A device and method for monitoring the state of a wind turbine blade

Abstract

The present invention relates to the field of infrared thermal imaging temperature measurement technology, and provides a device and method for monitoring the state of a fan blade of a wind turbine. The device includes a housing and a pan-tilt head. An inner sleeve is installed inside the housing, and a visible light module and an infrared thermal imaging module are installed inside the inner sleeve. The visible light module and the infrared thermal imaging module are installed to observe the same picture. The pan-tilt head fixedly installed on the wind turbine is connected to the bottom of the housing. The visible light module is used to collect the fan blade image when the wind turbine is operating, and the infrared thermal imaging module is used to collect the infrared image of the fan blade when the wind turbine is operating. This solution uses the infrared thermal imaging module in cooperation with the visible light module to monitor and analyze the wind turbine for the first time. The fan blade is imaged by the infrared thermal imaging module, so as to monitor the state of the fan blade in real time, and further guide the operation and maintenance of the wind turbine, improving the efficiency and reliability of the wind turbine.

Description

Technical Field

[0001] The present invention relates to the technical field of infrared thermal imaging temperature measurement, and particularly relates to a monitoring device and method for the state of a wind turbine blade. Background Art

[0002] In recent years, with the continuous promotion and construction of green energy projects, among which wind power generation is the main direction vigorously promoted and applied at present. A wind turbine is a wind energy power machine, and the wind turbine blades are the main components for receiving wind energy. The blade design requires efficient airfoils, reasonable installation angles, optimized lift-drag ratios, tip speed ratios, and blade twist laws, etc. Since wind turbines are installed at windward locations such as mountains, wildernesses, beaches, and islands, they are affected by irregular variable wind directions, variable load wind forces, and strong gust impacts, and are exposed to the influence of extreme heat, cold, lightning strikes, and extreme temperature differences all year round. There may be hidden cracks. In addition, in low-temperature conditions in mountainous areas, icing will occur on the blades. Icing will increase the operating load of the wind turbine and even damage the wind turbine. Therefore, the state of the wind turbine blades requires a real-time online monitoring system to monitor and sample the operating conditions of the blades, perform data statistical analysis, and conduct remote fault analysis to ensure the normal operation of the wind turbine blades.

[0003] The currently commonly used method is single visible light monitoring. Through the experience of monitoring personnel, it is judged whether there are abnormal conditions such as icing and hidden cracks on the fan blades in the collected visible light images, and whether icing will occur in the future is judged based on the temperature trend in the next few days, so as to decide whether to stop or inspect and maintain the wind turbine. This method relies on the experience of monitoring personnel. When the monitoring personnel cannot judge the state of the wind turbine blades or the future temperature trend is unclear, in order to protect the wind turbine, the wind turbine is often stopped. This not only cannot effectively judge the state of the wind turbine blades, but also greatly wastes the power generation efficiency of the wind turbine. Summary of the Invention

[0004] The present invention provides a monitoring device and method for the state of a wind turbine blade, and solves the technical problem of poor monitoring efficiency caused by low intelligent level of the state monitoring of the wind turbine blades as described above.

[0005] The present invention provides 1. A monitoring device for the state of a wind turbine blade. The device includes a housing and a pan-tilt head. An inner sleeve is installed inside the housing, and a visible light module and an infrared thermal imaging module are installed inside the inner sleeve. The visible light module and the infrared thermal imaging module are installed to observe the same picture;

[0006] The pan-tilt head fixedly installed on the wind turbine is connected to the bottom of the housing;

[0007] The visible light module is used to collect the blade images during the operation of the wind turbine generator, and the infrared thermal imaging module is used to collect the infrared images of the blades during the operation of the wind turbine generator.

[0008] Preferably, a centralized processing module is further provided inside the inner sleeve. The centralized processing module is used to analyze and process the blade images and infrared images of the blades, and determine whether there are dark stripes or icing on the blades.

[0009] Preferably, the outer shell is ellipsoidal.

[0010] Preferably, a temperature sensor and a semiconductor refrigerator are further provided inside the inner sleeve. The semiconductor refrigerator is used to adjust the temperature of the inner sleeve according to the temperature data of the inner sleeve monitored by the temperature sensor and control it within -40°C to 50°C.

