An Improved Multiobjective Genetic Algorithm in Optimization and its Application to High Efficiency and Low NOx Emissions Combustion
Xianyong Peng, Pei-hong Wang · 2009
To tackle the boiler combustion multiobjective optimization problem, an improved Pareto multiobjective genetic algorithm (IMOGA) is developed based on non-dominated sorting genetic algorithm II (NSGA-II). In proposed algorithm, its population classification mechanism and an outer sets container technique which contributes to maintaining diversity of the solutions and is the merit of SPEA2 is integrated. Studies on high efficiency and low NOxemissions combustion optimization were carried out by a previous purposed model of high efficiency and low NOxemissions and the IMOGA. In the comparison of IMOGA with two-weighted-objective genetic algorithm (TWOGA), the IMOGA shows good results and can find multiple Pareto optimal solutions in one single run. The optimization results obtained by two algorithms shows that they agree well with each other in the trend of optimal solutions and that of the IMOGA is better.