基于正态云-投影寻踪模型的火灾风险评估

Fire safety risk assessment using normal cloud-projection pursuit model

针对地下商场的火灾风险进行科学评估,提出正态云、投影寻踪模型相结合的方法。使用正态云模型确定其特征参数,得到商场火灾风险的隶属度值,并结合投影寻踪模型得到风险的最佳投影方向值,最终得到评估数值与排序,指出地下商场火灾风险水平。结果表明:用实际可得的数据进行科学处理,避免了主观评估方法因素的干扰,验证了该方法的可行性,为地下商场火灾风险评估提供了新的评估方法。

Aiming at the scientific evaluation of fire risk in underground shopping malls, this paper puts forward a method combining normal cloud model with projection tracking. The normal cloud model is used to identify the characteristic parameters and thus membership value of the fire risk in malls can be obtained. Combining with the projection tracking model, the best projection direction was gotten, so that the numerical evaluation and the sequence can finally be gained. And there upon this paper would point out the level of fire risk of the underground shopping malls.Results showed that the use of actual data available for scientific processing has prevented the interference of subjective evaluation, and has verified the feasibility of the method. Also, it has provided a new evaluation method for the fire risk assessment of underground shopping malls.