基于LSSVM建模和AFSA算法的燃气轮机燃烧优化

Combustion optimization of gas turbine based on LSSVM model and AFSA algorithm

针对当前我国燃气轮机建模存在的方法匮乏、燃烧调整策略不足等问题,提出了一种燃气轮机建模方法,该方法基于最小二乘法支持向量机(LSSVM),对燃气轮机燃烧室压力波动和NOx排放值进行建模,再通过人工鱼群优化算法,对所建模型进行优化。仿真结果表明,LSSVM方法对燃气轮机燃烧室压力波动和NOx排放值建模误差分别为0.397%和1.142%,人工鱼群算法对模型的优化值分别为0.485%和0.874%,基本满足建模优化要求,为燃气轮机建模与优化调整提供参考。

Due to the shortage of methods and combustion adjustment strategies in gas turbine modeling in China,a gas turbine modeling method is proposed based on the least squares support vector machine(LSSVM).The pressure fluctuation and the NOx emission value of gas turbine combustor are modeled,and the artificial fish swarm optimization algorithm is employed to optimize the model.The simulation results show that the model errors based on the LSSVM method for the pressure fluctuation and the NOx emission value of gas turbine combustion chamber are 0.397%and 1.142%respectively,and the optimized values by the artificial fish swarm algorithm are 0.485%and 0.874%respectively,which basically meet the requirements of model optimization,and provide references for gas turbine modeling and optimization.