Short-term stochastic forecasting of the dynamics of gas flows and reserves in the Unified Gas Supply System of Russia based on recurrent neural networks

Author(s):  
N.A. Kislenko ◽  
◽  
A.V. Belinsky ◽  
A.S. Kazak ◽  
O.I. Belinskaya ◽  
...  
2020 ◽  
Vol 219 ◽  
pp. 02001
Author(s):  
Nikolay Ilkevich ◽  
Tatyana Dzyubina ◽  
Zhanna Kalinina

This paper proposes taking into account new properties of gas supply systems in a mathematical model of flow distribution in comparison with the traditional formulation. The approach suggests introducing an arc coefficient, which allows for changes in the magnitude of gas flow passing along the arc, a vector of an increase in the arc throughput, and lower constraints on the gas flow along the arc. We also propose considering a new economic environment, namely, new criteria for optimizing the flow distribution and setting fictitious gas prices for consumers. These criteria enable us to take account of the priority gas supply to a definite group of consumers. As an example, the calculation of gas flows for the aggregated Unified Gas Supply System (UGSS) for 2030 is considered. This calculation takes into account the arc coefficients and the increase in the throughput of arcs.


Sign in / Sign up

Export Citation Format

Share Document