Monitoring system for health diagnosis of blade of wind turbine and its method

Abstract

A monitoring system for health diagnosis of a target blade of a wind turbine is provided. The system includes an infrared sensor, an image sensor, a laser sensor, a millimeter wave (mmWave) and at least a processing unit, wherein the at least processing unit is electrically connected with or communicates with the infrared sensor, the image sensor, and the specific sensor. Wherein, the at least a processing unit determines whether the target blade enters a sensing region according to sensing results of the infrared sensor and the image sensor, and when the at least a processing unit determines the target blade enters the sensing region, the specific sensor receives a specific reflect signal reflected from the target blade, and the at least a processing unit generates a monitoring result according to the specific reflect signal. Wherein the specific sensor includes at least a millimeter wave (mmWave) sensor and / or a laser sensor, and the specific reflect signal includes a mmWave reflect signal and / or a laser reflect signal.

Description

[Technical Field]

[0001] This invention relates to a monitoring system and method, and more particularly to a monitoring system and method for diagnosing the health of wind turbine blades. [Previous Technology]

[0002] With the advent of fifth-generation mobile communication technology (5G), millimeter-wave (mmWave) communication technology has been widely used. Compared to fourth-generation mobile communication technology (4G), 5G technology, which uses millimeter-wave communication technology, has advantages such as high transmission rate, low latency, and high transmission capacity, and can be applied to more industries. Currently, millimeter waves can be used for long-distance sensing, mostly in the automotive industry. For example, in the automatic following and driver assistance functions of cars, most of the time, the presence of objects in front of the car is sensed by emitting millimeter-wave signals and receiving reflected signals, or the distance between the car and the objects in front is calculated.

[0003] However, in other industries that require long-range sensing, such as the wind turbine industry, there is currently a lack of millimeter-wave applications and designs for the integration of millimeter-wave sensors with wind turbines, leaving much room for improvement. Furthermore, existing wind turbines lack reliable and energy-efficient blade health diagnostic methods.

[0004] Therefore, a novel monitoring system and method are needed to improve the above problems. [Summary of the Invention]

[0005] One object of the present invention is to provide a monitoring system for health diagnosis of wind turbine blades, used to monitor at least one target blade of a wind turbine. The monitoring system includes a specific sensor and at least one processing unit, wherein the at least one processing unit receives a specific reflection signal reflected by the at least one target blade from the specific sensor, and the at least one processing unit generates a monitoring result based on the specific reflection signal. The specific sensor includes at least one millimeter-wave sensor and / or at least one laser sensor, and the specific reflection signal includes a millimeter-wave reflection signal and / or a laser reflection signal.

[0006] One object of the present invention is to provide a monitoring system for health diagnosis of wind turbine blades, for monitoring at least one target blade of a wind turbine. The system includes an infrared sensor, an image sensor, a specific sensor, and at least one processing unit. The at least one processing unit is coupled to or communicatively connected to the infrared sensor, the image sensor, and the specific sensor. The at least one processing unit determines whether the at least one target blade has entered a sensing range based on the sensing results of the infrared sensor and the image sensor. When the at least one processing unit determines that the at least one target blade has entered the sensing range, the specific sensor receives a specific reflected signal reflected from the at least one target blade, and the at least one processing unit generates a monitoring result based on the specific reflected signal. The specific sensor includes at least one millimeter-wave sensor and / or at least one laser sensor, and the specific reflected signal includes a millimeter-wave reflected signal and / or at least one laser reflected signal.

[0007] One object of the present invention is to provide a monitoring method for health diagnosis of wind turbine blades, for monitoring at least one target blade of a wind turbine. The method is executed through a monitoring system for wind turbine blade health diagnosis, wherein the monitoring system includes an infrared sensor, an image sensor, a specific sensor, and at least one processing unit. The method includes the steps of: using at least one processing unit to determine whether the at least one target blade has entered a sensing range based on the sensing results of the infrared sensor and the image sensor; when the at least one processing unit determines that the at least one target blade has entered the sensing range, receiving a specific reflection signal reflected from the at least one target blade using the specific sensor; and using at least one processing unit to generate a monitoring result based on the specific reflection signal. The specific sensor includes at least one millimeter-wave sensor and / or at least one laser sensor, and the specific reflection signal includes a millimeter-wave reflection signal and / or at least one laser reflection signal.

Implementation Method

[0008] The exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. Wherever possible, the same element symbols are used in the drawings and description to denote the same or similar parts.

