Multi-objective performance optimization and control for gas turbine Part-load operation Energy-saving and NOx emission reduction
Yujia Ma,Jinfu Liu,Linhai Zhu,Qi Li,Yaqiong Guo,Huanpeng Liu,Daren Yu
DOI: https://doi.org/10.1016/j.apenergy.2022.119296
IF: 11.2
2022-08-15
Applied Energy
Abstract:This work aims to apply the multi-objective optimization method to gas turbine part-load energy-saving and NOx (Nitrogen Oxides) emission reduction problem. For power generation gas turbines, energy-saving is based on raising the thermal efficiency of the system. This is always based on regulating the variable geometries of the plant. However, in this process, the combustion condition would be changed, which would further change the NOx pollutant emission level. Therefore, in this paper, the conflict between gas turbine thermal efficiency enhancement and NOx emission reduction is discovered and analyzed with a nonlinear model. To solve this multi-objective optimization problem, GA (Genetic Algorithm), a global optimization algorithm, is applied. The two objectives are thermal efficiency for energy-saving and NOx Emission Index (EINOx) for pollution reduction. The fuel mass flow rate and compressor Inlet Guide Vane (IGV) position are the decision variables. Optimization is conducted for 50% to 90% nominal power levels and 5 °C to 25 °C ambient temperatures. With the non-inferior solutions on the Pareto Front and a trade-off decision, the selected gas turbine optimum working point can be obtained with a nonlinear converter module. For 15 °C, 70% nominal power working condition, increasing the fuel mass flow rate by 0.216% and IGV degree by 10.344% can bring a 46.273% reduction of EINOx, with the cost of a 0.212% decline in the thermal efficiency. The sensitivity analysis of the Pareto Frontiers to power level and ambient temperature is carried out. Then the level of energy-saving is analyzed by calculating the cost of fuel, and a trade-off suggestion is given. In the end, the values of the decision variables and objective functions under three trade-off scenarios are calculated and listed in a table, with which the final solution can be decided conveniently according to practical demands.
energy & fuels,engineering, chemical