论文部分内容阅读
从提高染色产品质量和效益的角度出发,综合考虑如染料浓度、温度、时间和助剂浓度等因素影响,构建了多目标染色工艺配方优化模型。针对传统遗传算法普遍存在的问题和缺陷,提出基于正交试验设计、自适应交叉操作及局部搜索等技术进行算法改进,并利用改进后的算法获得配方模型最优解的解决方法:。实践结果:证明,混合自适应遗传算法使种群更具有代表性和全面性,最大程度的继承了父代的优良特性,改善了算法的早熟现象并增强其寻优性能。最终以较少的计算量和较高的收敛速度对全局进行快速的搜索,比传统遗传算法得到的优化目标值降低了l0.8%左右。该方法:可推广应用于其他复杂过程的优化求解问题中。
From the perspective of improving the quality and efficiency of dyeing products, a multi-objective dyeing process formulation optimization model was constructed by comprehensively considering such factors as dye concentration, temperature, time and concentration of additives. Aiming at the common problems and defects of traditional genetic algorithm, this paper proposes the improvement of the algorithm based on orthogonal experimental design, adaptive crossover operation and local search techniques. The improved algorithm is used to obtain the optimal solution of the formulation model. The results of practice: Prove that hybrid genetic algorithm can make the population more representative and comprehensive, inherit the excellent characteristics of the parent to the greatest extent, improve the prematurity of the algorithm and enhance its performance. In the end, the global search is faster with less computational complexity and higher convergence rate, which is about 10% lower than that of the traditional genetic algorithm. This method can be generalized to the optimization problem of other complex processes.