中国安全科学学报 ›› 2017, Vol. 27 ›› Issue (9): 116-121.doi: 10.16265/j.cnki.issn1003-3033.2017.09.020

• 安全工程技术科学 • 上一篇    下一篇

基于BP算法的模糊Petri网HSE绩效评价方法

韩婧, 梁伟** 教授, 张来斌 教授   

  1. 中国石油大学(北京) 机械与储运工程学院,北京 102249
  • 收稿日期:2017-06-19 修回日期:2017-08-08 出版日期:2017-09-20 发布日期:2020-11-16
  • 通讯作者: **梁伟(1978—),男,陕西绥德人,博士,教授,博士生导师,主要从事油气管道安全监测、站场设备诊断与可靠性评估等方面的研究。E-mail: lw@cup.edu.cn。
  • 作者简介:韩 婧 (1992—),女,河北张家口人,硕士研究生,研究方向为安全评价技术。E-mail:cup_hanjing@163.com。
  • 基金资助:
    中国石油大学(北京)科研基金资助(2462015YQ0406)。

Performance evaluation of HSE with fuzzy Petri nets based on BP algorithm

HAN Jing, LIANG Wei, ZHANG Laibin   

  1. College of Mechanical and Transportation Engineering,China University of Petroleum-Beijing,Beijing 102249,China
  • Received:2017-06-19 Revised:2017-08-08 Online:2017-09-20 Published:2020-11-16

摘要: 为得到更加客观、有效的健康、安全和环境(HSE)绩效评价结果,增强安全管理效果,建立基于BP算法的模糊Petri网(FPN)HSE绩效评价模型。首先结合HSE体系要素层次结构,建立HSE绩效评价FPN模型;然后利用BP误差反传计算一阶梯度的特性对FPN推理过程中的变迁置信度和初始库所可信度进行调节,避免输入参数的主观性;其次输入相关参数,经过模糊推理,得到评价结果;最后以中石油某子公司为例,验证模型的有效性。结果表明:BPFPN模型适用于石油企业的HSE绩效评价,可以为HSE管理提供决策支持。

关键词: 健康、安全和环境(HSE)绩效评价, 模糊Petri网(FPN), BP算法, 变迁置信度, 库所可信度

Abstract: In order to obtain more objective and effective results of HSE performance evaluation and enhance the management effect, a FPN HSE performance evaluation model based on BP algorithm was built. Firstly, before building the model, a FPN model of HSE performance was built by combining the HSE elements hierarchy. Then the BP error back propagation algorithm was used to adjust the confidence of the transitions and the reliability of the initial places in the FPN reasoning process to avoid the subjectivity of input parameters. The evaluation results can be obtained by inputting the relevant parameters and fuzzy reasoning. Finally the effectiveness of BPFPN model was verified by applying it to a certain petrolic company as an example. The results show that the BPFPN model is suitable for HSE performance evaluation and can provide decision support for HSE management.

Key words: health, safety and environment(HSE) performance evaluation, fuzzy Petri nets(FPN), back propagation(BP), transitions confidence, places reliability

中图分类号: