The following refers to a method and a system for computer-implemented analysis of a wind farm comprising a number of wind turbines. Furthermore, the following refers to a corresponding computer program product and a corresponding computer program.
During the planning phase of a wind farm, a surrounding area around the planned location of the wind farm is analyzed. In this analysis, all impacts of wind turbines on humans and particularly residents in the neighborhood of the farm are evaluated. As a result, operation constraints are determined in order to comply with a curtailment limiting the adverse effects resulting from the operation of the wind farm.
A curtailment may refer to a maximum sound pressure level, particularly during night, at residential buildings around the wind farm. To comply with this curtailment, the wind farm has to be operated with reduced power resulting in less speed of the rotor blades of wind turbines and thus in less sound. Another curtailment may refer to a limited operation time of the wind turbines in order to reduce the adverse effects of shadow flickering for residents in the neighborhood of a wind farm. To comply with this curtailment, fixed operation intervals for the turbines are defined.
After having determined operation constraints in the planning phase, the wind farm launched after planning is usually operated based on the operation constraints for its complete life time even in case that sources of curtailment disappear in the meantime.
An aspect relates to a method for computer-implemented analysis of a wind farm in order to automatically determine an efficient operation of the farm.
The method of the invention provides a computer-implemented analysis of a wind farm comprising a number of wind turbines, i.e. at least one wind turbine. This wind farm has already been installed at a location on the earth's ground.
In method of the invention, the following steps i) and ii) are performed at each time point of one or more time points. In step i), an object detection algorithm is applied to a digital image showing the current state of the earth's ground in a surrounding area of the wind farm. This object detection algorithm results in the extraction and localization of a number of detected objects not belonging to the wind farm within the image, where each detected object is of an object type out of a number of object types. Relevant object types may be defined beforehand. As an object detection algorithm, any conventional art algorithm for analyzing images may be used. In an embodiment, the object detection algorithm is based on a trained data driven model where images comprising known objects of corresponding object types were used as training data so that objects of these object types can indeed be extracted. In a particular embodiment, the data driven model is a neural network. In an embodiment, a Convolutional Neural Network well known from the conventional art is used. Convolutional Neural Networks are particularly suitable for processing image data.
In a first variant of step ii) of the method, an information is determined on whether there is a change with respect to the number of detected objects (i.e. whether objects have disappeared or occurred and whether properties of existing objects have changed) in comparison to an earlier state (i.e. a state at a time point before the current state) of the earth's ground in the surrounding area of the wind farm, the change enabling an adaptation of the operation of the wind farm in compliance with a predetermined curtailment limiting one or more (adverse) effects resulting from the operation of the wind farm. Hence, in this first variant, only an indication of a possible adaptation of the operation of the wind farm due to environmental changes is given. E.g., this information may be used by the operator of the wind farm in order to obtain permission for an adapted operation of the wind farm from authorities. In an embodiment, the earlier state of the earth's ground is derived from a corresponding (earlier) image by applying an object detection algorithm, as it is the case for the current state of the earth's ground. However, the earlier state of the earth's ground may also be based on another digital description not being based on an image.
In a second variant of step ii) which can be performed alone or in combination with the first variant, a number of operation constraints for the wind farm is (automatically) determined based on the number of detected objects such that the wind farm generates maximum electric energy within a predetermined time interval on condition that a predetermined curtailment (corresponding to the above curtailment if the first and second variants are combined) limiting one or more (adverse) effects resulting from the operation of the wind farm is complied with. A predetermined time interval may refer e.g. to one year so that the maximum electric energy refers to the maximum annual energy production of the wind farm. Methods for automatically determining operation constraints based on objects in the neighborhood of a wind farm are known for a skilled person and, thus, will not be described in detail herein.
Embodiments of the invention are based on the finding that an automatic analysis of images of the earth's ground around a wind farm enables to derive objects relevant for operation constraints based on a predetermined curtailment defined for the wind farm. Hence, it is possible to update operation constraints based on newly acquired images. As a consequence, the operation of a wind farm can be adapted to changing environmental conditions.
In an embodiment, the one or more effects limited by the predetermined curtailment refer to sound and/or shadow flickering and/or ice throw caused by the wind farm. The predetermined curtailment includes those effects by defining corresponding restrictions, e.g. by defining a maximum sound pressure level at residential buildings in the surrounding area of the wind farm or by defining a maximum operation time of the wind turbines during day in order to limit sound propagation and/or shadow flickering or by defining a maximum rotation speed of the wind turbine rotors in order to avoid ice throw during winter.
In an embodiment, the above number of object types comprises the following types:
In a variant of the above embodiment, only the object types “building” and/or “traffic route” and/or “ground inclination” and/or “tree” may be defined.
In an embodiment, the one or more detected objects are associated with one or more properties thereof which are extracted by the object detection algorithm. This enables a very exact determination of operation constraints.
In another embodiment, the one or more properties of the detected objects refer to the height of a detected object which is particularly useful when considering sound and/or shadow flickering as the effects limited by the predetermined curtailment.
The above steps i) and ii) are performed several times based on certain criteria. In an embodiment, steps i) and ii) are repeated in case that an updated digital image is available, e.g. from a database storing those images. In another embodiment, steps i) and ii) are repeated in case that the predetermined curtailment has changed. This enables an adaptation of the operation of the wind farm to changed regulations of authorities.
The digital image processed by the method of embodiments of the invention are taken by a flying object over ground, particularly by a satellite or a plane or a drone. Nevertheless, the image may also be taken by one or more cameras installed on ground at the location of the wind farm, e.g. at the position of the nacelle of one or more wind turbines.
