Application of optimization algorithms to production management: Removing infrastructure constraints

Introduction. The task of improving the efficiency of late-stage field development is an urgent challenge, requiring a balance between costs and current production levels. It is proposed to use a genetic optimization algorithm to identify wells whose shutdown will allow to obtain an increase in oil production and (or) reduce operating costs

Aim. Reducing optimization costs of oil fields through production reallocation decisions using optimization algorithms

Materials and methods. The approach of simplified modelling of the system ‘reservoir — well — gathering system’ with the use of an optimization algorithm, the target function of which is daily oil production was implemented. In this work, an algorithm for preprocessing and analyzing the input data was used, as well as the tornado algorithm for sensitivity analysis. The study, modelling and optimization were performed based on a proxy integrated asset model adapted to the actual data.

Results. The result of approbation of the approach is a program of measures for stopping and implementation of hydraulic fracturing at the producing stock, corresponding to the maximum value of the selected target function.

Conclusion. It was determined that the main difficulties in using the genetic algorithm for optimization are the high dimensionality of the problem and the topology of the acquisition network. To eliminate these difficulties, a well filter was created and a sector model of the collection network was used. As a result, the genetic algorithm switched off  the most watered wells of the fund as the best option, which is a criterion of the optimizer’s correctness.

1. Gazizov T.T. Global Optimization Methods: Textbook, Tomsk: V-Spektr Publishing House, 2017. — 22 p.

2. Emelyanov V.V., Kureichik V.V., Kureichik V.M. Theory and Practice of Evolutionary Modeling. — Moscow: Fizmatlit, 2003

3. Gladkov L.A., Kureichik V.V., Kureichik V.M., Kureichik V.V. Bio-inspired methods in optimization: monograph. — Moscow: Fizmatlit, 2009

4. Akhtyamov O. V. Evaluation of the genetic algorithm effi ciency when changing the problem dimensionality Reshetnev Readings. — Krasnoyarsk: Siberian State Aerospace University named a` er academician M.F. Reshetnev, 2010.

5. User Manual for tNavigator. Adaptation and Optimization. Rock Flow Dynamics, 2024