Wind farm power production and fatigue load optimization based on dynamic partitioning and wake redirection of wind turbines

Due to the wake effect, there is a high degree of coupling between wind turbines in a wind farm. In order to optimize the power and load performance of wind farms, it is crucial to propose an optimization strategy to attenuate the wake interference between wind turbines by wake redirection. This study proposes a wake interference model that can consider wake deflection based on the yawed wake model. Accordingly, the interference relationship between turbines is represented as the topology of the graph. By solving for the connected components of the graph, the wind farm is divided into almost uncoupled partitions. In each partition, separate optimization problems with the objectives of power enhancement and load reduction are established. An improved Non-Dominated Sorting Genetic Algorithm with multiple crossover operators is used for solving the multi-objective optimization problem. Simulation results show that the proposed strategy can effectively lead to power enhancement and load reduction for the regular layout wind farm. For the curvilinear layout wind farm with an optimized layout considering wake effects, proposed strategy can further reduce wake disturbances to improve the performance of wind farms. •Establish a wake interference model among wind turbines considering yaw angle.•Dynamic partitioning of wind turbines via clustering nodes of undirected graph.•Optimize power production and fatigue load based on wind farm wake redirection.