Search |

.

The European Transmission Network (ETN) has been evolving significantly in the ways that network operators work and cooperate with each other, in particularly by setting up common operating procedures and tools such as network simulation tools . In the domain of power system state estimation and transmission network time simulation, very few innovations have emerged over the past decade (2000-2010) in Europe. Breakthroughs are therefore expected in power system simulation along four directions to serve the ETN: state estimation, optimization algorithms, time domain simulation and power system component modelling.

The PEGASE project aimed to remove several algorithmic barriers related to the monitoring, simulation and optimization of very large power systems, therefore paving the way for improved network operation.

Optimal reactive power and voltage control in electric networks require the solution of an optimization problem for the 24 h of the day before the dispatch day, for instance to minimize the daily system losses, while limiting the number of Load Tap Changers (LTC) and capacitor bank hourly switching operations. This optimization problem is basically a non-convex, mixed-integer non-linear programming (MINLP) problem, due to the discrete nature of the LTC and capacitor bank switching, as well as the nonlinear power flow constraints. A review on existing true mixed-integer non-linear programming [1] (MINLP) solvers [2] showed first that available solutions are not capable of working within reasonable computation times for the problems at stake for TSOs: this was confirmed by tests performed on the most well-known MINLP solvers.

A prototype optimization solver software has therefore been developed: it accounts for several different objective functions (like minimal deviation of generated active powers, active losses), while being able of taking into account the discrete behavior of equipment such as the taps of on-load tap-changing (OLTC) [3] transformers or Phase-Shifting Transformers (PST) , the switching process of capacitors/reactors banks, the shutdown/startup process of generators for very large systems like the pan European Transmission Network, but without contingency constraints. Moreover, reducing the complexity of the optimization problem requires an approach involving a sequence of optimization problems. It relaxes at each step integrality constraint for some discrete variables and combines with the use of a solution based on mathematical programming with equilibrium constraints [4] (MPEC): this consists in substituting the discrete variables by continuous variables with complementary constraints, which does not strictly guarantee optimum values for discrete variables, but least provides a feasible solution, “improved” values of discrete variables, and a local optimum for continuous variables.

Thus building realistic anticipated states of the pan-European grid can lean on such new optimization approaches. For such a large size system, solutions are proposed to deal with a large number of contingencies, but without discrete variables. For discrete variables, modelling the discrete behavior of several power devices is proposed, but optimization is addressed without any contingency.

According to the PEGASE Consortium Agreement, the prototype software is owned 100% by RTE, which will lead the Discrete Variable Optimization Software Prototype commercialization.

[1] See for instance www.gams.com

[2] See for instance “MINLP Solver Software” Michael R. Bussieck and Stefan Vigerske February 21, 2012

[3] For many power transformer applications, a supply interruption during a tap change is unacceptable: the transformer is then fitted with a more expensive and complex on-load tap-changing mechanism. On-load tap changers may be generally classified as either mechanical, electronically assisted, or fully electronic

[4] See for instance Z.-Q. Luo, J.-S. Pang and D. Ralph: Mathematical Programs with Equilibrium Constraints. Cambridge University Press, 1996

Add comments

Belonging to the same Cluster(s)

Page:

Share | . . . |