基于模糊识别的建筑物火灾危险性评价方法

Method of Fuzzy Recognition in Risk Assessment of Building Fire

笔者以传统的经典统计和模糊统计为基础 ,依据建筑物火灾危险性的影响因素 ,应用模糊识别理论及系统安全方法 ,建立了建筑物火灾危险性的评价指标体系 ;并应用集值统计方法对各项指标权重进行了处理。该方法改变了过去统计给出一个固定值的做法 ,而给出一个区间值 ,解决了对专家给出的指标“大约是多少”这一概念的定量化。为了验证了评价模型的准确性 ,将该理论应用到某购物商场火灾危险性评价中 ,取得了满意效果 ,为建筑物防火设计以及安全管理提供了可行的依据。

Referring to influential factors of the building fire risk and based on conventional fuzzy statistics and classical statistics,an index system for assessing building fire risk was established by applying the fuzzy recognition theory and system safety method. The weights of various indexes were calculated with the set-valued statistics. This method changed some fixed values from classical statistics to bound values,which altered some indexes provided by experts from estimated values to quantified values. In order to verify this method,it was applied to a supermarket with satisfactory result providing some feasible criteria for safety design and management of building fire.