Abstract: The present invention is a wind farm control method, wherein a reinforcement learning method (RL) is implemented in a decentralized manner (for each wind turbine). The reward (REC) is calculated according to a wake propagation delay (DEL). Thus, the reward is truly representative of the effect of the last action (previous yaw control for example).

Abstract: The present invention is a wind farm control method, wherein a reinforcement learning method (RL) is implemented in a decentralized manner (for each wind turbine). The reward (REC) is calculated according to a wake propagation delay (DEL). Thus, the reward is truly representative of the effect of the last action (previous yaw control for example).

Abstract: A system, method and apparatus for providing ancillary services to a power grid using a power consumption component at a customer premises. The apparatus may control a load with discrete, finite states of power or an on/off load at a customer premises using a controller. The controller may comprise a processor, which may be configured to receive a regulation signal associated with an ancillary service for the power grid; determine whether to modify a state of power of the power consumption component based on a probability function, local measurement at the premises, and the received regulation signal; and based at least in part on a determination to modify the state of power of the power consumption component, modify the state of power of the power consumption component. The system may be distributed, with different controllers at different customer premises independently making power state decisions.

Abstract: A system, method and apparatus for providing ancillary services to a power grid using a power consumption component at a customer premises. The apparatus may control a load with discrete, finite states of power or an on/off load at a customer premises using a controller. The controller may comprise a processor, which may be configured to receive a regulation signal associated with an ancillary service for the power grid; determine whether to modify a state of power of the power consumption component based on a probability function, local measurement at the premises, and the received regulation signal; and based at least in part on a determination to modify the state of power of the power consumption component, modify the state of power of the power consumption component. The system may be distributed, with different controllers at different customer premises independently making power state decisions.