[0011] Preferably, semiconductor refrigerators are installed at both the upper and lower parts of the inner sleeve, and each semiconductor refrigerator has a heating surface and a cooling surface;

[0012] The cooling surface of the semiconductor refrigerator located at the upper part faces the internal functional modules inside the inner sleeve, and the heating surface of the semiconductor refrigerator located at the lower part faces the internal functional modules inside the inner sleeve, forming a dual-system functional component for cooling and heating the temperature of the internal space of the inner sleeve.

[0013] Preferably, a wiper is installed on the outer shell, and the wiper is used to clean the windows of the visible light module and the infrared thermal imaging module.

[0014] Preferably, there is a gap between the outer shell and the inner sleeve, and a heat-insulating layer is provided on the outer layer of the inner sleeve.

[0015] The present invention also provides a method for monitoring the state of the blades of a wind turbine generator. The detection method is used to be realized by a device for monitoring the state of the blades of a wind turbine generator, and includes:

[0016] S1, collecting the visible light image and the infrared image of the blades of the wind turbine generator;

[0017] S2, performing unified input requirement conversion on the collected image data set to ensure the consistency of the subsequent processing results; specifically, performing preprocessing of color space and color features on the visible light image; performing image preprocessing of inversion, histogram equalization, denoising, and sharpening on the infrared image;

[0018] S3, using a convolutional neural network of deep learning to train the blade target object detector in the visible light module, and performing recognition and detection of the blade target image; when the intensity of the ambient light is lower than the threshold and is not sufficient to enable the visible light module to detect the blade target, the infrared thermal imaging module is used alone to perform recognition and detection of the target image;

[0019] S4. Divide the polar regions of each temperature area of the detected fan blade image, and establish corresponding infrared image training sets and test sets with temperature information for dark cracks, snowfall, regional icing, and overall icing.

[0020] S5. Use a long short-term memory network (LSTM) to learn and classify the fan blade icing area images in the infrared images with temperature information, and obtain the icing degree image classification of dark cracks, snowfall, small-area icing, large-area icing, overall icing, and overall low-temperature icing in the time series, thus completing the function of the image data analysis module.

[0021] S6. Alarm for the corresponding icing state or dark cracks according to the set conditions, and transmit the information to external devices to provide reference information for whether the wind turbine continues to operate.

[0022] Preferably, between S1 and S2, it further includes: separately establishing libraries for visible light images and infrared images, filtering the visible light images through a convolution filter to filter and activate the fan blade features in the images, and learning to judge and identify the fan blade target; filtering the infrared images through a convolution filter to filter and activate the thermal radiation information features of the icing and dark crack areas in the images, and learning to judge whether there are dark cracks and icing.

[0023] Beneficial effects: The present invention provides a device and method for monitoring the state of a wind turbine fan blade. The device includes a housing and a pan-tilt. An inner sleeve is installed inside the housing, and a visible light module and an infrared thermal imaging module are installed inside the inner sleeve. The visible light module and the infrared thermal imaging module are installed to observe the same picture; the pan-tilt fixedly installed on the wind turbine is connected to the bottom of the housing; the visible light module is used to collect fan blade images during the operation of the wind turbine, and the infrared thermal imaging module is used to collect infrared images of the fan blade during the operation of the wind turbine. This solution uses an infrared thermal imaging module in cooperation with a visible light module to monitor and analyze the wind turbine for the first time. By imaging the fan blade with the infrared thermal imaging module, and using the fact that the thermal radiation of dark cracks and icing is quite different from that of the surrounding normal areas to identify and analyze the dark crack and icing areas, so as to monitor the state of the fan blade in real time, and then guide the operation and maintenance of the wind turbine, improving the efficiency and reliability of the wind turbine.

