基于有向图的传统村落建筑群火灾蔓延风险分析

在传统村落开发保护过程中，建筑消防改造及建筑布局改变等会导致火灾风险发生变化. 火灾蔓延模拟工作量大、成本高及基础资料获取困难. 为克服这些难点，首先，将单体建筑视为有向图的节点、建筑间的火灾蔓延关系视为节点间的边，通过火灾蔓延模拟判断节点周边局部蔓延路径，并建立有向图的邻接矩阵，利用有向图遍历算法确定特定火灾场景下的蔓延范围，给出考虑多场景建筑群火灾蔓延的损失期望；然后，通过增删节点或边修正有向图邻接矩阵，反映建筑改造、增建或树木对建筑群火灾蔓延关系的变化，重新应用节点遍历算法确定变化后的建筑群火灾蔓延风险；最后，以某一传统村落为例验证方法的有效性. 研究结果表明:本文方法可快速计算动态变化的建筑群火灾蔓延风险，减少火灾蔓延再模拟的工作量；在火灾蔓延防控上，对危险性较高的6个节点进行消防加强改造，可使建筑群火灾蔓延的最大损失及损失期望分别降低61.9%、52.1%；在建筑规划或管理上，应避免新增建筑、树木等可燃物位于建筑密集区域或连通多个建筑密集区域的“桥节点”处. 

Directed Graph Based Risk Analysis of Fire Spread in Buildings in Chinese Traditional Villages

In the development and protection of traditional villages, building renovation and layout change for fire prevention may cause the changes in fire risk. Fire spread simulation is marked by large workloads, high costs and difficulty in acquiring basic data. To cope with these, first, the single building is regarded as the node of the directed graph, and the fire spread relationship between buildings as the edge between nodes. The local spread path around the node is judged through the fire spread simulation, the adjacency matrix of the directed graph is established, the spread range under a specific fire scene is determined by the traversing directed graph algorithm, and the loss expectation for the multiple fire spread scenarios is presented. The adjacency matrix of the directed graph is modified by adding and deleting nodes or edges to reflect the changes of building renovation and addition and trees on the fire spread relationships, and the traversing algorithm of nodes is reused to determine the fire spread risk of the buildings. Finally, a traditional village is used as an example to verify the proposed method. The results show that the proposed method can quickly calculate the fire spread risk of the dynamically changing village buildings and reduces the workload of fire spread re-simulation. In terms of fire spread prevention and control, the maximum loss and loss expectation of fire spread between buildings can be respectively reduced by 61.9% and 52.1% through improving fire preventions of 6 highly-risky nodes. For building planning or management, it is necessary to avoid the case of new buildings, trees and other combustible materials locating in densely built areas or link-nodes connecting multiple densely built areas.