Evaluation of the impact of active wake control techniques on ultimate loads for a 10 MW wind turbine

Wind farm control is one of the solutions recently proposed to increase the overall energy production of a wind power plant. A generic wind farm control is typically synthesized so as to optimize the energy production of the entire wind farm by reducing the detrimental effects due to wake–turbine interactions. As a matter of fact, the performance of a farm control is typically measured by looking at the increase in the power production, properly weighted through the wind statistics. Sometimes, fatigue loads are also considered in the control optimization problem. However, an aspect which is rather overlooked in the literature on this subject is the evaluation of the impact that a farm control law has on the individual wind turbine in terms of maximum loads and dynamic response under extreme conditions. In this work, two promising wind farm controls, based on wake redirection (WR) and dynamic induction control (DIC) strategy, are evaluated at the level of a single front-row wind turbine. To do so, a two-pronged analysis is performed. Firstly, the control techniques are evaluated in terms of the related impact on some specific key performance indicators, with special emphasis on ultimate loads and maximum blade deflection. Secondarily, an optimal blade redesign process is performed with the goal of quantifying the modification in the structure of the blade entailed by a possible increase in ultimate values due to the presence of wind farm control. Such an analysis provides for an important piece of information for assessing the impact of the farm control on the cost-of-energy model.

A Traffic-Aware Federated Imitation Learning Framework for Motion Control at Unsignalized Intersections with Internet of Vehicles

The wealth of data and the enhanced computation capabilities of Internet of Vehicles (IoV) enable the optimized motion control of vehicles passing through an intersection without traffic lights. However, more intersections and demands for privacy protection pose new challenges to motion control optimization. Federated Learning (FL) can protect privacy via model interaction in IoV, but traditional FL methods hardly deal with the transportation issue. To address the aforementioned issue, this study proposes a Traffic-Aware Federated Imitation learning framework for Motion Control (TAFI-MC), consisting of Vehicle Interactors (VIs), Edge Trainers (ETs), and a Cloud Aggregator (CA). An Imitation Learning (IL) algorithm is integrated into TAFI-MC to improve motion control. Furthermore, a loss-aware experience selection strategy is explored to reduce communication overhead between ETs and VIs. The experimental results show that the proposed TAFI-MC outperforms imitated rules in the respect of collision avoidance and driving comfort, and the experience selection strategy can reduce communication overheads while ensuring convergence.