[0024] The above description is only an overview of the technical solution of the present invention. In order to be able to understand the technical means of the present invention more clearly and implement it according to the content of the specification, the following takes the preferred embodiments of the present invention and combines the drawings to describe in detail as follows. The specific implementation manners of the present invention are given in detail by the following embodiments and their accompanying drawings. Brief Description of the Drawings

[0025] The accompanying drawings described herein are used to provide a further understanding of the present invention, and form a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention, and do not constitute an improper limitation of the present invention. In the drawings:

[0026] Figure 1 is a structural diagram of the fan blade status monitoring device of the wind turbine generator set of the present invention;

[0027] Figure 2 is an installation schematic diagram of the fan blade status monitoring device of the wind turbine generator set of the present invention;

[0028] Figure 3 is a schematic diagram of the principle of the fan blade status monitoring device of the wind turbine generator set of the present invention;

[0029] Figure 4 is a flowchart of the fan blade status monitoring method of the wind turbine generator set of the present invention. Detailed Embodiments

[0030] The principles and features of the present invention will be described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention. The present invention will be described more specifically by way of example in the following paragraphs with reference to the accompanying drawings. The advantages and features of the present invention will be clearer according to the following description and claims. It should be noted that the drawings are all in a very simplified form and use non-precise scales, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0031] It should be noted that when a component is referred to as "fixed to" another component, it can be directly on the other component or there can also be an intermediate component. When a component is considered to be "connected" to another component, it can be directly connected to the other component or there may be an intermediate component at the same time. When a component is considered to be "disposed on" another component, it can be directly disposed on the other component or there may be an intermediate component at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0033] Such as Figures 1 to 3As shown in the figure, the present invention provides a monitoring device for the state of the fan blades of a wind turbine generator. The device includes a housing 1, an inner sleeve 2, a visible light module 3, an infrared thermal imaging module 4, a centralized processing module 5, a temperature sensor 6, a semiconductor cooler 7, a network switch 8, a power supply 9, a wiper module 10, and a pan-tilt head.

[0034] The housing 1 is designed in an ellipsoidal shape with an overall streamlined shape. While carrying other internal modules, it can well resist the wind blowing from all directions, eliminate or reduce the shaking of the machine head caused by the wind, and keep the working environment of the device in a stable state.

[0035] The inner sleeve 2 is installed in the housing 1 and is designed with a heat-insulating layer of a certain thickness. There is a gap between the inner sleeve 2 and the housing 1. The heat-insulating layer and the gap can well separate the inner sleeve 2 from the housing 1, so that the temperature change of the housing 1 has no or as little impact as possible on the inner sleeve 2, further ensuring a stable working environment inside the device.

[0036] The visible light module 3 is installed in the inner sleeve 2 and is mainly used to collect the images of the fan blades during the operation of the wind turbine generator. The infrared thermal imaging module 4 is installed in the inner sleeve 2 and is used to collect the infrared images of the fan blades during the operation of the wind turbine generator.

[0037] The centralized processing module 5 is installed in the inner sleeve 2 and is used to analyze and process the visible light and infrared images of the fan blades, and judge whether there are dark stripes or icing on the fan blades according to the images.

[0038] The temperature sensor 6 is installed in the inner sleeve 2 and is used to collect the real-time temperature inside the device. The cold end of the semiconductor cooler 7 is below the inner sleeve 2 and is located below the visible light module 3 and the infrared thermal imaging module 4, and the hot end is above the inner sleeve 2. When the temperature sensor 6 collects that the temperature inside the housing 1 is lower than the minimum starting temperature of the visible light module 3 and the infrared thermal imaging module 4, the system will control the semiconductor cooler 7 to heat until the internal temperature of the device reaches the starting temperature; when the temperature sensor 6 collects that the temperature inside the housing 1 is higher than the maximum working temperature of the visible light module 3 and the infrared thermal imaging module 4, the system will control the semiconductor cooler 7 to refrigerate and reduce the internal temperature of the device to the normal working temperature.

[0039] The network switch 8 is installed in the inner sleeve 2 and is used for signal exchange and transmission; the power supply 9 is installed in the inner sleeve 2 and is used for power supply of the device; in addition, this device is designed with a wiper module 10, which is installed on the housing 1 and is used to clean the windows of the visible light module 3 and the infrared thermal imaging module 4, so that the visible light module 3 and the infrared thermal imaging module 4 have a clear field of view.

