Improving TSO's coordination for cross-border redispatch: a discussion of possible approaches in the European context

The paper first recalls the main features of the European market design. Next, all the present adjustment practices are defined and analysed in order to establish a basis for the evaluation of redispatching both in terms of adjustment mechanism and in terms of economic efficiency. Starting from the current European situation where almost all congestion management balancing measures are carried out individually by each area, even if some countermeasures as counter trading are already implemented in several areas, the paper points out the benefits brought by the extension of the coordination of redispatch measures to make a better use of the overall resources and thus to lower the cost of curative actions. This principle needs to be studied in-depth between TSOs, and this paper analyses the problems which must be overcome in order to improve, if necessary, the redispatch coordination. It proposes for this a possible implementation which has to be further discussed among TSOs and compared to other possible solutions. More precisely, the implementation difficulties of an extended redispatch coordination are analysed in terms of data which need to be exchanged between TSOs, optimality and calculation burden and an alternative is proposed to the fully decentralized redispatch. The main advantage is that it requires only the exchange of a small amount of adjustment bids which result from a local selection in each participating zone. Moreover, iterations are not necessary, since the solution calculation is carried out only by the TSO which has a congestion to solve, but the other participating TSOs may check whether the proposed solution is adequate in their own zones and, in consequence, to validate or not their participation to the redispatch. This alternative can be used with a small loss of the optimality of the fully centralized coordination since the participating bids can be a priori selected according to their influence on the physical flows of the congested flowgates. A simplified 15 buses, 4 zones of an academic European model is used to discuss and test those redispatch strategies