Modeling and Optimization of Boiler Low $\text{No}_{\mathrm{x}}$ Combustion Based on Adaptive Genetic Algorithm and Support Vector Machine

To optimize the control of boiler <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{NO}_{\mathbf{x}}$</tex> emission, the field operation data was analyzed, and the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{NO}_{\mathbf{x}}$</tex> emission characteristic model based on support vector machine was established. An improved adaptive genetic algorithm was used to optimize the model parameters, and the performance of the support vector machine model based on Sigmoid kernel function and RBF kernel function was compared. The results show that RBF kernel function has more advantages, and support vector machine algorithm has good generalization ability and prediction accuracy. By combining support vector machine modeling and adaptive genetic algorithm optimization, the adjustable parameters of the boiler were optimized, and the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{NO}_{\mathrm{x}}$</tex> emission of the furnace was reduced from <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{423}\mathbf{mg}/\mathbf{Nm}^{\mathbf{3}}$</tex> to <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{219.4}\mathbf{mg}/\mathbf{Nm}^{\mathbf{3}}$</tex>, with a decrease of 48.13%. The above method obtains the optimal operation parameter combination of low <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{NO}_{\mathbf{x}}$</tex> emission, which provides reference and basis for the operation optimization guidance and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{NO}_{\mathbf{x}}$</tex> emission control of power plant boiler.