[0040] As Figure 2As shown in the figure, the pan-tilt 202 is installed at the lower part of the nose 203 and on the outer shell of the wind turbine generator. The pan-tilt 202 is a two-axis pan-tilt, which can rotate 360° horizontally and has a 95° pitch. The pan-tilt 202 can adjust the observation direction of the device, so as to observe different positions of the fan blade 201 according to requirements.

[0041] The visible light module 3 and the infrared thermal imaging module 4 are installed to observe the same picture. Both use autofocus lenses. After the visible light module 3 and the infrared thermal imaging module 4 collect images of the fan blade 201, the centralized processing module 5 can respectively create libraries for the visible light images and infrared images collected by the visible light module 3 and the infrared thermal imaging module 4. The visible light images are filtered through a convolutional filter to filter and activate the fan blade features in the images, and learn to judge and identify the fan blade target; the infrared images are filtered through a convolutional filter to filter and activate the thermal radiation information features of the icing and dark crack areas in the images, and learn to judge whether there are dark cracks and icing; then the obtained information is transmitted to the wind turbine generator monitoring terminal to provide guidance for the operation and maintenance of the wind turbine generator.

[0042] Among them, the block diagrams of each module are as Figure 3 As shown in the figure, after the visible light module and the infrared thermal imaging module collect the fan blade images, they are transmitted backward to the switch for signal exchange, and then the images are transmitted to the centralized processing module. The centralized processing module processes the visible light images and infrared images respectively, identifies the fan blade area through the visible light images, and identifies the icing or dark cracks on the fan blade through the infrared images. At the same time, the centralized processing module will use a convolutional neural network to self-learn through a large number of collected visible light images and infrared images to improve the recognition rate of the visible light module for the fan blade and the recognition rate of the infrared thermal imaging module for the icing and dark cracks on the fan blade; in addition, the centralized processing module also receives the working environment temperature information collected by the temperature sensor and makes a judgment, and controls the hot end or cold end of the semiconductor refrigeration component to work according to the judgment result to adjust the internal working environment temperature of the device; according to the clarity of the visible light image background, the centralized processing module will also control the wiper to clean the windows of the visible light module and the infrared thermal imaging module to ensure that the image background collected by the visible light module and the infrared thermal imaging module is clean and free of interfering objects.

[0043] For a further solution, the installation position on the wind turbine generator is as Figure 3 As shown in the figure, the bottom of the pan-tilt of the device is installed at the top of the tail end of the equipment cabin of the wind turbine generator. The visible light module and the infrared thermal imaging module face the fan blade, and the precise observation angle can be adjusted through the pan-tilt at the wind turbine generator control end, improving the convenience of using the device.

[0044] As Figure 4 shown in the figure, the present invention also provides a method for monitoring the state of the fan blade of a wind turbine generator, which specifically includes the following:

[0045] (1) The visible light module and the infrared thermal imaging module collect the visible light image and the infrared image of the monitored fan blade target;

[0046] (2) Perform database building operations on the acquired real-time images to create visible light and infrared image datasets respectively. Usually, 25 frames of images can be collected per second for both visible light and infrared, and 90,000 specified images can be collected per hour. The established visible light and infrared image databases respectively include their own training set databases and test set databases;

[0047] (3) Preprocess the collected images. First, perform unified input requirement conversions (such as image size and channel conversions) on the collected image datasets to ensure the consistency of subsequent processing results. Perform preprocessing on the color space and color features of the visible light images; perform image preprocessing such as inversion, histogram equalization, denoising, and sharpening on the infrared images to obtain preprocessed visible light images and infrared images;

[0048] (4) Use the convolutional neural network of deep learning to train the fan blade target detector in the visible light module, identify and detect the fan blade target image, and obtain information such as the position and area of the fan blade target in the image.

