基于改进NSGA2-FLC的饱和含水燃油多目标优化控制

Multi-Objective Optimal Control of Water-Saturated Fuel Based on Improved NSGA-II-FLC

中文关键词:  改进第2代非支配遗传算法  模糊逻辑控制器  饱和含水燃油  多目标优化  第3方导热油

英文关键词:improved non-dominated sorting genetic algorithm-II  fuzzy logic controller  water-saturated fuel  multi-objective optimization  intermediate medium heat transfer oil

中文摘要:

      为了解决在燃油结冰适航试验饱和含水燃油制备中，燃油与导热油之间温差过大易导致热交换器壁面结冰问题，提出 一种将改进第2代非支配排序多目标优化算法（NSGA2）最优个体输入模糊控制器（FLC）进行动态优化的多目标优化方法，实现对 饱和含水燃油温度的精确控制。基于热液和两相流管路模型搭建含第3方导热油的饱和含水燃油温控系统，结合试验温控系统 验证仿真模型的可靠性，对不同温度工况下的系统进行仿真分析。优化在控制中定义的系统温度调节时间，并以温差控制的适应 度值为决策目标，引进佳点集法初始化种群并在迭代过程中采用变惯性权重的柯西变异（CM）法寻优，将改进NSGA2寻优获得的 全局最优控制参数输入模糊控制器中并自整定控制器中参数的变化率，实现对温控模型的动态优化。结果表明：改进NSGA2- FLC多目标优化控制能缩短调节时间，提高控温精度约50%，使温差接近且不超过13 ℃，满足适航要求。

英文摘要:

      To solve the problem of heat exchanger wall icing due to excessive temperature difference between fuel and heat transfer oil in the preparation of water-saturated fuel for fuel icing airworthiness tests, a multi-objective optimization method was proposed to improve the dynamic optimization of Non-dominated Sorting Genetic Algorithm-II (NSGA-II) optimal individual input fuzzy logic control? ler to achieve accurate temperature control of the water-saturated fuel. Based on the thermal hydraulicand two-phase pipe flow model, a water-saturated fuel temperature control system with intermediate medium heat transfer oil was built, and the reliability of the simulation model was verified with the test temperature control system, and the system under different temperature conditions was simulated and analyzed. The system temperature settling time defined in the control algorithm was optimized, and the fitness value of temperature differ? ence control was used as the decision objective, the Good Point Set method was introduced to initialize the population, and the Cauchy Mutation (CM) method with variable inertia weight was used to optimize the iterative process. The global optimal control parameters obtained from the improved NSGA-II optimization were fed into the fuzzy logic controller, and the Slope of the parameters in the controller were self-tuned to realize the dynamic optimization of the temperature control model. The results show that the improved NSGA-II-FLC multi-objective optimal control can shorten the temperature settling time, improve the temperature control accuracy by about 50%, and make the temperature difference close to but not exceeding 13℃, which meets the airworthiness requirements.