Method and device for monitoring icing of blades of wind generating set and electronic equipment

Abstract

The embodiment of the invention provides a wind generating set blade icing monitoring method and device, electronic equipment and a storage medium. Firstly, a first temperature value and / or a first humidity value detected by a temperature and humidity sensor can be obtained; then, when the first temperature value is smaller than the preset temperature value and / or the first humidity value is larger than the preset humidity value, a first image shot by the image shooting device can be obtained. And finally, extracting a target image from the first image, and inputting the target image into a pre-trained icing identification model to determine the icing state of the blade. According to the embodiment of the invention, the target image is processed by adopting the pre-trained icing identification model, the icing state of the blade is determined, and the target image comprises one or more of the image of the blade, the image of the temperature and humidity sensor and the image of the icing sensor, so that the accuracy of monitoring icing of the blade can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind power generation, and in particular to a wind turbine blade icing monitoring method and device, an electronic device, and a storage medium. BACKGROUND

[0002] Wind power generation technology has been widely used in the world due to its low pollution and abundant renewable wind energy. However, when a wind power generator is used in a cold region, the blades are prone to icing. Blade icing can cause changes in the aerodynamic performance of the wind turbine blades, leading to blade overload, uneven blade load distribution, and thus a decrease in wind capture capacity, affecting power generation. In addition, it is easy to cause overload and shorten the service life of components. Meanwhile, during the rotation of the blades, ice blocks are prone to falling off when the adhesion force of the ice layer decreases, causing operational accidents.

[0003] In related technologies, the blade icing state can be indirectly determined by analyzing the deviation of the actual output power of the wind turbine from the theoretically expected power. This method is the wind power mismatch method. The core logic is that icing can change the aerodynamic shape of the blades, increase wind resistance, and reduce wind energy capture efficiency, thereby causing power to decrease or fluctuate abnormally. However, there are many reasons for wind power mismatch, making it difficult to accurately determine whether the blade causes wind power mismatch due to icing.

[0004] Therefore, how to improve the accuracy of monitoring blade icing has become a technical problem to be solved at present. SUMMARY

[0005] In a first aspect, an embodiment of the present application provides a wind turbine blade icing monitoring method, comprising: obtaining a first temperature value and / or a first humidity value detected by a temperature and humidity sensor; in a case where the first temperature value is less than a preset temperature value and / or the first humidity value is greater than a preset humidity value, obtaining a first image captured by an image capturing device; the first image covers at least part of a region of a blade, part of a region of a temperature and humidity sensor, and / or part of a region of an icing sensor; extracting a target image from the first image; the target image includes at least one of an image of the blade, an image of the temperature and humidity sensor, and an image of the icing sensor; and inputting the target image into a pre-trained icing recognition model to determine an icing state of the blade.

[0006] Optionally, the distance between the image capturing device and the blade is greater than the distance between the image capturing device and the temperature and humidity sensor; and / or the distance between the image capturing device and the blade is greater than the distance between the image capturing device and the icing sensor.

[0007] Optionally, in a case that the first temperature value is less than a preset temperature value and / or the first humidity value is greater than a preset humidity value, the first image captured by the image capturing device is acquired, including: in a case that the first temperature value is less than the preset temperature value, determining whether the first humidity value is greater than the preset humidity value; if greater than the preset humidity value, acquiring the first image captured by the image capturing device.

[0008] Optionally, the target image is extracted from the first image, including: determining whether the image of the blade can be extracted from the first image; in a case that the image of the blade cannot be extracted, the image of the temperature and humidity sensor and / or the image of the icing sensor is extracted from the first image.

[0009] The target image is input into a pre-trained icing recognition model to determine the icing state of the blade, including: the image of the temperature and humidity sensor and / or the image of the icing sensor is input into the pre-trained icing recognition model to determine the icing state of the temperature and humidity sensor and / or the icing state of the icing sensor; and the icing state of the blade is determined according to the icing state of the temperature and humidity sensor and / or the icing state of the icing sensor.

[0010] Optionally, the image capturing device, the temperature and humidity sensor and / or the icing sensor are located on the surface of the nacelle of the wind turbine generator set or a support; and the support is located on the surface of the nacelle.

[0011] Optionally, the pre-trained icing recognition model is obtained by pre-training based on a first sample set; the first sample set includes images of the temperature and humidity sensor that has been iced and corresponding first labels, and images of the temperature and humidity sensor that has not been iced and corresponding second labels; the first labels are used to indicate different icing states, and the second labels are used to indicate an un-iced state; and / or, the first sample set includes images of the icing sensor that has been iced and corresponding third labels, and images of the icing sensor that has not been iced and corresponding fourth labels; the third labels are used to indicate different icing states, and the fourth labels are used to indicate an un-iced state; and / or, the first sample set includes images of the blade that has been iced and corresponding fifth labels, and images of the blade that has not been iced and corresponding sixth labels; the fifth labels are used to indicate different icing states, and the sixth labels are used to indicate an un-iced state.

