Gas network optimisation using Nash equilibria derived from dynamic non-cooperative non-zero sum game theory

Ramchandani, N-L

Abstract

Efficient management of the national gas grid in Great Britain involves a complex
optimisation task to resolve the sometimes conflicting requirements of security of
supplies to customers, contractual obligation to suppliers and the cost of transmission,
whilst at all times ensuring the integrity of the system. Existing computer
software aids for this task in routine use within British Gas divide the optimisation
into several parts which are treated separately before the results are brought
together for expert assessment and decision. A single, integrated suite of programs
to address all aspects of the optimisation would have obvious attractions. In this
work we describe a possible framework for such a development, based on a dynamic
non-cooperative game theoretic approach, in which the national transmission system
(NTS) is viewed as a closed trading market for the trade of gas related commodities.
The bulk of the thesis describes how the NTS is decentralized to be conceptually
viewed as a macro economy for the trade of pressure, linepack, transmission
capacity, mass flow rates of gas, and the like. A producer source terminal
oligopoly, trading with a second transmission compressor station oligopoly is
outlined to address the interaction between sourcing and transmission operations.
The consumers network/demand off-takes are shown to be the key elements
of the framework which maintain both the physical and economic integrity of the
NTS.
A conceptual framework is developed to implement the design to form HERO
— an experimental economic decision tool. The conceptual elegance of our game
theoretic approach is reflected in the computational simplicity, and hence efficiency,
associated with the natural computational parallelism and integral discrete and continuous
player decision making behaviour. In addition to implementation details,
results from test applications to small sections of the UK NTS are presented as evidence
of its potential value in allowing grid controllers to directly address network
operations from a strategic and tactical managerial view point.

Citation

Ramchandani, N. Gas network optimisation using Nash equilibria derived from dynamic non-cooperative non-zero sum game theory. (Thesis). University of Salford

• Thesis Type: Thesis
• Deposit Date: Jul 23, 2021
• Award Date: Mar 16, 1992