[0009] Certain terms are used throughout this specification and the appended claims to refer to specific components. Those skilled in the art will understand that sensing device manufacturers may use different names to refer to the same components. This document is not intended to distinguish between components that have the same function but different names. In the following specification and claims, words such as "containing," "comprising," and "including" are open-ended terms and should therefore be interpreted as "containing but not limited to...".

[0010] The terms “approximately,” “substantially,” or “roughly” are generally interpreted as being within 10% of a given value or range, or as being within 5%, 3%, 2%, 1%, or 0.5% of a given value or range.

[0011] The ordinal numbers used in the specification and claims, such as "first," "second," etc., to modify elements, do not in themselves imply or represent any prior ordinal number of that (or those) element, nor do they represent the order of one element with another, or the order of manufacturing methods. The use of these ordinal numbers is solely to clearly distinguish one named element from another element with the same name. The claims and specification may not use the same terminology; therefore, a first element in the specification may be a second element in the claims.

[0012] In this invention, the terms "given range is from the first value to the second value" and "given range falls within the range of the first value to the second value" indicate that the given range includes the first value, the second value, and other values ​​in between.

[0013] It should be understood that the features in the following embodiments can be replaced, recombined, or mixed to complete other embodiments without departing from the spirit of the present invention. Features between embodiments can be arbitrarily mixed and combined as long as they do not violate the spirit of the invention or conflict with it.

[0014] Unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It is understood that such terms, for example, as defined in commonly used dictionaries, should be interpreted as having a meaning consistent with the background or context of the relevant art and this invention, and should not be interpreted in an idealized or overly formal manner, unless specifically defined in the embodiments of this invention.

[0015] In addition, the word "adjacent" in the specification and claims is used to describe two things that are close to each other, and there may be contact or no contact between the two adjacent things.

[0016] Furthermore, descriptions such as "when..." or "...time" in this invention refer to states such as "at present, before, or after," and are not limited to simultaneous occurrences; this is stated in advance. Descriptions such as "set on..." in this invention refer to the corresponding positional relationship between two elements, and do not limit whether the two elements are in contact, unless specifically limited; this is stated in advance. Moreover, when this invention describes multiple functions, the use of the word "or" between functions indicates that the functions can exist independently, but does not exclude the possibility that multiple functions can exist simultaneously.

[0017] In this document, the term "electrical connection" or "coupled" includes various direct and indirect means of electrical connection, such as direct contact between two parties to transmit electrical signals, or the transmission of electrical signals between two parties through a third or more intermediaries. In this document, "mutual communication" may include data transmission between two parties through wired or wireless communication.

[0018] Figure 1 is a system architecture diagram of a monitoring system 1 for wind turbine blade health diagnosis (hereinafter referred to as monitoring system 1) according to an embodiment of the present invention. Monitoring system 1 is used to monitor at least one target blade 3 of a wind turbine 2, for example, monitoring the rotational speed, deflection, or offset of the at least one target blade 3, thereby assessing the health and fatigue state of the wind turbine 2, but not limited to this. Furthermore, there may be multiple target blades 3 being monitored, or only one blade may be monitored. For example, monitoring system 1 may simultaneously monitor multiple blades of the wind turbine 2, or it may only monitor some of the blades or one blade; for clarity, the following description uses the monitoring of one target blade 3 as an example.

[0019] As shown in Figure 1, the monitoring system 1 includes an infrared sensor 11, an image sensor 12, a specific sensor 13, and at least one processing unit 14. The at least one processing unit 14 can be coupled to the infrared sensor 11, the image sensor 12, and the millimeter-wave sensor 13 respectively, or the at least one processing unit 14 can communicate with the infrared sensor 11, the image sensor 12, and the specific sensor 13. In one embodiment, the monitoring system 1 may also include a storage device 15. The processing unit 14 may, for example, include a signal processing unit, and may have the functions of a signal processing unit. Furthermore, the wind turbine 2 may include a platform 21, a tower 22, a nacelle 23, a rotating shaft 24 (i.e., blade shaft), a fan assembly 25, and a generator 26. The platform 21 can be installed on land or at sea level, and the tower 22 is installed on the platform 21 and supports the nacelle 23. The generator 26 is installed inside the nacelle 23. The rotating shaft 24 can be connected between the fan assembly 25 and the generator 26. The above components are merely examples, and in practice, more or fewer components may be included. In one embodiment, a particular sensor 13 may include at least one millimeter-wave sensor 131 or at least one laser sensor 132, or it may simultaneously include at least one millimeter-wave sensor 131 and at least one laser sensor 132, and is not limited thereto.