[0012] Optionally, the icing state includes an icing type and / or an icing degree, the icing type is frost ice, clear ice or mixed ice, and the icing degree is heavy icing, moderate icing or light icing.

[0013] Optionally, the icing state comprises an icing type and / or an icing degree, the icing type is frost ice, clear ice or mixed ice, and the icing degree is heavy icing, medium icing or light icing.

[0014] The method further comprises: controlling the wind turbine to shut down when the icing type of the blade is frost ice or mixed ice and the icing degree is medium icing or heavy icing; and controlling the wind turbine to shut down when the icing type of the blade is clear ice and the icing degree is heavy icing.

[0015] The method further comprises: controlling the wind turbine to reduce the rotating speed to below a preset speed threshold when the icing type of the blade is clear ice and the icing degree is medium icing; and controlling the wind turbine to reduce the rotating speed to below a preset speed threshold when the icing degree is light icing.

[0016] Optionally, inputting the target image into the pre-trained icing recognition model to determine the icing state of the blade comprises: inputting the image of the blade into the pre-trained icing recognition model to determine the icing position of the blade.

[0017] Optionally, the pre-trained icing recognition model is obtained by pre-training based on a second sample set; the second sample set comprises images of blades with icing and corresponding seventh labels; the seventh label is used to indicate different icing positions on the blade; the icing position of the blade comprises a windward surface of the blade root, a leeward surface of the blade root, a windward surface in the blade, a leeward surface in the blade, a windward surface of the blade tip and a leeward surface of the blade tip.

[0018] Optionally, the method further comprises: determining a heater corresponding to the icing position according to the icing position of the blade; and controlling the heater to heat the icing position.

[0019] Optionally, the method further comprises: obtaining an icing detection result of the icing sensor; controlling the wind turbine to reduce the rotating speed to below a preset speed threshold and determining whether the icing state is icing when the icing detection result is icing; and controlling the wind turbine to shut down when the icing state is icing.

[0020] Optionally, controlling the wind turbine to reduce the rotating speed to below a preset speed threshold when the icing detection result is icing comprises: controlling the wind turbine to reduce the rotating speed to below a preset speed threshold when a duration of the icing detection result being icing reaches a preset duration.

[0021] Secondly, embodiments of this application provide a monitoring device for icing on wind turbine blades, comprising: a first acquisition module for acquiring a first temperature value and / or a first humidity value detected by a temperature and humidity sensor; a second acquisition module for acquiring a first image captured by an image capturing device when the first temperature value is less than a preset temperature value and / or the first humidity value is greater than a preset humidity value; the first image at least covers a portion of the blade, a portion of the temperature and humidity sensor, and / or a portion of the icing sensor; an extraction module for extracting a target image from the first image; the target image includes at least one of the following: an image of the blade, an image of the temperature and humidity sensor, and an image of the icing sensor; and a determination module for inputting the target image into a pre-trained icing recognition model to determine the icing state of the blade.

[0022] Thirdly, embodiments of this application provide an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the electronic device implements any of the aforementioned methods.

[0023] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the aforementioned methods.

[0024] This application provides a method, apparatus, electronic device, and storage medium for monitoring icing on wind turbine blades. First, a first temperature value and / or a first humidity value detected by a temperature and humidity sensor can be acquired. Then, if the first temperature value is less than a preset temperature value and / or the first humidity value is greater than a preset humidity value, a first image captured by an image capturing device can be acquired. Finally, a target image can be extracted from the first image and input into a pre-trained icing recognition model to determine the icing state of the blades. This application uses a pre-trained icing recognition model to process the target image to determine the icing state of the blades. The target image includes one or more of the following: an image of the blade, an image from a temperature and humidity sensor, and an image from an icing sensor; therefore, the accuracy of monitoring blade icing can be improved. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1A schematic flowchart illustrating a method for monitoring icing on wind turbine blades provided in this application embodiment;

[0027] Figure 2 This is another flowchart illustrating the monitoring method for wind turbine blade icing provided in the embodiments of this application;

[0028] Figure 3 A schematic diagram of a monitoring device for monitoring icing of wind turbine blades provided in an embodiment of this application;

[0029] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0030] In this article, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The symbol " / " in this article indicates that the related objects are in an "or" relationship; for example, A / B means A or B.

[0031] The terms "first" and "second," etc., used in the specification and claims herein are used to distinguish different objects, not to describe a specific order of objects. For example, "first response message" and "second response message," etc., are used to distinguish different response messages, not to describe a specific order of response messages.

[0032] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0033] In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more, for example, multiple processors means two or more processors, multiple elements means two or more elements, etc.

[0034] To provide a clearer description of the embodiments of this application, the basic structure of a wind turbine generator set will first be described.