[0049] (5) When the intensity of the ambient light enables the visible light module to detect the fan blade target, use the visible light module to perform target image recognition, and map the coordinate information of the identified fan blade target or the fan blade target area information to the infrared image coordinates to improve the detection rate of the fan blade target in the infrared image; when the intensity of the ambient light is insufficient to enable the visible light module to detect the fan blade target, use the infrared thermal imaging module alone to perform target image recognition;

[0050] (6) The detected fan blade infrared image contains target temperature information (i.e., thermal radiation information). Therefore, the temperature regions can be divided, and corresponding infrared image training sets and test sets for dark cracks, snowfall, regional icing, and full-area icing with temperature information can be established.

[0051] (7) Use the long short-term memory network (LSTM) to learn and classify the fan blade icing region images in the infrared images with temperature information, and obtain the icing degree image classification of dark cracks, snowfall, small-area icing, large-area icing, full-area icing, and full-area low-temperature icing in the time series to complete the functions of the image data analysis module.

[0052] After obtaining the classified image information, alarms for the corresponding icing state or dark cracks are implemented according to the set conditions, and the information is transmitted to external devices (such as the wind turbine control console) to provide reference information for whether the wind turbine continues to operate. By statistically analyzing the detection, recognition, and classification results of the fan blades multiple times in the time series, this system can achieve the function of predicting the icing state within a period of time and can identify the dark cracks existing on the fan blades.

[0053] The present invention provides a dual-light system device and a monitoring and analysis method for monitoring and analyzing the state of fan blades of large wind turbines. The fan blades of the wind turbine are identified through a visible light module, and real-time non-contact thermal imaging sampling of the fan blades of the wind turbine is performed through an infrared thermal imaging module. A large amount of collected image data is trained using a self-deep learning method to obtain the temperature and images of different regions on the fan blades. Whether the fan blades are iced can be judged by the temperature, and whether there are cracks or hidden damages on the surface of the fan blades can be judged by the shadows, dark lines, etc. on the thermal imaging images, realizing the monitoring and analysis of the state of the fan blades. For the first time, an infrared thermal imaging module is used in cooperation with a visible light module to monitor and analyze a wind turbine, solving the problem that the existing visible light monitoring system cannot be detected in environments with insufficient light (such as at night, rain, snow, heavy fog). Moreover, the temperature at night is generally lower than that during the day, and the fan blades of the wind turbine are more likely to ice. Therefore, the infrared thermal imaging module can better monitor the fan blades at night; a monitoring and analysis method based on multi-layer deep learning of neural networks is adopted, and self-learning is carried out through a large number of collected images to continuously improve the image recognition accuracy, thereby reducing the dependence on the observation experience of monitoring personnel. The present invention provides a method for monitoring and analyzing the state of fan blades of large wind turbines, which can scientifically guide the operation and maintenance of wind turbines and improve the equipment utilization efficiency and stability.

[0054] Generally speaking, compared with the prior art through the above technologies proposed by the present invention, the following advantages or benefits can be obtained:

[0055] For the first time, an infrared thermal imaging module is used in cooperation with a visible light module to monitor and analyze a wind turbine. The fan blades are imaged through the infrared thermal imaging module, and the dark cracks and icing areas are identified and analyzed by using the large difference in thermal radiation between the dark cracks and the iced areas and the surrounding normal areas, so as to monitor the state of the fan blades in real time, and then guide the operation and maintenance of the wind turbine, improving the efficiency and reliability of the wind turbine;

[0056] By using the infrared thermal imaging module, the problem that the existing visible light monitoring system cannot be detected in environments with insufficient light (such as at night, rain, snow, heavy fog) can be solved. The temperature at night is generally lower than that during the day, and the fan blades of the wind turbine are more likely to ice. Therefore, the infrared thermal imaging module can better monitor the fan blades at night;

[0057] Adopt a monitoring and analysis method based on multi-layer deep learning of neural networks, and through self-learning with a large number of collected images, continuously improve the image recognition accuracy, thereby reducing the dependence on the observation experience of monitoring personnel.