[0020] In one embodiment, the infrared sensor 11, the image sensor 12, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) may be disposed on the cabin 23. In one embodiment, the infrared sensor 11, the image sensor 12, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) are disposed such that the sensing range of the infrared sensor 11, the imaging range of the image sensor 12, and the sensing range of the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) at least partially overlap. In one embodiment, the infrared sensor 11, the image sensor 12, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) may correspond to a specific sensing range R1. This specific sensing range R1 is defined as the range within which the infrared sensor 11, image sensor 12, and specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) can sense or image the target blade 3 when it enters the specific sensing range R1, and is not limited thereto. In one embodiment, the specific sensing range R1 may be the sensing range of the infrared sensor 11, but is not limited thereto. In another embodiment, the monitoring system 1 may also include another set or multiple sets of infrared sensors 11, image sensors 12, and specific sensors 13 (millimeter-wave sensors 131 and / or laser sensors 132). This other set of infrared sensors 11, image sensors 12, and specific sensors 13 (millimeter-wave sensors 131 and / or laser sensors 132) may be installed on the nacelle 23, on the platform 21, or at any suitable location on the wind turbine 2 for sensing the target blade 3, and is not limited thereto.

[0021] Next, the details of the aforementioned sensor will be explained.

[0022] Regarding the infrared light sensor 11. In one embodiment, the infrared light sensor 11 can be used to sense whether an object enters a specific sensing range R1. In one embodiment, the infrared light sensor 11 may be, for example, an infrared thermal imager, or the infrared light sensor 11 may have a sensing element, wherein the type of sensing element may include a long-wave infrared (LWIR) sensing element, a short-wave infrared (SWIR) sensing element, a middle-wave infrared (MWIR) sensing element, a near-infrared (NIR) sensing element, or a far-infrared (FIR) sensing element, and is not limited thereto. Depending on the type of sensing element, the sensing range of the infrared light sensor 11 will also be different. Furthermore, in one embodiment, the infrared light sensor 11 can be continuously activated and continuously sense whether an object enters the specific sensing range R1, but in other implementations, the infrared light sensor 11 may also be set to be activated periodically, or only activated at preset time points. In one embodiment, the infrared light sensor 11 may transmit its sensing results to the processing unit 14, but is not limited thereto.

[0023] Regarding the image sensor 12. In one embodiment, the image sensor 12 may include a charge-coupled device (CCD) or a complementary metal oxide semiconductor (CMOS) photosensitive element, but may also include other types of photosensitive elements, such as a contact image sensor (CIS), and is not limited thereto. In one embodiment, the image sensor 12 can be used to acquire a visible light image within its imaging range and transmit the acquired visible light image to the processing unit 14 so that the processing unit 14 can analyze the visible light image, and is not limited thereto. In one embodiment, when image acquisition is not required, the image sensor 12 may be preset to an off state or a sleep state (or a standby state), and when image acquisition is required, the image sensor 12 will be activated or deactivated from the sleep state (or deactivated from the standby state), and is not limited thereto.

[0024] Regarding the specific sensor 13. The specific sensor 13 used in this application may include a millimeter-wave sensor 131 and / or a laser sensor 132. First, the laser sensor 132 will be described. The laser sensor 132 can emit a laser beam, wherein the laser beam is collected in parallel by a lens inside the laser sensor 132 based on the time-of-flight (TOF) measurement technology, and then emitted to the target object. Then, the reflected light from the object can be received by the lens of the laser sensor 132 and processed by an image sensor (which may be additionally set in this system, or the laser sensor 132 itself may have an image sensor). In one embodiment, the distance moved by the target object will change the distance of the landing image sensed by the image sensor, thereby generating a current inside the laser sensor 132, and the amount of this current can be used to calculate the distance. Generally, laser sensors, also known as LADAR (Laser Detection and Ranging) or LiDAR (Light Detection and Ranging), are commonly used in radio detection and ranging applications. In one embodiment, the laser sensor 132 may employ a point-based distance measurement method, which involves direct distance measurement by aligning it with a specific target or reflector. Sensors performing one-dimensional (distance) operations in this manner are called one-dimensional sensors, or 1D laser sensors, and the technology using 1D laser sensors is called 1D laser sensing technology. In one embodiment, the laser sensor 132 may rotate or move the measurement beam in a plane to obtain distance and angle data, thereby producing two-dimensional results. Sensors used in this type of measurement application are generally called 2D laser sensors or 2D-LiDAR sensors, and the technology using 2D laser sensors is called 2D laser sensing technology. Measurements are taken sequentially, with measurement time intervals typically equal. Furthermore, by adding a rotating LiDAR sensor, it becomes a 3D laser sensor for three-dimensional (3D) operations, and the technology using a 3D laser sensor can be called 3D laser sensing technology. This provides distance and position information along the x-axis, as well as position information along the y and z-axis. In one embodiment, if the multiple light-projecting and light-receiving systems in the laser sensor 132 move and scan along different horizontal angles, similar information about various spatial parameters can also be obtained. In 2D and 3D applications, a focused projected laser beam can achieve high resolution, thereby scanning distant or finely structured objects. Typically, such sensors are called multilayer scanners. The laser sensor 132 of this application may include the various states described above, or any combination of the states described above, and is not limited thereto.

