基于网格化Logistic回归模型的区域突发环境事件应急决策方法

针对突发环境事件不确定性较高,应急决策方案复杂,实行成本较高的问题,提出一种基于网格化Logistic回归模型的区域突发环境事件应急决策方法。利用德尔菲方法选取区域突发环境事件影响变量,以变量为输入,构建基于Logistic回归模型的区域突发环境事件发生概率计算模型,将研究区域划分为单元格形式,划分区域等级,明确安全区范围。在多个约束条件下,利用蚁群算法求取目标函数,得出最优人员疏散路径,完成应急决策方案设计。实验结果表明: 该方法能够有效应对突发环境事件,且从建模至完成人员疏散的时间约0.5 h,总耗时较短,实现了区域突发环境事件应急决策,从而保障了人员和财产安全。研究对区域突发环境事件的防控具有指导意义。

Emergency decision method of regional environmental emergencies based on gridding logistic regression model

In view of the high uncertainty, complex emergency decision-making scheme and high implementation cost of environmental emergencies, the authors in this research proposed a emergency decision method for regional environmental emergencies based on grid logistic regression model. The Delphi method is used to select the impact variables of regional environmental emergencies, and a probability calculation model of regional environmental emergencies based on logistic regression model is constructed, with the variables as input. The study area was divided into cell forms, and the regional grades were divided to define the scope of the safety area. Under multiple constraints, the objective function was obtained by using the ant colony algorithm, and the optimal evacuation route was obtained to complete the emergency decision scheme design. The experimental results show that this research method is effective for environmental emergencies, and the duration between model establishment and people evacuation is about half an hour. This short time-consuming means the emergency decision is effective for regional environmental emergencies, which would ensure the safety of personnel and property. This research has guiding significance for the prevention and control of regional environmental emergencies.