Oxygen content prediction method based on real-time measurement data

Abstract

The invention discloses an oxygen content prediction method based on real-time measurement data. The oxygen content prediction method comprises the following steps: (1) additionally arranging a grid multi-point oxygen content measurement device behind a flue in-situ oxygen content measurement point; (2) collecting operation data of the power station boiler; (3) extracting a part of data as a training sample of the artificial neural network, and taking the remaining data as a verification sample; (4) the prediction time advance is set to be t0, and training is carried out with the unit loads, the total coal feeding amount, the total air volume and the flue gas flow sample data at the multiple different moments t as input and the grid multi-point oxygen amount sample data at the moment t + t0 as prediction output; (5) testing the trained artificial neural network through a verification sample; and (6) embedding the verified artificial neural network serving as an oxygen compensation prediction model into a measurement system to be responsible for oxygen prediction. The method has the advantages of accuracy of actual oxygen amount hard measurement and soft measurement prediction, and can better meet the requirement of oxygen amount auxiliary adjustment in automatic air supply.

Description

technical field [0001] The invention relates to a method for predicting oxygen content based on real-time measurement data, belonging to the field of flue gas oxygen content monitoring. Background technique [0002] At present, thermal power still occupies a dominant position in my country's power generation system. The optimization direction of thermal power generation is mainly to improve the combustion efficiency of the boiler and reduce the emission of pollutants. The oxygen content of power plant flue gas is a parameter directly related to boiler efficiency and pollutant emission measurement, and its measurement accuracy plays an important role in guiding production and operation. [0003] If the measured oxygen content of the flue gas is too high, it indicates that the ratio of the air volume to the fuel volume is too high, the heat loss of the exhaust gas will increase, the efficiency of the boiler will decrease, and the generation of nitrogen oxides will increase, w...

AI Technical Summary

Problems solved by technology

The disadvantage is that it is necessary to know the low-level calorific value of coal-fired coal, and it is also required that the state parameters of the boiler should remain unchanged as much as possible during the operation period. For example, the air leakage cannot change. If there is a change, the soft measurement cannot obtain accurate measurement results.

Images