[0025] Regarding the millimeter-wave sensor 13. In one embodiment, the millimeter-wave sensor 13 can be used to transmit millimeter-wave signals, and when an object enters the sensing range of the millimeter-wave sensor 13, the millimeter-wave signal can be transmitted to the object, and the millimeter-wave sensor 13 can receive a millimeter-wave reflected signal reflected back from the object, and is not limited thereto. In one embodiment, the millimeter-wave sensor 13 can transmit the millimeter-wave reflected signal to the processing unit 14, so that the processing unit 14 can analyze the millimeter-wave reflected signal, and is not limited thereto. In one embodiment, the millimeter-wave sensor 13 can be continuously activated to continuously transmit millimeter-wave signals, but the millimeter-wave sensor 13 can also be set to activate periodically, or can be activated during a preset time period, and is not limited thereto. Furthermore, it should be noted that the millimeter-wave sensor 13 can only sense the target blade 3 when the wind blows and the target blade 3 rotates to a specific position (e.g., when it enters a specific sensing range R1). In other words, when the target blade 3 has not yet entered the specific sensing range R1, the millimeter-wave signal emitted by the millimeter-wave sensor 13 cannot be transmitted to the target blade 3.

[0026] Regarding the processing unit 14 and the storage device 15. In one embodiment, the processing unit 14 may include, for example, a physical processor that can execute software or firmware, but may also include a virtual processor in software form, and is not limited thereto. In one embodiment, the processing unit 14 may be paired with the storage device 15, wherein the storage device 15 may store at least one instruction, and the processing unit 14 may execute the instruction. When the processing unit 14 executes the instruction, the processing unit 14 may perform special operations to achieve special functions such as analyzing images or signals, signal filtering, etc., and is not limited thereto. In one embodiment, the storage device 15 may include, for example, a non-transitory computer-readable medium, wherein "non-transitory computer-readable medium" may include, for example, memory, hard disk, flash drive, virtual memory, cloud hard drive, or other similar devices, and is not limited thereto. Furthermore, in one embodiment, the instructions may be presented in software or firmware form, and are not limited thereto. In one embodiment, the processing unit 14 and the storage device 15 may be located in various suitable places, such as inside the wind turbine 2, or integrated into the infrared light sensor 11, the image sensor 12 and / or a specific sensor 13 (millimeter wave sensor 131 and / or laser sensor 132), or in a cloud server or a remote computer or mobile device, and are not limited thereto. In another embodiment, the infrared light sensor 11, the image sensor 12, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) can share the same processing unit 14. For example, the signals sensed by the infrared light sensor 11, the image sensor 12, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) can all be processed by the same processing unit 14. However, in other embodiments, the signals sensed by the infrared light sensor 11, the image sensor 12, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) can also be processed by different processing units 14, or some of the sensors may correspond to the same processing unit 14, while other sensors may correspond to another processing unit 14, and the embodiments are not limited to this.

[0027] Furthermore, it is assumed that the infrared light sensor 11, the image sensor 12, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) can be considered as a set of sensing elements. In one embodiment, the monitoring system 1 may have multiple sets of sensing elements. In this case, one set of sensing elements may be disposed on the cabin 23, and another set of sensing elements may be disposed on the platform 21, and is not limited thereto. It should be noted that if the sensing element set is to be disposed on the platform 21, the size of the platform 21 must be large enough so that the sensing element set can be disposed in an appropriate position. Here, "appropriate position" means that the position of the sensing element set on the platform 21 is set such that when the target blade 3 rotates to a specific orientation, it can enter the sensing range of the sensing element set, and is not limited thereto. In addition, in one embodiment, multiple sets of sensing elements may also be disposed on the cabin 23.

