摘要
重大危险源事故风险预警技术是事故控制技术中的一大研究重点。本文通过对比分析安全生产领域常用事故预警模型的优缺点,总结了事故预警模型的发展趋势,提出了基于模糊综合评判技术和动态模糊神经网络技术的重大危险源事故风险预警模型,详细陈述其构建过程,最后对液化石油气蒸气云爆炸事故应用该模型,得到了较好的预测效果,实现了风险程度的定量化预警。该预警模型具有快速的自学习能力和容错能力,能够同时处理多种风险因素、自动生成模糊规则并满足预警系统的实时性要求,可应用于重大危险源在线监控预警系统,为企业端和政府端监管者提供有效的重大危险源事故风险定量化预警信息,为遏制重大灾难事故的发生、减少死亡人、数受伤人数和直接经济损失提供先进的理论和技术支撑。
The pre-warning technology for accidents risk of major hazards is a major research focus of accident control technologies.By analyzing advantages and disadvantages of common pre-warning models used in the field of safety production,this paper summarized the development trend of pre-warning models.Then,it puts forward a new pre-warning model based on fuzzy comprehensive evaluation technology and dynamic fuzzy neural network technology,and states its building process.At last,this model is applied to LPG vapor cloud explosion accident,a good predictive validity can be gained,and risk can be well predicted in the form of quantization.This model has rapid learning ability and fault-tolerant ability,can deal with multiple risk factors simultaneously,generate fuzzy rules automatically,and meet the real time requirement of early warning system.It can be applied to major hazards monitoring and early-warning systems,provide effective quantitative accident risk for regulators of government and enterprises,and provide advanced theoretical and technological support to prevent major disasters from occurring and to reduce the number of deaths and injuries and direct economic losses.
出处
《中国安全生产科学技术》
CAS
北大核心
2010年第2期44-50,共7页
Journal of Safety Science and Technology
基金
广东省科技计划项目(编号:2008A030203002)
关键词
安全
重大危险源
风险预警
动态模糊神经网络
蒸气云爆炸
safety
major hazards
risk pre-warning
dynamic fuzzy neural network
vapor cloud explosion
