A method for controlling a wind power installation is provided. The method includes operating the installation at a normal operating point at which the installation is not operated in a throttled fashion if there is no request for throttling, and operating the installation at a throttled operating point, in response to a throttle request, with output power which is throttled in comparison with the normal operating point. The method includes changing the operation from the throttled operating point to a reserve operating point at which the installation is operated with higher output power in response to a power increase request. The throttled operating point has a higher tip speed ratio than that of the reserve operating point and is positioned on an associated iso-characteristic curve in the λθ diagram having, at the throttled operating point, a negative characteristic curve gradient.

Methods and apparatus for reducing peak power consumption of a grid connected power plant having a plurality of wind turbines. In response to determining that a power production value of the power plant is below a power threshold, one method includes: after a first time delay of a first group of one or more wind turbines, control the first group to operate in a power saving mode for a predefined first power saving period; and after a first time delay of a second group of one or more other wind turbines, control the second group to operate in the power saving mode for a predefined second power saving period. The first time delay of the first group is less than the first time delay of the second group and the power saving mode inhibits a power consuming activity for the wind turbines operating in the power saving mode.

A method for operating a wind farm having a plurality of wind power installations is provided. The installations each comprise an aerodynamic rotor, and the rotors each have an aerodynamic characteristic value. The method includes: obtaining a setpoint power value of the wind farm, in particular a setpoint value of the electrical power of the wind farm to be fed in, ascertaining an actual power value of the wind farm as the sum of actual electrical powers of the operated wind power installations, determining a permissibility of a power-reduced operating mode of each of the wind power installations of the wind farm on the basis of the associated aerodynamic characteristic value, and operating the wind power installations of the wind farm such that each operated wind power installation is operated in a permissible operating mode and the ascertained actual power value does not exceed the obtained setpoint power value.

A multistage wind turbine or network of wind turbines with improved and optimized wind-directing, wind-shaping, and wind-power conversion features indicates that the shapes of these features directly affect the ability of the multistage wind turbine to use the power of moving air, such as wind, to spin a rotor and create torque on a rotor shaft to generate electricity. The wind-power-conversion mechanical efficiency described significantly improves upon previous designs by conversion of wind energy into electrical power at a superior price-to-performance ratio compared with existing alternative energy technologies.

A method and an apparatus for controlling noise of a wind turbine. The method includes: determining a noise-influencing sector of the wind turbine based on a position of the wind turbine and a position of a noise-influencing site; acquiring a current wind direction; determining whether the wind turbine under the current wind direction operates in the noise-influencing sector; and limiting output power of the wind turbine, and increasing, after the wind turbine goes out from the noise-influencing sector and the output power reaches a rated power, the output power, in a case that the determination is positive.

A method includes receiving, via one or more turbine-level controllers, an indication of at least one of a communication loss between the one or more turbine-level controllers and a farm-level controller, a detection of an absence of reactive power regulation by the farm-level controller, or a reactive power command of the farm-level controller being equal to or above a saturation threshold during transitioning between a baseline operational mode and reactive power mode, the reactive power mode being characterized in that only reactive power is generate. Upon receipt of the indication, the method includes adjusting a reactive power response of one or more reactive power regulators of the one or more turbine-level controllers so as to avoid an overshoot reactive power event or an undershoot reactive power event at the point of interconnection.

The objective of the present invention is to diagnose an abnormality in the amount of electricity generated by a distributed power source while restricting an increase in the number of sensors. A packet generating unit 200 generates electricity generation amount packet data D2a from electricity generation amount data D1 and identification data D6; an electricity generation feature quantity calculating unit 201 calculates electricity generation feature quantity data D3a, D3b indicating a feature quantity of the amount of electricity generated by each distributed power source, from each item of and a result output unit 203 consolidates abnormalities in the amounts of electricity generated by the distributed power sources, indicated by the determination result data D5, as abnormality diagnosis result data D7, and outputs the same to a display unit 206.

Disclosed is an apparatus that adapts the rate of its computational work to match the availability of energy harvested from a stochastic energy source; and, with respect to some types of energy harvesting, regulates the rate of energy capture, the rate of energy conversion, and the rate of consumption of stored potential energy, through its alteration, regulation, and/or adjustment, of that same computational work load.

A computer-implemented reinforcement learning method includes determining, based on a target probability of satisfaction of a constraint condition related to a state of a control object and a specific time within which a controller causes the state of the control object not satisfying the constraint condition to be the state of the control object satisfying the constraint condition, a parameter of a reinforcement learner that causes, in a specific probability, the state of the control object to satisfy the constraint condition at a first timing following a second timing at which the state of control object satisfies the constraint condition; and determining a control input to the control object by either the reinforcement learner or the controller, based on whether the state of the control object satisfies the constraint condition at a specific timing.

A method for identifying underperforming agents in a multi-agent cooperative system includes receiving information relating to the performance of each agent in the multi-agent system, calculating an estimated extracted resource value of each agent based on the received information, comparing the estimated extracted resource value of each agent to a threshold value, calculating a performance index based on the comparison and identifying an agent as an under-performing agent based on the performance index. A system for identifying under-performing agents in a plurality of agents in a multi-agent cooperative system includes a performance analyzing processor, a communications port for receiving state information for each agent and control information for each agent, a classifier for identifying a subset of agents in the plurality of agents that are performance comparable and an optimizer configured to identify an under-performing agent of performance comparable agents and generate updated control information for the identified under-performing agent.