[0035] A wind turbine is the core equipment for converting wind energy into electrical energy. Its complex and highly integrated structure mainly comprises the following key components: a rotor system, a transmission and power generation system, a support and structural system, and an auxiliary and control system. Among these, the rotor system is the core energy capture unit of the wind turbine, and its design directly affects the wind energy conversion efficiency. The rotor system consists of blades and a hub. For example, three blades can be connected to the same hub. The blades generate lift through aerodynamic airfoil design (similar to an aircraft wing), converting wind energy into rotational kinetic energy. The hub connects the blades to the main shaft, transmits mechanical energy, and adjusts the blade angle through a pitch system to control the rotational speed.

[0036] This application provides a method, device, electronic equipment, and storage medium for monitoring icing on wind turbine blades, which can improve the accuracy of monitoring blade icing.

[0037] Figure 1 This is a flowchart illustrating a method for monitoring icing on wind turbine blades provided in an embodiment of this application. The executing entity in this embodiment is an electronic device, which can be a server, desktop computer, laptop, mobile phone, etc., and this embodiment does not impose any limitations on this.

[0038] like Figure 1 As shown in the embodiments of this application, the method for monitoring icing on wind turbine blades may include the following steps:

[0039] S11, acquire the first temperature value and / or the first humidity value detected by the temperature and humidity sensor.

[0040] Specifically, a temperature and humidity sensor, also known as a temperature and humidity data acquisition device, is a device that converts ambient temperature and humidity into measurable electrical signals. These measurable electrical signals characterize the ambient temperature value (first temperature value) and the ambient humidity value (first humidity value). This temperature and humidity sensor can be connected to the aforementioned electronic device. In one example, the temperature and humidity sensor can send the first temperature value to the electronic device. In another example, the temperature and humidity sensor can send the first humidity value to the electronic device. In yet another example, the temperature and humidity sensor can send both the first humidity value and the first temperature value to the electronic device.

[0041] The temperature and humidity sensor can send the measurable electrical signal to the electronic device, thereby enabling the electronic device to obtain the temperature and humidity values ​​of the environment in which the temperature and humidity sensor is located.

[0042] S12, when the first temperature value is less than the preset temperature value and / or the first humidity value is greater than the preset humidity value, a first image captured by the image capturing device is acquired; the first image at least covers a part of the leaf, a part of the temperature and humidity sensor and / or a part of the icing sensor.

[0043] In this step, icing will only occur on the blades under specific temperature and humidity conditions. For example, icing will only occur when the ambient temperature is below 0 degrees Celsius and the humidity is above 80%. The same applies to the temperature and humidity sensors and the icing sensor, which will not be detailed here.

[0044] Therefore, in the first implementation, the electronic device can first determine whether the first temperature value is less than a preset temperature value (e.g., 0 degrees Celsius). Only when it is less than the preset temperature value will it control the image capturing device to capture an image of the leaf to generate a first image. This first image can cover part or all of the leaf. Next, the image capturing device can send the first image to the electronic device.

[0045] In the second implementation, the only difference from the first is that the image capturing device captures an image of the temperature and humidity sensor, rather than the leaf, to generate the first image. Accordingly, this first image can cover part or all of the temperature and humidity sensor area.

[0046] In the third implementation, the only difference from the first is that the image capturing device captures an image of the icing sensor, rather than the blade itself, to generate the first image. Accordingly, this first image can cover part or all of the icing sensor. For example, the icing sensor can be a capacitive sensor.

[0047] Furthermore, the electronic device can also execute the content described in the aforementioned three implementation methods when the first humidity value is greater than the preset humidity value. The electronic device can also execute the content described in the aforementioned three implementation methods when the first temperature value is less than the preset temperature value and the first humidity value is greater than the preset humidity value.

[0048] In this way, the image capturing device can capture images of the blades, temperature and humidity sensors, and / or icing sensors only when the ambient temperature and / or humidity meet preset conditions, thereby reducing the power consumption of the image capturing device.

[0049] S13, extract the target image from the first image; the target image includes at least one of the following: an image of the blade, an image of the temperature and humidity sensor, and an image of the icing sensor.

[0050] In this step, since the first image includes one or more images of the leaf, the temperature and humidity sensor, and the icing sensor, the electronic device can use semantic segmentation technology to filter the background in the first image to obtain the target image, which is one or more of the images of the leaf, the temperature and humidity sensor, and the icing sensor.

[0051] S14, the target image is input into a pre-trained icing recognition model to determine the icing state of the blade.

[0052] In this step, the target image can be input into a pre-trained icing recognition model to determine the icing state of the blades.

[0053] In one specific implementation, the pre-trained icing recognition model can be obtained by pre-training based on a first sample set.

[0054] In the first scenario, during the creation of the first sample set, the user can manually select images of iced temperature and humidity sensors, manually identify the icing state of the sensors, create corresponding labels based on the icing state, and associate the two. Thus, the first sample set will include images of iced temperature and humidity sensors and their corresponding labels (i.e., the first label). This first label indicates the specific icing state of the temperature and humidity sensors in the aforementioned images. For example, the specific icing state can be either open ice or a light degree of icing. To enable the pre-trained icing recognition model to identify various icing states of temperature and humidity sensors, the first sample set should include images of temperature sensors in various icing states and their corresponding labels.

