论文部分内容阅读
为解决工程应用中的多目标优化问题,提出一种双层多目标遗传算法(two-layer multi-objective genetic algorithm,TLMOGA)。该算法根据个体间的支配关系将种群分成2层,并分别采用快速k最邻近算法和净强度函数法为这2层中的个体分配适应度。在此基础上,设计相应的个体排序和种群修剪策略,并确定了算法的整体流程。通过与传统多目标遗传算法进行比较,证明TLMOGA能够很好地保持解的收敛性和分布性,同时也具有较高的运算效率。最后,以ALSTOM气化炉基准控制器的参数优化整定为工程应用实例,进一步验证TLMOGA的有效性。仿真试验的结果表明,经优化后的控制系统,控制品质有了显著提高,达到了ALSTOM气化炉基准测试的要求。
To solve the multi-objective optimization problem in engineering application, a two-layer multi-objective genetic algorithm (TLMOGA) is proposed. The algorithm divides the population into two layers according to the dominance relation among the individuals, and assigns the fitness to individuals in the two layers using fast k-nearest neighbor algorithm and net intensity function method respectively. On this basis, the corresponding individual ranking and population pruning strategy are designed and the overall process of the algorithm is determined. Compared with the traditional multi-objective genetic algorithm, it proves that TLMOGA can well maintain the convergence and distribution of the solution, and also has a high computational efficiency. Finally, the parameter optimization of ALSTOM gasifier reference controller is set as an example of engineering application to further verify the effectiveness of TLMOGA. The simulation results show that the optimized control system has significantly improved the control quality and met the ALSTOM gasifier benchmark test requirements.