论文部分内容阅读
为突破遗传算法(GA)在危险化学品泄漏事故应用中存在早熟收敛等不足,有针对性地引入淘汰者基因库,提高种群多样性,避免算法过早陷入局部极值。同时,借鉴粒子群算法的跟随思想,引入启发信息,强化收敛域内的局部搜索力度,最终整理得到改进型遗传算法(MGA)。统计结果表明,MGA的计算结果更准确,误差适应性更强,可为泄漏事故现场的应急决策提供快速有效的数据支持。
In order to break through the genetic algorithm (GA), there are some problems such as premature convergence in the application of dangerous chemical spill. Targeted introduction of knockout gene bank, increasing population diversity and avoiding premature falling into local extreme. At the same time, referring to the following idea of particle swarm optimization, we introduce heuristic information to enhance the local search in the convergence domain, finally finishing the improved genetic algorithm (MGA). The statistical results show that the MGA results are more accurate and error-adaptive, which can provide fast and effective data support for emergency decision-making on the scene of a spill accident.