Generative deep learning for decision making in gas networks

DOI10.1007/S00186-022-00777-XMaRDI QIDQ2155386zbMATHOpenAlexFDO

Authors Lovis Anderson, Thorsten Koch, Mark G. Turner

Publication date 15 July 2022

Published in Mathematical Methods of Operations Research (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2102.02125

zbMATH Keywords

deep learningmixed integer programmingprimal heuristicgas networksgenerative modelling

Mathematics Subject Classification ID

Deterministic network models in operations research (90B10) Mixed integer programming (90C11)

Abstract: A decision support system relies on frequent re-solving of similar problem instances. While the general structure remains the same in corresponding applications, the input parameters are updated on a regular basis. We propose a generative neural network design for learning integer decision variables of mixed-integer linear programming (MILP) formulations of these problems. We utilise a deep neural network discriminator and a MILP solver as our oracle to train our generative neural network. In this article, we present the results of our design applied to the transient gas optimisation problem. With the trained network we produce a feasible solution in 2.5s, use it as a warm-start solution, and thereby decrease global optimal solution solve time by 60.5%.

Recommendations

• A deep learning model for gas storage optimization
• Physics informed neural networks: a case study for gas transport problems
• Generative Adversarial Network for Probabilistic Forecast of Random Dynamical Systems
• Deep learning in turbulent convection networks
• Embedding decision diagrams into generative adversarial networks
• Convolutional -- recurrent neural network proxy for robust optimization and closed-loop reservoir management
• Deep Generative Models and Inverse Problems

Cites work

• scientific article; zbMATH DE number 7124428 (Why is no real title available?)
• A decision support system for a real vehicle routing problem.
• A decision support system for cyclic master surgery scheduling with multiple objectives
• A mixed integer approach for the transient case of gas network optimization.
• Deep learning
• Foundations of bilevel programming
• Local branching
• Machine learning for combinatorial optimization: a methodological tour d'horizon
• Maximizing the storage capacity of gas networks: a global MINLP approach
• Optimal operation of transient gas transport networks
• Pyomo -- optimization modeling in Python
• The voice of optimization

Cited in

(4)

• Efficient simulation of coupled gas and power networks under uncertain demands
• A deep learning model for gas storage optimization
• Enhancing cut selection through reinforcement learning
• Generating neural archetypes to instruct fast and interpretable decisions

Describes a project that uses

Uses Software

• SCIP
• Gurobi
• Pyomo
• Inception-v4

This page was built for publication: Generative deep learning for decision making in gas networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2155386)