Cleaner production for continuous digester processes based on hybrid Pareto genetic algorithm

Pulping production process produces a large amount of wastewater and pollutant emitted, which has become one of the main pollution sources in pulp and paper industry. To solve this problem, it is necessary to implement cleaner production by using modeling and optimization technology. This paper studies the modeling and multi\|objective genetic algorithms for continuous digester process. First, model is established, in which environmental pollution and saving energy factors are considered. Then hybrid genetic algorithm based on Pareto stratum\|niche count is designed for finding near\|Pareto or Pareto optimal solutions in the problem and a new genetic evaluation and selection mechanism is proposed. Finally using the real data from a pulp mill shows the results of computer simulation. Through comparing with the practical curve of digester,this method can reduce the pollutant effectively and increase the profit while keeping the pulp quality unchanged.

面向汽车投产排序的混合多目标网格遗传算法

汽车投产排序时,希望同时实现零部件消耗均衡化、车型调整费用最小化、工位作业位置精准化三个目标,为此提出一种基于Pareto层级的混合多目标网格遗传算法(HmoGA)。先将个体排斥机制加入到Pareto层级构造中,使非支配解的分布更均匀,再融合Pareto层级划分、网格拥挤度评价与相邻个体几何距离计算,设计一种多目标自适应网格选择机制,用于从动态变化的父代种群中选择较优个体构成进化种群、获取交叉运算的父代基因、改善非支配解集的分布质量。混合双基因位的迁移算子对非支配解进行邻域搜索,适时扩大搜索空间,跳出局部最优。利用三组不同规模的测试问题集,从非支配率、非支配解数量和相邻个体距离偏差三个指标方面进行比较,实验证明HmoGA算法在收敛性、解的数量和分布性方面都比NSGA-Ⅱ算法有显著优势。