NO x There is lag and inaccuracy in concentration measurement, which improves the control effect of ammonia injection

Images