Research on Optimization and Adjustment of Gas Turbine Combustion Based on XGBoost and NSGA-II

In view of the lack of experience in the combustion adjustment of gas turbines in China, a model of gas turbine combustion optimization is proposed, which can ensure that the gas turbine runs in a safe and stable area with low NOx emission. Based on the real-time operating data of the gas turbine, the NOx emission and pressure fluctuation prediction models of XGBoost are established respectively. In order to make the prediction model more accurate, the Bayesian optimization algorithm is used to optimize the model parameters of XGBoost; then NSGA-II optimization algorithm is used to solve the combustion optimization problem of gas turbine, so as to obtain the Pareto solution set of the gas turbine running in a safe, stable and low-emission environment. The experimental results show that the XGBoost combined with the NSGA-II optimization model is used for multi-objective optimization of gas turbine combustion optimization problem. The optimal operating parameter setting combination obtained from Pareto solution concentration has certain reference significance for the stable operation of gas turbine and reducing NOx emission, and accumulated experience for domestic combustion adjustment technology.