[0055] Regarding the state of freezing, in one example, the state of freezing can include the type of freezing, which can specifically be frost, clear ice, or a mixture of both. In another example, the state of freezing can include the degree of freezing, which can specifically be heavy freezing, moderate freezing, or light freezing. In yet another example, the state of freezing can include both the type of freezing and the degree of freezing from the previous two examples.

[0056] Furthermore, during the creation of the first sample set, users can manually select images of temperature and humidity sensors that are not iced, create corresponding tags, and associate the two. In this way, the first sample set will include images of temperature and humidity sensors that are not iced and their corresponding tags (i.e., the second tags). These second tags are used to indicate the icy state of the temperature and humidity sensors in the aforementioned images.

[0057] After training the icing recognition model using the first sample set, in actual use, the images from the temperature and humidity sensors obtained in step S13 can be input into the icing recognition model to determine the current icing state of the temperature and humidity sensors. Since both the icing sensor and the blades are located in the wind turbine generator, their icing states are almost identical, thus the icing state of the blades can be determined.

[0058] Similarly, in the second scenario, the first sample set may include images of icing sensors that have already formed ice and their corresponding labels (i.e., the third label), which indicates different icing states. The implementation is similar to the description above, the only difference being that here it uses images of icing sensors that have already formed ice, rather than images of icing temperature and humidity sensors. The first sample set may also include images of icing sensors that have not formed ice and their corresponding labels (i.e., the fourth label), which indicates the non-icing state. The implementation is similar to the description above, the only difference being that here it uses images of icing sensors that have not formed ice, rather than images of icing temperature and humidity sensors.

[0059] In this case, after training the icing recognition model using the first sample set, the image of the icing sensor obtained in step S13 can be input into the icing recognition model during actual use, thereby determining the current icing state of the icing sensor. Since both the icing sensor and the blade are located in the wind turbine generator, their icing states are almost identical, thus the icing state of the blade can be determined.

[0060] Similarly, in the third case, the first sample set may include images of icy leaves and corresponding labels (i.e., the fifth label), which indicates different icing states. Its implementation is similar to the description above, the only difference being that here it uses images of icy leaves, rather than images of icy temperature and humidity sensors. The first sample set may also include images of icy leaves and corresponding labels (i.e., the sixth label), which indicates the non-icing state. Its implementation is similar to the description above, the only difference being that here it uses images of non-iced leaves, rather than images of non-iced temperature and humidity sensors.

[0061] In this case, after training the icing recognition model using the first sample set, the image of the blade obtained in step S13 can be input into the icing recognition model during actual use, thereby determining the current icing state of the blade.

[0062] It should be noted that the first sample set may include images and corresponding labels from the two or three of the aforementioned cases. Using this first sample set to train the icing recognition model helps to improve the accuracy of the icing recognition model in identifying whether the leaves are icy.

[0063] In a specific example, the first sample set includes images and corresponding labels from all three scenarios mentioned above. After training the icing recognition model using this first sample set, the model can process images of leaves to determine their icing state, process images of temperature and humidity to determine their icing state, and process images from icing sensors to determine their icing state. When the target image in step S14 includes all three types of images, the icing recognition model can integrate the three judgment results to ultimately determine the icing state of the leaf, which helps improve the accuracy of the judgment.

[0064] The above description is merely an example. When the first sample set includes images and corresponding labels from the two scenarios mentioned above, the pre-trained icing recognition model can also integrate the two judgment results to ultimately determine the icing status of the leaves, which helps improve the accuracy of the icing recognition model in identifying whether the leaves are icy.

[0065] In some embodiments of this application, the distance between the image capturing device and the blade can be greater than the distance between the image capturing device and the temperature and humidity sensor. This allows the image captured by the image capturing device of the temperature and humidity sensor to be clearer than the image of the blade, thereby enabling the pre-trained icing recognition model to more accurately determine the icing state of the blade when processing the image of the temperature and humidity sensor.

[0066] Similarly, the distance between the image capturing device and the blade can be greater than the distance between the image capturing device and the icing sensor. This allows the pre-trained icing recognition model to more accurately determine the icing state of the blade when processing images from the icing sensor.

[0067] Based on the above embodiments, when performing the extraction of the target image from the first image (step S13), the electronic device can first determine whether the image of the leaf can be extracted from the first image. If the image of the leaf cannot be extracted, it indicates that the leaf image contained in the first image is not clear. For example, when the air visibility is poor (e.g., in rainy weather), the clarity of the leaf image in the first image is poor.

[0068] Since the temperature and humidity sensor and / or icing sensor are closer to the image capturing device than the blades, the electronic device can extract the images from the first image and / or the icing sensor, and input these images into a pre-trained icing recognition model to determine the icing state of the temperature and humidity sensor and / or the icing sensor, thereby determining the icing state of the blades. In this way, the electronic device can still determine the icing state of the blades using the images from the temperature and humidity sensor and / or the icing sensor even when the image capturing device cannot obtain a clear image of the blades, expanding the application scenarios of the embodiments of this application.