[0028] Next, the operation process of monitoring system 1 will be explained.

[0029] Figure 2 is a flowchart of the main operation of the monitoring method executed by the monitoring system 1 according to an embodiment of the present invention, and please refer to Figure 1 at the same time. As shown in Figure 2, firstly, step S1 is executed, and the infrared sensor 11 senses whether an object has entered the specific sensing range R1. When the infrared sensor 11 senses that an object has entered the specific sensing range R1, step S2 is executed, and the processing unit 14 starts or wakes up the image sensor 12 and proceeds to step S3. Alternatively, in another embodiment, if the infrared sensor 11 does not sense an object entering the infrared specific sensing range R11, then if the image sensor 12 (e.g., periodically turned on, or already turned on) senses the target image, then step S3 can also be executed. Step S3 is: the image sensor 12 obtains the visible light image of the object, and the processing unit 14 determines whether the object in the visible light image is the target blade 3. When processing unit 14 determines that the object in the visible light image is the target blade 3, step S4 is executed, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) receives specific reflection signals (e.g., millimeter-wave reflection signals and / or laser reflection signals) reflected from the target blade 3. Conversely, when the object is not the target blade 3, step S1 is re-executed. Then step S5 is executed, and processing unit 14 analyzes the specific reflection signals (millimeter-wave reflection signals and / or laser reflection signals) and generates monitoring results based on the specific reflection signals (millimeter-wave reflection signals and / or laser reflection signals). In some other embodiments, steps S1 and S2 can also be modified so that the infrared sensor 11 and the image sensor 12 are continuously or periodically activated. In this case, if the image sensor 11 detects a target image but does not detect an object entering the specific sensing range R11 of the infrared sensor, step S3 can still be performed, and vice versa, and is not limited to this.

[0030] Regarding steps S1 and S2, since the blades of the wind turbine 2 rotate due to wind, the position of the target blade 3 changes over time. Therefore, the target blade 3 may not necessarily be located within the specific sensing range R1. When the target blade 3 is located outside the specific sensing range R1, the object within the specific sensing range R1 may be a non-target blade, or no object may have entered the specific sensing range R1. In this case, the millimeter-wave reflected signal received by the millimeter-wave sensor 13 may not originate from the target blade 3, or the millimeter-wave sensor 13 may not receive the millimeter-wave reflected signal. Therefore, the processing unit 14 does not need to analyze the signal. In other words, the processing unit 14 only needs to analyze the millimeter-wave reflected signal when the target blade 3 is located within the specific sensing range R1. By using steps S1 and S2, the infrared sensor 11 and the image sensor 12 first determine whether the object within the specific sensing range R1 is the target blade 3, ensuring that the millimeter-wave reflected signal is reflected from the target blade 3. This avoids the processing unit 14 processing signals that do not need to be processed. In addition, in one embodiment, the processing unit 14 may also use artificial intelligence deep learning technology to remove signals from non-target blades.

[0031] In addition, the image sensor 12 usually consumes a lot of power and is not suitable for long-term operation. By using the infrared sensor 11 in step S1 to determine whether an object has entered the specific sensing range R1 before deciding whether to start the image sensor 12, it can be ensured that the image sensor 12 will only be started or operated when it is necessary to capture an image, and unnecessary power consumption can be avoided.

[0032] Regarding step S3, in one embodiment, the storage device 15 may, for example, pre-store an image of the target blade 3. Therefore, the processing unit 14 can use image recognition technology (including artificial intelligence technology) to compare and determine whether the object in the visible light image obtained by the image sensor 12 is the target blade 3. However, it is not limited to this, and the present invention can also employ various other suitable image recognition methods. In one embodiment, the image sensing used in the present invention can also be used to identify and assess the damage state of the wind turbine blades or the adhesion of rainwater and snowflakes, and is not limited to this.

[0033] Regarding steps S4 and S5, in one embodiment, the monitoring result may include a deformation or flexural displacement monitoring result of the target blade 3, or a rotational speed monitoring result, and is not limited thereto. In other words, the processing unit 14 may analyze whether the target blade 3 has deformed or flexed based on millimeter-wave reflection signals and / or laser reflection signals, or may calculate the deformation or displacement of the target blade 3, or the processing unit 14 may also analyze the rotational speed of the target blade 3 based on millimeter-wave reflection signals and / or laser reflection signals, and is not limited thereto. In addition, in one embodiment, the image sensor 12 may also acquire image signals at different time points, and the processing unit 14 may identify the target blade 3 in the image signals and compare the target blade 3 at different time points to determine whether the target blade 3 has deformed or shifted, but is not limited thereto.

