中国安全科学学报 ›› 2018, Vol. 28 ›› Issue (2): 181-186.doi: 10.16265/j.cnki.issn1003-3033.2018.02.031

• 公共安全 • 上一篇    下一篇

神经网络范式下硐室群施工安全风险预警研究

江新1,2 教授, 杜海文1,2, 袁轩1,2, 胡文佳1,2, 郑霞忠1,2, 孙正熙3   

  1. 1 三峡大学 湖北省水电工程施工与管理重点实验室,湖北 宜昌 443002;
    2 三峡大学 水利与环境学院,湖北 宜昌 443002;
    3 广东省水文局 茂名水文分局,广东 茂名 525000
  • 收稿日期:2017-12-10 修回日期:2018-01-20 出版日期:2018-02-28 发布日期:2020-11-17
  • 作者简介:江新 (1966—),男,安徽寿县人,硕士,教授,主要从事工程项目群管理(安全管理)、系统决策理论及安全评价等研究。
  • 基金资助:
    长江科学院开放研究基金资助 (CKWV2016382/KY);国家自然科学基金资助(51379110);水电工程施工与管理湖北省重点实验室(三峡大学)开放基金资助(2016KSD20)。

Study on early warning of safety risk in underground cavern group construction under neural network paradigm

JIANG Xin1,2, DU Haiwen1,2, YUAN Xuan1,2, HU Wenjia1,2, ZHENG Xiazhong1,2, SUN Zhengxi3   

  1. 1 Hubei Key Laboratory of Construction and Management in Hydropower Engineering,China Three Gorges University,Yichang Hubei 443002,China;
    2 College of Hydraulic Environmental Engineering,China Three Gorges University,Yichang Hubei 443002,China;
    3 Maoming Hydrology Branch,Guangdong Province Hydrographic Bureau,Maoming Guangdong 525000,China
  • Received:2017-12-10 Revised:2018-01-20 Online:2018-02-28 Published:2020-11-17

摘要: 为保障地下硐室群施工人员安全、减少安全事故,针对地下硐室群施工安全风险,利用BP神经网络方法,开展预警研究。首先通过分析地下硐室群施工安全的影响因素,构建预警指标体系;收集1 000组训练样本后,设置并调试网络参数,网络训练成功后,用以进行地下硐室群施工期安全风险预警分析;然后以某地一个在建水电站地下硐室群为例,验证该预警方法的有效性。结果表明:用该预警方法能够快速获知地下硐室群当前施工安全状态以及风险变化趋势,提前采取有效的应对措施,提高施工安全管理水平。

关键词: 地下硐室群, 施工安全, 风险, 预警指标, BP神经网络

Abstract: In order to protect the safety of underground cavern group construction workers and prevent accidents, the BP neural network method was used to study early warning of safety risks in the construction of underground caverns. Firstly, through the analysis of the influencing factors of construction safety of underground caverns, a warning index system was constructed. After collecting 1 000 sets of training samples, the network parameters were set and debugged. After the success of the network training, a safety risk early warning analysis for the underground cavern group during construction was carried out. Then the underground caverns under construction in a certain place were taken as an example to verify the effectiveness of the early warning method. The results show that the early warning method can be used to quickly know the current safety status of and the trend in variation of risk in the underground caverns construction, and is helpful in taking effective measures in advance for improving the safety management level of underground caverns construction.

Key words: underground cavern group, construction safety, risk, early warning indicator, BP neural network

中图分类号: