SCI和EI收录∣中国化工学会会刊

中国化学工程学报 ›› 2021, Vol. 33 ›› Issue (5): 211-220.DOI: 10.1016/j.cjche.2020.12.022

• Process Systems Engineering and Process Safety • 上一篇    下一篇

Neural network-based source tracking of chemical leaks with obstacles

Qiaoyi Xu1, Wenli Du1,2, Jinjin Xu1, Jikai Dong1   

  1. 1 Key Laboratory of Advanced Control and Optimization for Chemical Processes, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China;
    2 Shanghai Institute of Intelligent Science and Technology, Tongji University, Shanghai 200092, China
  • 收稿日期:2020-04-12 修回日期:2020-11-02 出版日期:2021-05-28 发布日期:2021-08-19
  • 通讯作者: Wenli Du
  • 基金资助:
    The work was supported by the National Natural Science Foundation of China (Basic Science Center Program:61988101; 21706069), Natural Science Foundation of Shanghai (17ZR1406800) and National Science Fund for Distinguished Young Scholars (61725301).

Neural network-based source tracking of chemical leaks with obstacles

Qiaoyi Xu1, Wenli Du1,2, Jinjin Xu1, Jikai Dong1   

  1. 1 Key Laboratory of Advanced Control and Optimization for Chemical Processes, Ministry of Education, East China University of Science and Technology, Shanghai 200237, China;
    2 Shanghai Institute of Intelligent Science and Technology, Tongji University, Shanghai 200092, China
  • Received:2020-04-12 Revised:2020-11-02 Online:2021-05-28 Published:2021-08-19
  • Contact: Wenli Du
  • Supported by:
    The work was supported by the National Natural Science Foundation of China (Basic Science Center Program:61988101; 21706069), Natural Science Foundation of Shanghai (17ZR1406800) and National Science Fund for Distinguished Young Scholars (61725301).

摘要: The leakage of hazardous gases poses a significant threat to public security and causes environmental damage. The effective and accurate source term estimation (STE) is necessary when a leakage accident occurs. However, most research generally assumes that no obstacles exist near the leak source, which is inappropriate in practical applications. To solve this problem, we propose two different frameworks to emphasize STE with obstacles based on artificial neural network (ANN) and convolutional neural network (CNN). Firstly, we build a CFD model to simulate the gas diffusion in obstacle scenarios and construct a benchmark dataset. Secondly, we define the structure of ANN by searching, then predict the concentration distribution of gas using the searched model, and optimize source term parameters by particle swarm optimization (PSO) with well-performed cost functions. Thirdly, we propose a one-step STE method based on CNN, which establishes a link between the concentration distribution and the location of obstacles. Finally, we propose a novel data processing method to process sensor data, which maps the concentration information into feature channels. The comprehensive experiments illustrate the performance and efficiency of the proposed methods.

关键词: Obstacle, Optimization, Neural networks, Feature extraction, Source term estimation, Computational fluid dynamics (CFD)

Abstract: The leakage of hazardous gases poses a significant threat to public security and causes environmental damage. The effective and accurate source term estimation (STE) is necessary when a leakage accident occurs. However, most research generally assumes that no obstacles exist near the leak source, which is inappropriate in practical applications. To solve this problem, we propose two different frameworks to emphasize STE with obstacles based on artificial neural network (ANN) and convolutional neural network (CNN). Firstly, we build a CFD model to simulate the gas diffusion in obstacle scenarios and construct a benchmark dataset. Secondly, we define the structure of ANN by searching, then predict the concentration distribution of gas using the searched model, and optimize source term parameters by particle swarm optimization (PSO) with well-performed cost functions. Thirdly, we propose a one-step STE method based on CNN, which establishes a link between the concentration distribution and the location of obstacles. Finally, we propose a novel data processing method to process sensor data, which maps the concentration information into feature channels. The comprehensive experiments illustrate the performance and efficiency of the proposed methods.

Key words: Obstacle, Optimization, Neural networks, Feature extraction, Source term estimation, Computational fluid dynamics (CFD)