[0034] In some cases, the output data of the millimeter-wave sensor 131 (e.g., information on millimeter-wave reflected signals) and / or the output data of the laser sensor 132 (e.g., information on laser reflected signals) may contain noise or have irregular characteristics, which may affect the signal analysis of the processing unit 14. In one embodiment, the processing unit 14 may be coupled to or communicatively connected to a signal filter (not shown), or the processing unit 14 itself may include the function of a signal filter. The signal filter can be used to filter the millimeter-wave reflected signals to reduce the complexity of signal analysis. In one embodiment, the type of signal filter may include a Savitzky-Golay filter, an elliptic filter, or a Kalman filter, and is not limited thereto. In one embodiment, this application may use one of the above filters alone, but in other embodiments, at least two of the above three filters may be used simultaneously.

[0035] Regarding the Savitzky-Golay filter. In one embodiment, the Savitzky-Golay filter can smooth data and increase the precision of detecting phase changes in the data. It works by using a polynomial to fit a window to adjacent data. Generally, increasing the window size can increase the smoothness of the output data. However, an excessively large window size can cause important features in the data to be lost. Furthermore, a higher polynomial order allows for the capture of more complex data; however, overfitting can make it more sensitive to noise, potentially leading to insufficient filtering of the output data. If both are increased—that is, the polynomial order and the window size are both increased—the computational complexity also increases significantly. In one embodiment, the Savitzky-Golay filter used in this invention can have a window size of 13 and a polynomial order of 3, providing better accuracy, but is not limited thereto.

[0036] Regarding Elliptic filters. In one embodiment, a bandpass Elliptic filter can effectively extract phase shifts from the signal, enabling faster transitions between passbands. The steep roll-off of the Elliptic filter provides the bandpass filter with different cutoff frequencies; if the cutoff frequency is too wide, there is a risk of obtaining unwanted frequencies, while conversely, if the cutoff frequency is too narrow, valuable information will be suppressed. In one embodiment, the Elliptic filter used in this invention is set such that its cutoff frequency updates with the window, and its adaptive adjustment ensures that the filter remains constant, exhibiting adaptability and responsiveness to different conditions.

[0037] Regarding the Kalman filter. In one embodiment, the Kalman filter can be used to observe whether the values ​​calculated based on the millimeter-wave reflected signal contain unpredictable errors. In this case, the Kalman filter can be applied to values ​​such as range, velocity, and azimuth calculated by the processing unit 14 based on the millimeter-wave reflected signal to ensure that these values ​​are noise-reduced or unaffected by noise or irregularities. The Kalman filter is designed for a constant acceleration model, thus adapting to dynamic changes and changes in time. This adaptive characteristic of the Kalman filter is particularly important and valuable when dealing with unexpected data in the real world. Since the measured values ​​may change or be erroneous, the Kalman filter contributes to the robustness and accuracy of the estimation process by iteratively updating the estimates and model predictions based on the data.

[0038] Thus, the main operation of the monitoring system 1 can be understood. Next, details of the deformation or offset monitoring results and rotational speed monitoring results of the target blade 3 generated by the processing unit 14 will be explained separately.

[0039] First, we will explain how the processing unit 14 generates the deformation or deflection monitoring results of the target blade 3. Figure 3A is a flowchart of the detailed steps of the monitoring method of the first embodiment of the present invention. Please also refer to Figures 1 and 2. Steps S1 to S4 are applicable to the explanation in Figure 2, so they will not be described in detail here. In addition, steps S501 to S503 are an example of detailed sub-steps of step S5. Figure 3A illustrates the case where the specific sensor 13 is a millimeter-wave sensor 131. The case where the specific sensor 13 is a laser sensor 132 will be described in a later paragraph with Figure 3B.

[0040] As shown in Figure 3A, after steps S1 to S4 are completed, step S501 can be executed, and the millimeter-wave sensor 131 receives millimeter-wave reflected signals at different time points. Then, step S502 is executed, and the processing unit 14 tracks at least one specific part on the target blade 3 based on the millimeter-wave reflected signals at different times. Then, step S503 is executed, and the processing unit 14 generates a deformation or deflection monitoring result of the target blade 3 based on the distance change, velocity change, and azimuth angle change of the at least one specific part on the target blade 3 at different times. The deformation or deflection monitoring result may be whether the target blade 3 has undergone deformation or deflection, or the deformation or deflection amount of the target blade 3, and is not limited to these.

[0041] Regarding step S501, in one embodiment, the millimeter-wave sensor 131 continuously emits millimeter-wave signals and continuously receives reflected signals. Therefore, when the target blade 3 enters the sensing area R1, the millimeter-wave sensor 131 can receive the millimeter-wave reflected signals reflected back from the target blade 3 at different time points (which can be regarded as the reflected signals when the target blade 3 rotates to different positions). In one embodiment, in order for the processing unit 14 to determine whether the target blade 3 has deformed or deflected, the millimeter-wave sensor 131 must acquire the millimeter-wave reflected signals of the target blade 3 at at least two time points, so that the processing unit 14 can generate a judgment result based on the millimeter-wave reflected signals at these two time points.

[0042] Regarding step S502, please refer to Figures 3A and 4 simultaneously. Figure 4 is a schematic diagram of the distribution of millimeter-wave signals at different points in time according to an embodiment of the present invention. In one embodiment, when the millimeter-wave signal sent by the millimeter-wave sensor 13 is transmitted to the target blade 3, due to the shape design of the target blade 3 and the angle of signal transmission, the signal intensity of the millimeter-wave signal may be concentrated in a plurality of regions on the target blade 3 (e.g., 3A~3E in Figure 4, but not limited to these). The signal intensity of the millimeter-wave reflected signal received by the millimeter-wave sensor 13 will also be grouped and reflected roughly according to the plurality of regions 3A~3E. Furthermore, the plurality of regions 3A~3E have a positional order, for example, region 3A is at the top and region 3E is at the bottom. Therefore, the plurality of regions 3A~3E can be used as the basis for the processing unit 14 to identify specific parts on the target blade 3. In this way, the processing unit 14 can know the approximate location of each part of the target blade 3 at different time points, and make individual comparisons for each part accordingly, but is not limited thereto. In one embodiment, the tracking of the specific part in step S502 can be achieved at least through a trajectory tracking algorithm, such as a particle image velocimetry algorithm, but is not limited thereto.

[0043] Regarding step S503, in one embodiment, the processing unit 14 can obtain, at each time point, distance information (e.g., the relative distance between the millimeter-wave sensor 13 and the at least one specific part at that time point), speed information (e.g., the relative speed between the millimeter-wave sensor 13 and the at least one specific part at that time point), and horizontal angle information (e.g., the horizontal angle of the at least one specific position relative to the millimeter-wave sensor 13 at that time point) of the at least one specific position based on the millimeter-wave signal and the millimeter-wave reflection signal, and compare the distance information, speed information, and horizontal angle information of the at least one specific position at different time points to obtain the distance change, speed change, and angle change in space of the at least one specific position. Next, in one embodiment, the processing unit 14 can use various suitable algorithms to convert the distance changes, velocity changes, and spatial angular changes of the at least one or more specific parts of the target blade 3 into the deformation or deflection of the target blade 3, and determine whether the target blade 3 has deformed or deflected based on the deformation or deflection, or output the deformation or deflection, and is not limited thereto. In one embodiment, whether deformation or deflection has occurred can be determined by comparing the distance changes, velocity changes, and horizontal angular changes of different parts of the target blade 3 to see if outliers appear. For example, when the azimuth angular change of part 3A of the target blade 3 is much greater than the azimuth angular change of other parts, the probability of deformation of part 3A is relatively high. Therefore, the processing unit 14 can determine that the target blade 3 may deform, but the determination method of the present invention is not limited to this.

[0044] Further, in one embodiment, the processing unit 14 can also determine the degree of blade wear (which can also be regarded as a blade health diagnosis) based on the calculated deformation or flexural offset. In one embodiment, the degree of wear can be evaluated, for example, by pre-classifying the deformation or flexural offset and using a lookup table or threshold comparison to evaluate the calculated deformation or flexural offset, but it is not limited thereto, and other judgment methods can also be applied to the present invention.

[0045] Thus, the process by which the processing unit 14 generates the deformation or flexural displacement monitoring results of the target blade 3 can be understood.