In an embodiment, the information on whether there is a change with respect to the number of detected objects in comparison to an earlier state of the earth's ground in the surrounding area of the wind farm and/or the number of operation constraints for the wind farm and/or the maximum electric energy as determined in step ii) are output via a user interface. This user interface may be accessible for staff of the operator of the wind farm. Hence, the operator has the option to change the operation of the wind farm or to negotiate with authorities in order to achieve an adapted operation resulting in higher energy output.
Besides the above method, embodiments of the invention refer to a system for computer-implemented analysis of a wind farm comprising a number of wind turbines, where the system comprises a processor configured to carry out the method according to embodiments of the invention or according to one or more embodiments of the invention.
The invention also refers to a computer program product (non-transitory computer readable storage medium having instructions, which when executed by a processor, perform actions) with program code, which is stored on a non-transitory machine-readable carrier, for carrying out the method according to the invention or according to one or more embodiments of the invention, when the program code is executed on a computer.
Furthermore, embodiments of the invention refer to a computer program with program code for carrying out the method according to embodiments of the invention or according to one or more embodiments of the invention, when the program code is executed on a computer.
The method as described in the following refers to the analysis of a wind farm where the operation of the wind farm is subjected to a curtailment limiting one or more adverse effects for humans resulting from the operation of the wind farm. In an embodiment, the curtailment refers to restrictions given by an authority in order to limit sound and shadow flickering of the wind farm for residential buildings in the surrounding area of the wind farm. Sound and shadow flickering is caused by the rotation of the wind turbine rotors of the wind farm.
The above curtailment can be defined in various ways. E.g., a curtailment with respect to sound can be such that a sound pressure level caused by the wind farm at the location of a residential building must not exceed a certain threshold. Furthermore, the sound curtailment may also be coupled to certain time periods, e.g. the above threshold may only be applicable for the operation of the wind farm during night or different thresholds for night and day may be defined.
Furthermore, a curtailment with respect to shadow flickering may be given by a maximum amount of operation hours of the wind farm in a given time interval, e.g. within one year. Further criteria may apply with respect to shadow flickering, e.g. the restriction of the operation hours may only be applicable during day time or in case that the sun is shining. The condition that the sun is shining can be determined based on corresponding sensors positioned at the location of the wind farm.
Another curtailment may refer to the operation of the wind farm during possible icing conditions, e.g. in order to limit ice throw caused by the rotation of the rotor blades of the wind turbines. In this case, a curtailment with respect to a maximum rotation speed of the rotor blades of the wind turbines may be given during winter or during periods with low temperatures in case that humans are expected in the surrounding area of the wind farm, e.g. in case that a street is located near the wind farm.
In the embodiment described herein, a wind farm already in operation is analyzed based on satellite images in order to get information about changes in the surrounding area of the wind farm. Those changes may allow an adaptation of the operation of the wind farm resulting in higher electric energy production whilst a given curtailment is still complied with.
The method according to
The satellite images IM are taken from a central database which may be a database publicly available. Furthermore, the method of embodiments of the invention have access to a curtailment CU in the form of digital data given by an authority in order to limit adverse effects resulting from the operation of the wind farm, e.g. the above described effects concerning sound, shadow flickering and ice throw. This curtailment will be processed in step S2 of
According to the method of
In the method of
In the embodiment described herein, a well-known algorithm based on a Convolutional Neural Network CNN is used for detecting objects within the image. The Convolutional Neural Network is trained based on training images having known objects of known object types shown therein. As known from the conventional art, a Convolutional Neural Network comprises convolutional layers followed by pooling layers as well as fully connected layers in order to extract and classify the objects within an image.
After having performed step S1, the information about the detected objects is used in order to determine one or more operation constraints OC of the wind farm taking into account the above mentioned curtailment CU provided as digital data in step S2. The operation constraints OC for the wind farm are defined such that the wind farm generates maximum electric energy within a predetermined time interval on condition that the curtailment CU is complied with. In the embodiment described herein, the maximum electric energy refers to the maximum annual energy production AEP.
The result of the method of
The derivation of operation constraints OC based on step S2 of
Embodiments of the invention as described in the foregoing have several advantages. The conditions with respect to the operation of a wind farm can be checked regularly based on updated images of the wind farm. In one embodiment, those images are satellite images. Nevertheless, the images may also refer to images taken by other flying objects (plane or drone) or by a camera installed on ground at the location of the wind farm. As a result of embodiments of the invention, a higher energy output of a wind farm may be achieved in case that objects relevant for a predetermined curtailment are no longer present or have been changed. Furthermore, it is also possible to perform the method of embodiments of the invention in case that a curtailment given by an authority has been changed (particularly relaxed) in order to evaluate if the wind farm can be operated with higher energy output.
Although the present invention has been disclosed in the form of preferred embodiments and variations thereon, it will be understood that numerous additional modifications and variations could be made thereto without departing from the scope of the invention.
For the sake of clarity, it is to be understood that the use of “a” or “an” throughout this application does not exclude a plurality, and “comprising” does not exclude other steps or elements.
Number | Date | Country | Kind |
---|---|---|---|
19156942.5 | Feb 2019 | EP | regional |
This application claims priority to PCT Application No. PCT/EP2020/053151, having a filing date of Feb. 7, 2020, which is based on EP Application No. 19156942.5, having a filing date of Feb. 13, 2019, the entire contents both of which are hereby incorporated by reference.
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP2020/053151 | 2/7/2020 | WO | 00 |