Distributive Model of Maximum Permissible Emissions of Enterprises into the Atmosphere and Its Application

All-Russian research-to-practice conference "Ecology and safety in the technosphere: current problems and solutions"

工业技术（总论）

Meshechkin, V.V.^1 ; Bykov, A.A.^2 ; Krutikov, V.N.^1 ; Kagan, E.S.^1

Kemerovo State University, 6, Krasnaya Street, Kemerovo

650000, Russia^1

Institute of Computational Technologies SB RAS, 21 Rukavishnikov Street, Kemerovo

650025, Russia^2

关键词: Direct minimization;    Harmful substances;    Inequality constraint;    Linear programming models;    Linear programming problem;    Minimization methods;    Pollution regulation;    Reducing emissions;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/224/1/012019/pdf
DOI  :  10.1088/1755-1315/224/1/012019

学科分类：工业工程学

来源: IOP

【 摘 要 】

The paper discusses the problem of limiting enterprises' emissions of harmful substances into the atmosphere taking into account the air pollution regulations. The cost model for reducing emissions of individual enterprises is formulated. The distributive model of maximum permissible emissions of enterprises is defined as the problem of minimizing the total cost of reducing emissions, the limitations of which are determined by the requirements for air pollution in residential areas. The disadvantages of the distributive model in the form of a linear programming problem are discussed. The presented model, in contrast to the linear programming model, is more resistant to data inaccuracies and provides a proportional distribution of the volume of permissible emissions for enterprises with identical characteristics. The solution of the direct minimization problem with inequality constraints is irrational due to its high dimension. A method for solving the direct problem based on the duality theory is obtained. As a rule, the dual problem has far less dimension. To solve it, the effective subgradient minimization method with stretching-compression of space is used. The distributive model is tested on real data. The example of solving an applied problem is given.