Kurzbeschreibung
(Englisch)
|
The proposed project deals with the development of a multi-level hierarchical control structure for wind energy generation. Starting from the low level of controlling individual wind turbines we will build up to the wind farm coordination and control level, and finally to the integration of wind farms into the overall power grid (in terms of stability and pricing). The main objective is to provide a step towards the reliable and uninterrupted operation of power networks, one of the most vital infrastructures of the modern society. Hence the outcome of this study, which will involve both theoretical and simulation based analysis, will directly support the objectives of the IntelliCIS COST Action; the results will be integrated into the Inte-liCIS workprogram through the participation of the two Principal Investigators in the COST action acitvities.
|
Partner und Internationale Organisationen
(Englisch)
|
BE, BG, CH, CY, CZ, DE, DK, EL, ES, FI, FR, HR, HU, IE, IL, IT, LT, LV, NL, NO, PL, PT, RO, RS, SE, SI, TR, UK
|
Abstract
(Englisch)
|
Wind energy occupies a prominent position among renewable energy sources, and continuously gains significance as governments worldwide strive to reduce the environmental footprint in the energy sector. We propose a control architecture that starts from the low level of individual wind turbines, builds up to the level of coordination and wind farm control, and finally investigates the integration of wind farms to the power network. The latter highlights the necessity of revisiting security and scheduling problems, which in turn give rise to stability and pricing issues. This leads to one of the greatest challenges in the power grid arena from the point of view of the utilities and the regulators. The objective of our work is to provide a flexible framework for modeling and control of distributed generation with focus on wind farms. We start by developing a model for each wind turbine, which is rich enough to capture the different modes of operation, and can be also integrated in the control strategies of each level of the hierarchy. To achieve this, and since the wind turbine involves both continuous dynamics and discrete modes of operation, we adopt a modeling framework based on hybrid automata. At a next step, we employ this model for control purposes, where a novel algorithm, based on a combination of nonlinear and predictive control techiques, is applied to yield the desired power tracking performance. At the wind farm level, one objective is to quantify the potential improvement in the wind power production, if wind turbines are placed optimally on a given site. Using the developed wind turbine model, this is achieved by means of randomized optimization. Another objective is to improve wind power prediction, by using algorithms based on Monte Carlo methods and taking into account the temporal and spatial correlation of the wind. Finally, at the integration layer, we mainly concentrate on the problem of computing a secure dispatch for the generating units, and determining the minimum cost reserves so as to satisfy a given demand level, while avoiding undesirable load shedding and wind generation spillage. To achive this, we treat the problem in a probabilistic framework, which results in a chance constrained optimization program. Using then efficient constraint-sampling techniques, allows us to extend our approach for large scale networks.
|
