建筑火灾全局动态疏散路径规划研究  

Research on global dynamic path planning for evacuation during building fires

作　　者：李明海[1] 兰亚乐 马骁[1] 何鑫 杨一帆[1] LI Minghai;LAN Yale;MA Xiao;HE Xin;YANG Yifan(College of Resources Engineering,Xi'an University of Architecture and Technology,Xi'an 710055,China)

机构地区：[1]西安建筑科技大学资源工程学院,西安710055

出　　处：《安全与环境学报》2025年第1期205-215,共11页Journal of Safety and Environment

基　　金：陕西省“2024年重点研发计划”项目(2024SF-ZDCYL-05-16)。

摘　　要：传统A^(*)算法被广泛应用于路径规划研究中,但该算法在处理复杂环境时存在搜索效率低和寻优路径质量不高的问题。为克服这些问题,提出了一种改进A^(*)算法,该算法结合了启发式搜索与实时动态规划的思想,能在保留A^(*)算法优势的同时显著提升其搜索效率和路径质量。在改进算法中,设计了一种新型启发式函数,该函数不仅考虑了火灾场景下的危险因素,还引入了实时动态规划策略以引导搜索过程,从而生成更高效的疏散路径。将改进算法与原始算法进行性能对比测试以及建筑火灾模拟疏散仿真对比试验,以验证改进算法的寻优性能。对比测试和试验结果表明,改进A^(*)算法在提高路径规划效率方面具有显著优势。与传统A^(*)算法相比,改进A^(*)算法生成的应急疏散路径中拐点数量少,扩展节点的数量减少96.49%,路径计算速度提升95.68%。验证了改进A^(*)算法在复杂场景下的优越性能,表明改进A^(*)算法在实际应用中具有广阔的前景。To address the issues of low search efficiency and poor path quality associated with the traditional A^(*)search algorithm in complex environments,this paper proposes an improved A^(*)algorithm that integrates heuristic search with real-time dynamic programming.This enhancement aims to boost both search efficiency and path quality while preserving the strengths of the original A^(*)algorithm.It also considers the practicality of the algorithm in specific fire scenarios.Building on the traditional A^(*)algorithm,the improved A^(*)algorithm introduces a new heuristic function that integrates the Jump Point Search(JPS)strategy,node update rules,and an optimization method for setting danger zones.Utilizing real-time fire detection and personnel location data transmitted by building sensors,the improved A^(*)algorithm dynamically updates fire scene information by integrating it with FDS fire simulation data.It employs a multi-head attention mechanism to assess the hazardous conditions of the fire scene and incorporates a real-time dynamic planning strategy to guide the search process.The improved A^(*)algorithm significantly enhances the efficiency and safety of evacuation paths.Compared to the traditional A^(*)algorithm,the results demonstrate clear advantages in path planning efficiency,with a 95.68%reduction in search time,a 96.49%decrease in the number of extended nodes,and a 36.84%reduction in the number of inflection points.The experimental comparison of application cases clearly demonstrates the superiority of the proposed algorithm.The results indicate that,compared to the traditional A^(*)algorithm,the proposed algorithm reduces the number of extended nodes by approximately 76%and increases the speed of acquiring path distance by about 85%.The results of the comparative test between the A^(*)algorithm and the improved A^(*)algorithm demonstrate that the improved A^(*)algorithm offers significant advantages in complex fire scenarios.

关 键 词：安全工程 A^(*)算法 启发式搜索 动态规划 火灾场景 

分 类 号：X93[环境科学与工程—安全科学]