The invention relates to energy generation and water conservation. There are many water systems which might lend themselves to energy extraction, such as canal systems. However, despite canal systems being around for hundreds of years, practical solutions for using the energy available have not been developed. The invention provides, among various examples, a system which can be associated with a lock in a canal and provides a flow control strategy responsive to water availability upstream of the lock. Thus electricity may be generated selectively in response to the lock state and energy demands but without adversely affecting the canal system by taking excess water. The system may be applied to multiple locks and may incorporate machine learning to evolve a strategy for a canal based on lock usage and energy demand.

A method and an apparatus for controlling noise of multiple wind turbines. The method includes: determining a noise-influencing sector of each of the multiple wind turbines, based on positions of the multiple wind turbines and a position of a noise-influencing site; acquiring a current wind direction; determining whether there is at least one wind turbine of the multiple wind turbine under the current wind direction operating in the noise-influencing sector; and limiting output power of the at least one wind turbine, in a case that the determination is positive.

A method for planning repowering of a wind energy plant is disclosed. The wind energy plant is positioned at a site and comprises a plurality of original wind turbines arranged at distributed positions within the site. Historical data related to the original wind turbines is retrieved. The historical data is collected over a previous time period and during operation of the original wind turbines. Meteorological conditions at the site of the wind energy plant are estimated based on the retrieved historical data. The repowering of the wind energy plant is planned based on the estimated meteorological data, including planning replacement of the original wind turbines by replacement wind turbines to be positioned at the positions of the original wind turbines.

The invention provides a method for controlling a wind turbine, including predicting behaviour of one or more wind turbine components such as a wind turbine tower over a prediction horizon using a wind turbine model that describes dynamics of the one or more wind turbine components or states. The method includes determining behavioural constraints associated with operation of the wind turbine, wherein the behavioural constraints are based on operational parameters of the wind turbine such as operating conditions, e.g. wind speed. The method includes using the predicted behaviour of the one or more wind turbine components in a cost function, and optimising the cost function subject to the determined behavioural constraints to determine at least one control output, such as blade pitch control or generator speed control, for controlling operation of the wind turbine.

According to an embodiment, a method of controlling a temperature of a blade includes generating a first power production curve based on current weather conditions and generating a second power production curve based on future weather conditions. The method also includes, in response to determining that the second power production curve reduces a net power production loss of the blade more than the first power production curve, adjusting a heating cycle of the blade based on the second power production curve rather than the first power production curve.

Some embodiments are directed to a method for estimating a periodic or substantially periodic force present in a mechanical or electromechanical system, the method comprising: estimating, by a processing device, one or more harmonic frequencies of an acceleration signal representing an acceleration in the system, the substantially periodic force contributing to said acceleration; and estimating, by the processing device, the force on the basis of a dynamic model of the system, the dynamic model being defined by said one or more estimated harmonic frequencies.

A method for controlling a power output of a power generating unit is disclosed. An accumulated power output of the power generating unit during a predefined time interval is forecasted. An actual power output of the power generating unit is measured during the predefined time interval, and an actual accumulated power output is estimated for the predefined time interval on the basis of the measured actual power output of the power generating unit. A difference between the forecasted accumulated power output and the estimated actual accumulated power output is derived. The power output of the power generating unit is boosted, in the case that the estimated actual accumulated power output is below the forecasted accumulated power output, and the difference between the forecasted accumulated power output and the estimated actual accumulated power output is larger than a predefined threshold value.

A method of parameterizing a controller of a first wind energy installation wherein the controller sets a manipulated variable of the wind energy installation as a function of an input variable. An artificial intelligence determines at least one value of a parameter of the controller for at least one state/degree of being iced up of the wind energy installation based on a power curve, load, and/or downstream flow of the wind energy installation predicted with a mathematical model of the wind energy installation for at least one state/degree of being iced up, and/or determines at least one value of a parameter of the controller for at least one state/degree of being iced up of the wind energy installation, based on at least one determined state/degree of being iced up and a power, load, and/or downstream flow of the wind energy installation and/or at least one second wind energy installation.

A computer system and method for improving the accuracy of predictions of the amount of renewable energy, such as solar energy and wind energy, available to an electric utility, and/or refine such predictions, by providing improved integration of meteorological forecasts. Coefficient values are calculated for a renewable energy generation model by performing a regression analysis with the forecasted level of renewable energy posted by the utility, forecasted weather conditions and measures of seasonality as explanatory variables. Accuracy is further enhanced through the inclusion of a large number of time series variables that reflect the systematic nature of the energy/weather system. The model also uses the original forecast posted by the system operator as well as variables to control for season.

An apparatus and method is disclosed for over-rating a wind turbine using turbulence prediction. Weather forecast information is used to determine whether there is a risk of turbulent conditions occurring at the site of the wind turbine. The wind turbine is over-rated if turbulent conditions are not predicted, and conversely over-rating is cancelled or reduced if turbulent conditions are expected. This allows an increase in the annual energy production of the wind turbine to be realised. The weather forecast information may be combined with real time measurements of operating conditions to supplement the predictions.