[0069] The nacelle of a wind turbine generator is located at the top of the tower and connected to the tower via a yaw system. It is a key structural unit housing the core mechanical and electrical equipment. The nacelle integrates the drivetrain (main shaft, gearbox, generator, etc.), control system, and auxiliary equipment. A support frame can be installed on the upper surface of the nacelle, with the height of the support frame approximately the same as the height of the nacelle's upper surface. In this embodiment, the image capturing device, temperature and humidity sensor, and icing sensor can all be located on the surface of the wind turbine generator's nacelle. This ensures that the temperature and humidity sensor and the icing sensor are at approximately the same altitude as the blades. Therefore, the icing status of the temperature and humidity sensor and the icing sensor is essentially the same as the icing status of the blades. Thus, after the pre-trained icing recognition model outputs the icing status of the temperature and humidity sensor and the icing sensor, the icing status of the blades can be determined more accurately.

[0070] Furthermore, since the image capturing device is relatively close to the temperature and humidity sensor and the icing sensor, it can obtain clear images of both sensors. This helps improve the accuracy of the icing status determination by the temperature and humidity sensor and the icing sensor, thereby improving the accuracy of the icing status determination of the blades.

[0071] It should be noted that brackets can be installed on the surface of the cabin, and image capturing devices, temperature and humidity sensors, and icing sensors can also be arbitrarily installed on the surface of the cabin and on the brackets. The resulting technical effects are basically the same as those of the aforementioned solutions.

[0072] Furthermore, in the case where the embodiments of this application only include one of the temperature and humidity sensor and the icing sensor, the image capturing device and the temperature and humidity sensor can be arbitrarily set on the surface of the cabin and the bracket, or the image capturing device and the icing sensor can be arbitrarily set on the surface of the cabin and the bracket, and the resulting technical effect is basically the same as the technical effect of the above-mentioned solution.

[0073] In some embodiments of this application, the icing state may include icing type and / or icing degree, wherein the icing type is frost ice, clear ice or mixed ice, and the icing degree is heavy icing, moderate icing or light icing.

[0074] Specifically, when the icing type on the blades is frost or mixed ice, and the degree of icing is moderate or severe, ice blocks are prone to falling off the blades during rotation. Therefore, electronic equipment can control the wind turbine to shut down to avoid safety accidents caused by falling ice blocks.

[0075] When the icing type on the blades is open ice and the degree of icing is severe, ice blocks are prone to falling off the blades during rotation. Therefore, electronic equipment can control the wind turbine to shut down to avoid safety accidents caused by falling ice blocks.

[0076] When the icing type on the blades is open ice and the degree of icing is moderate, the risk of ice chunks falling off the blades during rotation is low. Therefore, the electronic equipment controls the wind turbine to reduce its rotational speed to below a preset speed threshold to avoid safety accidents caused by falling ice. At the same time, this ensures that the wind turbine can continue generating electricity. This preset speed threshold can be set according to actual needs.

[0077] In cases of light icing, the risk of ice fragments falling off the blades during rotation is low. Therefore, electronic control systems reduce the wind turbine's rotation speed to below a preset threshold to prevent accidents caused by falling ice. This also ensures the wind turbine continues to generate electricity.

[0078] Therefore, the embodiments of this application can adopt different control strategies according to different icing states of the blades, which can ensure the safe operation of the wind turbine generator set and generate as much power as possible.

[0079] In some embodiments of this application, when performing step S14, the specific implementation method can be to input the image of the blade into a pre-trained icing recognition model to determine the icing location of the blade. Therefore, the icing location of the blade can be accurately determined.

[0080] The pre-trained icing recognition model can be obtained by pre-training based on a second sample set. The second sample set includes images of icy leaves and corresponding seventh labels, which indicate different icing locations on the leaves. A leaf can include three parts: the leaf root, the leaf middle, and the leaf tip. The icing locations on the leaf include six parts: the windward side of the leaf root, the windward side of the leaf root, the windward side of the leaf middle, the windward side of the leaf middle, the windward side of the leaf tip, and the windward side of the leaf tip.

[0081] Therefore, electronic devices can use a pre-trained icing recognition model to determine which of the six parts the icing on the blades is located in.

[0082] After the electronic equipment determines the specific location of icing on the blade, it can identify the corresponding heater. The electronic equipment can then control this heater to heat the icing area, thereby achieving rapid ice melting.

[0083] For example, six heaters can be installed in the blade, each corresponding to a part of the blade. For instance, heater A is installed below the windward side of the blade root. When the icing location of the blade is determined to be the windward side of the blade root, the electronic equipment can activate heater A to heat the windward side of the blade root, thereby achieving a de-icing effect. The principle is the same when the icing location of the blade is in other parts, and will not be elaborated further here.

