基于Pareto蚁群算法的双目标路径规划研究

Research on double objective path planning based on Pareto ant colony algorithm

针对复杂建筑环境人员应急疏散单一路径不能满足火灾环境变化需求的问题,基于改进蚁群算法,结合Pareto双目标解集思想,提出一种组合优化解集的双目标蚁群算法,通过排序优化的思想,实现人员多路径动态疏散规划。在构造Pareto解集的阶段协同考虑疏散路径长度以及火灾风险程度2个优化目标,计算各个解之间的支配关系。利用排序优化蚁群算法的正反馈机制将各组解的信息素按一定比例作为最优路径信息素的积累,加快解集的寻找。最后将其与传统双目标蚁群算法相比较,结果表明:优化后的双目标算法更加适合复杂建筑人员疏散路径规划问题,在寻找多组满足要求解的同时展示目标之间的利弊关系,供决策者选择合适的路径,提高疏散效率。

Aiming at the problem that the single path of emergency evacuation of personnel in complex built environment cannot meet the changing requirements of fire environment,based on the improved ant colony algorithm and the idea of Pareto multi-objective solution set,a multi-path ant colony algorithm with combined optimization solution set was proposed.Through the idea of sequencing optimization,the dynamic evacuation planning of multi-path personnel was realized.In the stage of constructing the Pareto solution set,the two optimization objectives of evacuation path length and fire risk degree were considered together,and the dominant relationship between each solution was calculated.The positive feedback mechanism of the ranking optimization ant colony algorithm was used to accumulate the pheromones of each group of solutions as the optimal path in a certain proportion,which speeds up the search for the solution set.Compared with the traditional multi-objective ant colony algorithm,the results show that the optimized algorithm can find multiple sets of solutions that meet the requirements,which demonstrates the advantages and disadvantages between the objectives while searching for multiple sets of solutions to satisfy the requirements for the decision maker to choose the appropriate path and improve the evacuation efficiency.