Neural network modeling of air pollution in tunnels according to indirect measurements

2016 Joint IMEKO TC1-TC7-TC13 Symposium: Metrology Across the Sciences: Wishful Thinking?

Kaverzneva, T.^1 ; Lazovskaya, T.^1 ; Tarkhov, D.^1 ; Vasilyev, A.^1

Peter the Great St-Petersburg Polytechnic University, 29 Politechnicheskaya Str, Saint-Petersburg

195251, Russia^1

关键词: Approximate solution;    Identification problem;    Indirect measurements;    Initial boundary problems;    Mass transfer process;    Neural network model;    Problem regularization;    Ventilation systems;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/772/1/012035/pdf
DOI  :  10.1088/1742-6596/772/1/012035

来源: IOP

【 摘 要 】

The article deals with the problem of providing the necessary parameters of air of the working area in dead-end tunnels in the case of ventilation systems powered off. An ill-posed initialboundary problem for the diffusion equation is used as a mathematical model for a description and analysis of mass transfer processes in the tunnel. The neural network approach is applied to construct an approximate solution (regularization) of the identification problem in the case of the approximate measurement data and the set of interval parameters of the modeled system. Two types of model measurements included binary data are considered. The direct problem solution and the inverse problem regularization for the offered neural network approach are constructed uniformly.