Linearization of an MINLP model and its application to gas distribution optimization. (1st March 2018)

• Record Type: Journal Article
• Authors: Mikolajková, Markéta; Saxén, Henrik; Pettersson, Frank
• Is Part Of: Energy. Volume 146(2018)
• Page Start: 156
• Page End: 168
• Publication Date: 2018-03-01
• Subjects: Optimization -- MILP -- Linearization -- Gas supply chain -- Natural gas -- LNG; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2017.05.185
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Natural gas is currently regarded as a "bridge fuel", a more environmental-friendly alternative to fossil fuels such as oil or coal while in the transition towards the near-zero emission future. New gas sources allow using natural gas more extensively. Growing popularity of LNG, biogas or SNG might cause an increase in the natural gas consumption, especially in the power production or in transportation. Transportation of such gas fuels should be considered when designing the new extensions of the pipeline networks. Mathematical modelling and optimization are important tools used to solve the complex problem of a gas distribution optimization. The objective of this work is the development and application of a linearized MILP (mixed integer linear programming) formulation of a natural gas distribution network while covering all the technical and operational aspects such as the mass flows, pressures, compressor duties or changes of the mass flow directions. The multi-period formulation allows studying the influence of the seasonal changes and demand fluctuations on the optimal network structure. Use of the model is demonstrated by optimizing the existing Finnish natural gas network and its prospective extensions. The results provide a detailed insight into the interdependencies of the features within a gas network. Highlights: Gas networks are optimized by mixed integer linear programming (MILP). Multi-period formulation considers operational and technical aspects appropriately. The model considers the possibility to change the gas flow direction. Integration of gas sources and network extensions are demonstrated to be feasible.