[0084] In some embodiments of this application, such as Figure 2 As shown in the embodiments of this application, the method for monitoring icing on wind turbine blades may further include:

[0085] S15, Obtain the icing detection result of the icing sensor;

[0086] S16, if the icing detection result is that the wind turbine has iced, control the wind turbine to reduce its speed to below a preset speed threshold, and determine whether the icing state is that the wind turbine has iced.

[0087] S17, if the icing state is already frozen, control the wind turbine to stop.

[0088] Specifically, an icing sensor is an electronic device used to detect the thickness of ice on the surface of an object, and it is widely used in aviation, power, transportation, and building safety monitoring. Its core function is to convert icing signals into electrical signals to achieve real-time monitoring and early warning. In this embodiment, the icing sensor is used to detect whether the surface of a blade is iced. For example, the icing sensor can be a capacitive sensor.

[0089] If the icing detection result indicates that the blades are likely icy, the electronic equipment can control the wind turbine to reduce its speed to below a preset speed threshold, thereby reducing potential safety risks. Furthermore, the electronic equipment can also determine whether the icing status in step S14 is indeed icy. If it is, it indicates a high probability that the blades are already icy, and therefore the wind turbine can be shut down to ensure safe operation of the wind turbine generator set.

[0090] In one specific implementation, when the icing detection result indicates icing, the electronic equipment can monitor the duration of this condition. When the duration reaches a preset time (e.g., 2 hours), the wind turbine can be controlled to reduce its speed to below a preset speed threshold. This improves the accuracy of the icing sensor in determining the icing status.

[0091] like Figure 3 As shown in the embodiment of this application, a monitoring device for icing on wind turbine blades is also provided. This device may include:

[0092] The first acquisition module is used to acquire the first temperature value and / or the first humidity value detected by the temperature and humidity sensor;

[0093] The second acquisition module is used to acquire a first image captured by the image capturing device when the first temperature value is less than a preset temperature value and / or the first humidity value is greater than a preset humidity value; the first image at least covers a part of the leaf, a part of the temperature and humidity sensor and / or a part of the icing sensor.

[0094] An extraction module is configured to extract a target image from the first image; the target image includes at least one of the following: an image of the leaf, an image of the temperature and humidity sensor, and an image of the icing sensor;

[0095] The determination module is used to input the target image into a pre-trained icing recognition model to determine the icing state of the blade.

[0096] This application provides a monitoring device for icing on wind turbine blades. First, a first temperature value and / or a first humidity value detected by a temperature and humidity sensor can be acquired. Then, if the first temperature value is lower than a preset temperature value and / or the first humidity value is higher than a preset humidity value, a first image captured by an image capturing device can be acquired. Finally, a target image can be extracted from the first image and input into a pre-trained icing recognition model to determine the icing state of the blades. This application uses a pre-trained icing recognition model to process the target image to determine the icing state of the blades. The target image includes one or more of the following: an image of the blade, an image from a temperature and humidity sensor, and an image from an icing sensor; therefore, the accuracy of monitoring blade icing can be improved.

[0097] Optionally, the distance between the image capturing device and the blade is greater than the distance between the image capturing device and the temperature and humidity sensor; and / or, the distance between the image capturing device and the blade is greater than the distance between the image capturing device and the icing sensor.

[0098] Optionally, the second acquisition module is specifically used to: determine whether the first humidity value is greater than the preset humidity value when the first temperature value is less than the preset temperature value; if it is greater than the preset humidity value, then acquire the first image captured by the image capturing device.

[0099] Optionally, the extraction module is specifically used to: determine whether the image of the leaf can be extracted from the first image; if the image of the leaf cannot be extracted, extract the image of the temperature and humidity sensor and / or the image of the icing sensor from the first image;

[0100] The extraction module is specifically used to: input the image of the temperature and humidity sensor and / or the image of the icing sensor into a pre-trained icing recognition model to determine the icing state of the temperature and humidity sensor and / or the icing state of the icing sensor; and determine the icing state of the blade based on the icing state of the temperature and humidity sensor and / or the icing state of the icing sensor.

[0101] Optionally, the image capturing device, the temperature and humidity sensor, and / or the icing sensor are located on the surface of the nacelle or on the support of the wind turbine generator set; the support is located on the surface of the nacelle.

[0102] Optionally, the pre-trained icing recognition model is pre-trained based on a first sample set; the first sample set includes images of iced temperature and humidity sensors and corresponding first labels, and images of uniced temperature and humidity sensors and corresponding second labels; the first label indicates different icing states, and the second label indicates an uniced state; and / or, the first sample set includes images of iced icing sensors and corresponding third labels, and images of uniced icing sensors and corresponding fourth labels; the third label indicates different icing states, and the fourth label indicates an uniced state; and / or, the first sample set includes images of iced leaves and corresponding fifth labels, and images of uniced leaves and corresponding sixth labels; the fifth label indicates different icing states, and the sixth label indicates an uniced state.

[0103] Optionally, the icing state includes icing type and / or icing degree, wherein the icing type is frost ice, clear ice, or mixed ice, and the icing degree is heavy icing, moderate icing, or light icing.

