一种基于LSTM的燃气管道施工破坏风险预测模型
|
摘要
随着城市建设步伐的加快,大型施工增多,施工破坏已经成为燃气管道安全越来越不可忽视的问题。传统的燃气管道风险评估方法存在主观性太强和数据采集难度大两个主要问题,本文通过分析燃气管道施工破坏的特点,发现其在时间维度上表现出特有的规律,并从时间维度提取出一套实用的燃气管道施工破坏风险因子,然后基于大数据,使用LSTM建立燃气管道施工破坏可能性的预测模型,最后使用燃气管道安全管理公司提供的数据对模型进行有效性验证。
With the rapid development of large-scale construction in cities,construction damage has become an increasingly important issue for gas pipeline safety. Traditional assessment methods for gas pipeline risk have two main problems: strong subjectivity and difficulty in data collection. This paper analyzes the characteristics of construction damage in gas pipelines and finds that it shows a unique law in the time dimension. We extract a set of risk factors of construction damage from the time dimension. Then based on big data,we use LSTM to establish a prediction model for the failure probability of gas pipelines caused by construction damage. Finally,we use the data provided by gas pipeline safety management company to verify the validity of the model.
With the rapid development of large-scale construction in cities,construction damage has become an increasingly important issue for gas pipeline safety. Traditional assessment methods for gas pipeline risk have two main problems: strong subjectivity and difficulty in data collection. This paper analyzes the characteristics of construction damage in gas pipelines and finds that it shows a unique law in the time dimension. We extract a set of risk factors of construction damage from the time dimension. Then based on big data,we use LSTM to establish a prediction model for the failure probability of gas pipelines caused by construction damage. Finally,we use the data provided by gas pipeline safety management company to verify the validity of the model.
引文
[1]W. Kent Muhlbauer.管道风险管理手册[M].北京:中国石化出版社,2005.
[2]李栋学,刘茂,刘付衍华.基于多层次可拓评价法的城市燃气管线风险评价研究[J].防灾减灾工程学报,2010,30(1):92-97.
[3]刘红芳,刘成敏,王海宁.我国天然气风险及评价方法分析[J].中国安全生产科学技术,2014(2):86-92.
[4]赵欧,张鹏.长输天然气与城市燃气管道风险评价指标体系比较分析[J].管道技术与设备,2012(1):12-14.
[5]徐火力.城市埋地钢制燃气管道风险评价技术研究及应用[J].质量技术监督研究,2015(1):8-12.
[6]顾晓婷,王秋妍,孙萍萍,等.大数据在管道安全评估中的应用[J].消防科学与技术,2017(3):398-401.
[7]韩小明,苗绘,王哲.基于大数据和神经网络的管道完整性预测方法[J].油气储运,2015,34(10):1042-1046.
[8]袁鹏斌,刘凤艳,舒江,等.大数据时代油气管道的安全与防护[J].无损检测,2015,37(4):51-55.
[9]冯庆善.基于大数据条件下的管道风险评估方法思考[J].油气储运,2014,33(5):457-461.
[10]张永明.大数据思维模式下的地下燃气管道风险等级评估[J].城市燃气,2016(3):30-36.
[11]孙伟.城市燃气设施第三方施工破坏的脆性源分析与评价研究[D].天津:天津大学,2014.
[12]周立国,姚安林,蒋宏业,等.城镇燃气管道第三方施工损伤风险评价方法研究[J].中国安全生产科学技术,2015(12):123-128.
[13]Hochreiter S,Schmidhuber J. Long short-term memory[J]. Neural computation,1997,9(8):1735-1780.
[14]Bertolami R,Bunke H,Fernandez S,et al. A Novel Connectionist System for Improved Unconstrained Handwriting Recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(5).
[2]李栋学,刘茂,刘付衍华.基于多层次可拓评价法的城市燃气管线风险评价研究[J].防灾减灾工程学报,2010,30(1):92-97.
[3]刘红芳,刘成敏,王海宁.我国天然气风险及评价方法分析[J].中国安全生产科学技术,2014(2):86-92.
[4]赵欧,张鹏.长输天然气与城市燃气管道风险评价指标体系比较分析[J].管道技术与设备,2012(1):12-14.
[5]徐火力.城市埋地钢制燃气管道风险评价技术研究及应用[J].质量技术监督研究,2015(1):8-12.
[6]顾晓婷,王秋妍,孙萍萍,等.大数据在管道安全评估中的应用[J].消防科学与技术,2017(3):398-401.
[7]韩小明,苗绘,王哲.基于大数据和神经网络的管道完整性预测方法[J].油气储运,2015,34(10):1042-1046.
[8]袁鹏斌,刘凤艳,舒江,等.大数据时代油气管道的安全与防护[J].无损检测,2015,37(4):51-55.
[9]冯庆善.基于大数据条件下的管道风险评估方法思考[J].油气储运,2014,33(5):457-461.
[10]张永明.大数据思维模式下的地下燃气管道风险等级评估[J].城市燃气,2016(3):30-36.
[11]孙伟.城市燃气设施第三方施工破坏的脆性源分析与评价研究[D].天津:天津大学,2014.
[12]周立国,姚安林,蒋宏业,等.城镇燃气管道第三方施工损伤风险评价方法研究[J].中国安全生产科学技术,2015(12):123-128.
[13]Hochreiter S,Schmidhuber J. Long short-term memory[J]. Neural computation,1997,9(8):1735-1780.
[14]Bertolami R,Bunke H,Fernandez S,et al. A Novel Connectionist System for Improved Unconstrained Handwriting Recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(5).