This is the home page for projects in the School of Mathematics involving
the application of optimization and operational research methods to energy problems.
We organise talks which are a mixture of reading seminars where some interesting
published work with be discussed, and informal lead discussion of
ongoing work, and more formal seminars.
See Talks for details and background material
The Operations Research Group (ORG) in the School of Mathematics currently has a number of research projects related
to Optimization problems in operation of Power System networks. In detail these are
 Preventing widearea blackouts through adaptive islanding of transmission networks: This is an EPSRC
funded project (reference EP/G060169/1) in collaboration with the Institute for Energy Systems in the Edinburgh
University School of Engineering, the Power Systems Engineering group at Durham University and the Analytic &
Geometric Methods in Group Theory group of Southampton University. The purpose of the project is to explore if
cascading widearea power systems blackouts can be prevented by purposely disconnecting the power systems into
distinct stable islands at short notice. The role of the ORG is in the formulation of suitable Optimization models
and the development of numerical algorithms to solve these models in a short time frame.
 Stochastic Unit Commitment with network constraints: Unit Commitment is concerned with identifying optimal
heating up/cooling down decisions for power generation plants in order to meet future anticipated demand at minimum
cost and robustly. Emphasis is on the inclusion of significant stochasticity (due to increasing wind generation)
and accurately modelling network constraints.

Specialist solvers for Security Constrained Optimal Power Flow: Any generation schedule for a power system
must be robust against sudden line failures. For large systems the sheer number of possible failure events that
ought to be considered poses problems for current solution techniques. This project aims to develop improved
algorithm that exploit the structure of these problems in the linear algebra of the optimization solver and also by
early identification of crucial failure scenarios. Collaboration with Power Systems Engineers from the University of
Durham.
 Risk Based Optimal Power Flow (OPF): This project investigates risk based OPF as an alternative
to the standard security constrained OPF. A extension to the PSAT dynamic simulator has been implemented to
investigate cascading failures, and this can be used as a tool to calibrate the risk terms in the OPF
 Building a library of electricity system test cases: We are assembling test cases and data for different
electricity system optimization problems. See below for current status. We plan to make those that are nonconfidential
available on the web.
 Building a library of verified local optima of OPF problems:
There are no well documented published cases of local optima in OPF
problems which are within realistic voltage limits. We have identified several examples and
are making the data for these available on the web (see section below).
source of
Electricity Network Data Sets

We are gathered data for a wide range of electricity power
system test networks. A summary of the systems is here
Summary of power system data sets
Local Optima in Optimal Power Flow problems

We have found cases of local optima in for the Optimal Power Flow Problem:
Data for some of these cases is given here:
Local Optima Archive