[0104] Optionally, the icing state includes icing type and / or icing degree, wherein the icing type is frost ice, clear ice, or mixed ice, and the icing degree is heavy icing, moderate icing, or light icing;

[0105] The monitoring device for icing on wind turbine blades is also used to: control the wind turbine to shut down when the icing type of the blades is frost or mixed ice, and the icing degree is moderate or severe; control the wind turbine to shut down when the icing type of the blades is clear ice, and the icing degree is severe; control the wind turbine to reduce its speed to below a preset speed threshold when the icing type of the blades is clear ice, and the icing degree is moderate; and control the wind turbine to reduce its speed to below a preset speed threshold when the icing degree is light.

[0106] Optionally, the determining module is specifically used to: input the image of the blade into a pre-trained icing recognition model to determine the icing location of the blade.

[0107] Optionally, the pre-trained icing recognition model is pre-trained based on a second sample set; the second sample set includes images of iced leaves and corresponding seventh labels; the seventh label is used to indicate different icing locations on the leaves; the icing locations of the leaves include the windward side of the leaf root, the leeward side of the leaf root, the windward side of the leaf, the leeward side of the leaf, the windward side of the leaf tip, and the leeward side of the leaf tip.

[0108] Optionally, the monitoring device for icing of wind turbine blades is further configured to: determine the heater corresponding to the icing location based on the icing location of the blade; and control the heater to heat the icing location.

[0109] Optionally, the monitoring device for icing of wind turbine blades is further configured to: acquire the icing detection result of the icing sensor; if the icing detection result indicates that icing has occurred, control the wind turbine to reduce its rotational speed to below a preset speed threshold and determine whether the icing state is confirmed; if the icing state is confirmed, control the wind turbine to shut down.

[0110] Optionally, when the icing detection result indicates that the wind turbine is icing, and the wind turbine is controlled to reduce its speed to below a preset speed threshold, the wind turbine blade icing monitoring device is specifically used to: control the wind turbine to reduce its speed to below a preset speed threshold when the duration of the icing detection result indicating that the icing state has reached a preset duration.

[0111] The monitoring method and device for wind turbine blade icing provided in this application embodiment can perform the actions performed by the electronic equipment in the above embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0112] like Figure 4As shown in the embodiments of this application, an electronic device is also provided. The electronic device includes a memory and a processor. The memory is used to store a computer program. When the computer program is executed by the processor, it can implement the monitoring method for icing of wind turbine blades as described above. For details, please refer to the description of the foregoing embodiments.

[0113] Specifically, at the hardware level, the electronic device may include a processor, an internal bus, and memory. The memory may include main memory and non-volatile memory. The processor reads the corresponding computer program from the non-volatile memory into main memory and then executes it. Those skilled in the art will understand that... Figure 4 The structure shown is for illustrative purposes only and does not limit the structure of the electronic device described above. For example, the electronic device may also include components that are larger than... Figure 4 The components shown may include more or fewer components, such as other processing hardware like a GPU (Graphics Processing Unit) or external communication ports. Of course, this application does not exclude other implementation methods besides software implementations, such as logic devices or a combination of hardware and software.

[0114] In this embodiment, the processor may include a central processing unit (CPU) or a graphics processing unit (GPU), and may also include other microcontrollers, logic gates, integrated circuits, or appropriate combinations thereof with logic processing capabilities. The memory described in this embodiment can be a storage device for storing information. In digital systems, a device capable of storing binary data can be a memory; in integrated circuits, a circuit without physical form but with storage function can also be a memory, such as RAM or FIFO; in a system, a storage device with physical form can also be called a memory. In implementation, this memory can also be implemented using a cloud storage method; the specific implementation method is not limited in this specification.

[0115] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the monitoring method for icing of wind turbine blades as described above.

[0116] This application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the method for monitoring icing on wind turbine blades as described above.

[0117] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0118] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for monitoring icing on wind turbine blades, characterized in that, include: Acquire the first temperature value and / or first humidity value detected by the temperature and humidity sensor; When the first temperature value is less than the preset temperature value and / or the first humidity value is greater than the preset humidity value, a first image is acquired by the image capturing device; the first image at least covers a part of the leaf, a part of the temperature and humidity sensor and / or a part of the icing sensor. Extract a target image from the first image; the target image includes at least one of the following: an image of the leaf, an image of the temperature and humidity sensor, and an image of the icing sensor; The target image is input into a pre-trained icing recognition model to determine the icing state of the blade.

2. The method according to claim 1, characterized in that, The distance between the image capturing device and the blade is greater than the distance between the image capturing device and the temperature and humidity sensor; and / or, the distance between the image capturing device and the blade is greater than the distance between the image capturing device and the icing sensor.

3. The method according to claim 1, characterized in that, When the first temperature value is less than a preset temperature value and / or the first humidity value is greater than a preset humidity value, acquiring a first image captured by the image capturing device includes: If the first temperature value is less than the preset temperature value, determine whether the first humidity value is greater than the preset humidity value; If the humidity is greater than the preset humidity value, then the first image captured by the image capturing device is obtained.