[0058] As mentioned above, it is only the preferred embodiment of the present invention, and there is no any formal limitation to the present invention; any ordinary technician in this industry can smoothly implement the present invention according to what is shown in the attached drawings of the specification and what is described above; however, any equivalent changes such as slight modifications, decorations and evolutions made by those skilled in the art within the scope of the technical solution of the present invention by using the technical content disclosed above are all equivalent embodiments of the present invention; at the same time, any equivalent changes, modifications and evolutions made to the above embodiments based on the essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A monitoring device for the state of the fan blades of a wind turbine generator, characterized in that, the device includes a housing and a pan-tilt head. An inner sleeve is installed inside the housing. A visible light module and an infrared thermal imaging module are installed inside the inner sleeve, and the visible light module and the infrared thermal imaging module are installed to observe the same picture; the pan-tilt head fixedly installed on the wind turbine generator is connected to the bottom of the housing; the visible light module is used to collect the fan blade images during the operation of the wind turbine generator, and the infrared thermal imaging module is used to collect the infrared images of the fan blades during the operation of the wind turbine generator; a centralized processing module is further provided inside the inner sleeve, and the centralized processing module is used to analyze and process the fan blade images and infrared images of the fan blades, and judge whether there are dark stripes and icing on the fan blades; a temperature sensor and a semiconductor refrigerator are further provided inside the inner sleeve, and the semiconductor refrigerator is used to adjust the temperature of the inner sleeve according to the temperature data of the inner sleeve monitored by the temperature sensor and control it within -40°C to 50°C; semiconductor refrigerators are installed on both the upper and lower parts of the inner sleeve, and each semiconductor refrigerator has a heating surface and a cooling surface; the cooling surface of the semiconductor refrigerator located in the upper part faces the internal functional modules inside the inner sleeve, and the heating surface of the semiconductor refrigerator located in the lower part faces the internal functional modules inside the inner sleeve, forming a dual-function component for cooling and heating the temperature of the internal space of the inner sleeve.

2. The monitoring device for the state of the fan blades of a wind turbine generator according to claim 1, characterized in that, the housing is ellipsoidal.

3. The monitoring device for the state of the fan blades of a wind turbine generator according to claim 1, characterized in that, a wiper is installed on the housing, and the wiper is used to clean the windows of the visible light module and the infrared thermal imaging module.

4. The monitoring device for the state of the fan blades of a wind turbine generator according to claim 1, characterized in that, a gap is left between the housing and the inner sleeve, and a heat insulation layer is provided on the outer layer of the inner sleeve.

5. A method for monitoring the state of the fan blades of a wind turbine generator, characterized in that, the monitoring method is used to be realized by a monitoring device for the state of the fan blades of a wind turbine generator according to any one of claims 1-4, and includes: S1, collecting the visible light image and the infrared image of the fan blades of the wind turbine generator; S2, performing a unified conversion of the input requirements on the collected image data set to ensure the consistency of the subsequent processing results; specifically, performing preprocessing of the color space and color features on the visible light image; performing image preprocessing of inversion, histogram equalization, denoising, and sharpening on the infrared image; S3, using a convolutional neural network of deep learning to train the fan blade target object detector in the visible light module, and performing the recognition and detection of the fan blade target image; when the intensity of the ambient light is lower than the threshold and is not sufficient to enable the visible light module to detect the fan blade target, then the infrared thermal imaging module is used alone to perform the recognition and detection of the target image; S4, dividing the polar regions of each temperature region of the detected fan blade image, and establishing corresponding infrared image training sets and test sets with temperature information for dark cracks, snowfall, regional icing, and global icing; S5. Use a long short-term memory network (LSTM) to learn and classify images of the icing areas of the fan blades in infrared images with temperature information, and obtain the icing degree image classification of dark cracks, snowfall, small-area icing, large-area icing, full-area icing, and full-area low-temperature icing in the time series, completing the function of the image data analysis module; S6. Alarm the corresponding icing state or dark cracks according to the set conditions, and transmit the information to external devices to provide reference information for whether the wind turbine continues to operate; Between S1 and S2, it also includes: building libraries for visible light images and infrared images respectively, filtering the visible light images through a convolutional filter to filter and activate the fan blade features in the images, and learning to judge and identify the fan blade targets; filtering the infrared images through a convolutional filter to filter and activate the thermal radiation information features of the icing and dark crack areas in the images, and learning to judge whether there are dark cracks and icing.

Images