A prediction model optimization method for scr denitrification system based on machine learning

Abstract

The present invention provides a machine learning-based SCR denitrification system prediction model optimization method, comprising the following steps: Step S1: collecting real-time sample data of the NOx concentration at the boiler outlet in the SCR denitrification system and related indicators affecting the NOx concentration; Step S2: Use principal component analysis to perform dimension reduction processing; Step S3: Establish a support vector machine model; Step S4: Introduce an exponential decay model to iteratively update the step size value of the beetle whisker algorithm, and optimize the vector machine parameters; Step S5: Support vector machine simulation; Step S6: Repeat steps S1-S5. The invention provides a prediction model optimization method for SCR denitrification system based on machine learning, which solves the problem that it is difficult to realize the precise control of ammonia injection in existing thermal power plants. The invention reduces the sample data based on principal component analysis (PCA). Dimensional processing, through the introduction of exponential decay model iteratively update the step value, improve the beetle beetle algorithm (BAS) optimization to obtain the optimal support vector machine model parameters, and establish an optimized support vector machine regression (SVM) model.

Description

technical field [0001] The invention is a machine learning-based SCR denitrification system prediction model optimization method, which relates to the fields of geotechnical engineering and tunnel engineering, and specifically relates to the fields of real-time monitoring and forecasting of foundation pit excavation deformation and stability analysis. Background technique [0002] More than half of my country's nitrogen oxide (NOx) emissions come from coal combustion, and one of the key industries for NOx pollution control is the thermal power industry. Selective Catalytic Reduction technology (Selective Catalytic Reduction, SCR) is under the condition of a catalyst, select the flue downstream of the boiler with a flue temperature of 300 ~ 4000 ℃, inject the reducing agent into the NOx reaction in the flue gas, and reduce the NOx , reduced to pollution-free N2 and H20. The SCR denitrification system has become an important equipment for large thermal power units to achieve ...

AI Technical Summary

Problems solved by technology

[0007] The technical problem to be solved by the present invention is to provide a machine learning-based SCR denitrification system prediction model optimization method to solve the problem that it is difficult to realize the precise control of ammonia injection in existing thermal power plants. The present invention is based on the principal component analysis method ( PCA) performs dimensionality reduction processing on the sample data, iteratively updates the step size value by introducing an exponential decay model, improves the beetle whisker algorithm (BAS) to obtain the optimal support vector machine model parameters, and establishes an optimized support vector machine regression (SVM) Model

Images