4. The method according to claim 2, characterized in that, Extracting the target image from the first image includes: Determine whether the image of the leaf can be extracted from the first image; If the image of the leaf cannot be extracted, the image of the temperature and humidity sensor and / or the image of the icing sensor are extracted from the first image; The target image is input into a pre-trained icing recognition model to determine the icing state of the blades, including: The image of the temperature and humidity sensor and / or the image of the icing sensor are input into a pre-trained icing recognition model to determine the icing state of the temperature and humidity sensor and / or the icing state of the icing sensor. The icing state of the blade is determined based on the icing state of the temperature and humidity sensor and / or the icing state of the icing sensor.

5. The method according to claim 1, characterized in that, The image capturing device, the temperature and humidity sensor, and / or the icing sensor are located on the surface of the nacelle or on the support of the wind turbine generator set; the support is located on the surface of the nacelle.

6. The method according to claim 1, characterized in that, The pre-trained icing recognition model is obtained by pre-training based on the first sample set; The first sample set includes images of iced temperature and humidity sensors and corresponding first tags, and images of uniced temperature and humidity sensors and corresponding second tags; the first tags are used to indicate different icing states, and the second tags are used to indicate uniced states. And / or, the first sample set includes images of icing sensors that have frozen and corresponding third tags, and images of icing sensors that have not frozen and corresponding fourth tags; the third tags are used to indicate different freezing states, and the fourth tags are used to indicate non-icing states; And / or, the first sample set includes images of icy leaves and corresponding fifth tags, and images of uniced leaves and corresponding sixth tags; the fifth tags are used to indicate different icing states, and the sixth tags are used to indicate uniced states.

7. The method according to claim 6, characterized in that, The freezing state includes freezing type and / or freezing degree, wherein the freezing type is frost ice, clear ice or mixed ice, and the freezing degree is heavy freezing, moderate freezing or light freezing.

8. The method according to claim 1, characterized in that, The freezing state includes freezing type and / or freezing degree, wherein the freezing type is frost ice, clear ice or mixed ice, and the freezing degree is heavy freezing, moderate freezing or light freezing; The method further includes: If the icing type of the blades is frost or mixed ice, and the icing degree is moderate or severe, the wind turbine will be shut down. If the icing type of the blades is open ice and the icing degree is severe icing, the wind turbine will be shut down. When the icing type of the blade is open ice and the degree of icing is moderate, the wind turbine is controlled to reduce its speed to below a preset speed threshold. When the icing level is light icing, the wind turbine is controlled to reduce its rotational speed to below a preset speed threshold.

9. The method according to claim 1, characterized in that, The target image is input into a pre-trained icing recognition model to determine the icing state of the blades, including: The image of the blade is input into a pre-trained icing recognition model to determine the icing location of the blade.

10. The method according to claim 9, characterized in that, The pre-trained icing recognition model was obtained by pre-training based on the second sample set; The second sample set includes images of frozen leaves and corresponding seventh tags; The seventh label is used to indicate different icing locations on the blade; The icing locations of the blades include the windward side of the blade root, the leeward side of the blade root, the windward side of the middle of the blade, the leeward side of the middle of the blade, the windward side of the blade tip, and the leeward side of the blade tip.

11. The method according to claim 9, characterized in that, The method further includes: Based on the icing location of the blades, determine the heater corresponding to the icing location; The heater is controlled to heat the icing location.

12. The method according to claim 1, characterized in that, The method further includes: Obtain the icing detection results from the icing sensor; If the icing detection result indicates that icing has occurred, the wind turbine is controlled to reduce its speed to below a preset speed threshold, and it is determined whether the icing state is indeed icing. If the icing condition is confirmed to be frozen, the wind turbine will be shut down.

13. The method according to claim 12, characterized in that, If the icing detection result indicates that icing has occurred, controlling the wind turbine to reduce its speed to below a preset speed threshold includes: If the duration of the icing detection result indicating that the state of icing has lasted for a preset duration, the wind turbine is controlled to reduce its rotational speed to below a preset speed threshold.

14. A monitoring device for icing on wind turbine blades, characterized in that, include: The first acquisition module is used to acquire the first temperature value and / or the first humidity value detected by the temperature and humidity sensor; The second acquisition module is used to acquire a first image captured by the image capturing device when the first temperature value is less than a preset temperature value and / or the first humidity value is greater than a preset humidity value. The first image covers at least a portion of the leaf, a portion of the temperature and humidity sensor, and / or a portion of the icing sensor. An extraction module is configured to extract a target image from the first image; the target image includes at least one of the following: an image of the leaf, an image of the temperature and humidity sensor, and an image of the icing sensor; The determination module is used to input the target image into a pre-trained icing recognition model to determine the icing state of the blade.

15. An electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it causes the electronic device to implement the method as described in any one of claims 1 to 13.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 13.

Images