[0046] Next, the state when the specific sensor 13 is a laser sensor 132 will be described. FIG3B is a detailed flowchart of the monitoring method of the second embodiment of this application. Please refer to FIG1 and FIG2 simultaneously. Steps S1 to S4 are applicable to the description in FIG2, so they will not be described in detail. In addition, steps S511 to S513 are an example of detailed sub-steps of step S5. FIG3B is illustrated with the example of the specific sensor 13 being a laser sensor 132.

[0047] Laser sensors can be used for non-contact measurement. Their greatest advantage is their applicability to objects with almost any characteristics, offering numerous application options in industry, such as logistics (transportation processes, etc.), road traffic flow detection, and automation of container loading and unloading processes in ports. The reflected light output of a laser pulse directly depends on the physical characteristics of the object being measured and the distance. Because the laser pulse diffuses in a right-angled plane along the projection direction (i.e., diverges), the light output per unit area reaching the object will be correspondingly reduced, depending on the distance. The diffusion condition also applies to reflected light. Furthermore, when aligning with the surface to be irradiated, it is not necessary to ensure that all reflected light returns along the sensor direction. Typically, only a small portion of the reflected light reaches the laser sensor's receiver. Laser sensors can be used as point-based distance measurement systems. Sensors that perform one-dimensional (distance) operations by aligning with a specific target or reflector are called one-dimensional sensors, or 1D sensors. Rotating or moving the measurement beam on a plane provides distance and angle data, thus providing two-dimensional results. Sensors used in this type of measurement application are generally called 2D laser scanners. Adding a rotating laser sensor enables three-dimensional operation. This provides distance and position information along the x-axis, as well as position information along the y and z-axis. If multiple light-emitting and receiving systems within the sensor move and scan along different horizontal angles, similar information about various spatial parameters can be obtained. Such sensors are called multilayer scanners or 3D sensors.

[0048] As shown in Figure 3B, after steps S1 to S4 are completed, step S511 can be executed. The laser sensor 132 receives laser reflection signals at different time points using laser sensing technology, which can be 1D laser sensing technology, 2D laser sensing technology, or 3D laser sensing technology. Then, step S512 is executed, and the processing unit 14 tracks at least one specific part on the target blade 3 based on the laser reflection signals at different times. Then, step S503 is executed, and the processing unit 14 generates deformation or deflection monitoring results of the target blade 3 based on the distance changes, speed changes, and azimuth angle changes of the at least one specific part on the target blade 3 at different times. The deformation or deflection monitoring results can be whether the target blade 3 has deformed or deflected, or the deformation amount or deflection amount of the target blade 3, and are not limited to these.

[0049] Regarding step S511, in one embodiment, the laser sensor 132 continuously uses 1D laser sensing technology, 2D laser sensing technology, or 3D laser sensing technology for sensing. Therefore, when the target blade 3 enters the sensing area R1, the laser sensor 132 can receive the laser reflection signal reflected back from the target blade 3 at different time points (which can be regarded as the reflection signal when the target blade 3 rotates to different positions). In one embodiment, in order for the processing unit 14 to determine whether the target blade 3 has deformed or deflected, the laser sensor 132 must acquire the laser reflection signal of the target blade 3 at at least two time points so that the processing unit 14 can generate a judgment result based on the laser reflection signal at these two time points. In one embodiment, the implementation of 1D laser sensing technology, 2D laser sensing technology, or 3D laser sensing technology can be referred to the description in the preceding paragraph (e.g., paragraph

[0025] ), but is not limited thereto.

[0050] Regarding steps S512 and S513, refer to the description of steps S502 and S503, and replace the millimeter wave signal and millimeter wave reflection signal in steps S502 and S503 with laser signal and laser reflection signal, but not limited thereto.

[0051] Next, we will explain how the processing unit 14 generates the rotational speed monitoring result of the target blade 3. Figure 5 is a flowchart of the detailed steps of the monitoring method of the third embodiment of the present invention. Please refer to Figures 1 and 2 simultaneously. Steps S1 to S4 are applicable to the description in Figure 2, so they will not be described in detail here. In addition, steps S511a to S513 are an example of the detailed sub-steps of step S5.

[0052] As shown in Figure 5, after step S4 is executed, step S511a is executed, and the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) detects that the target blade 3 has rotated to a first rotation position and a second rotation position. Alternatively, step S511b is executed, and the infrared sensor 11 detects that the target blade 3 has rotated to a first rotation position and a second rotation position. Alternatively, step S511c is executed, and the image sensor 12 detects that the target blade 3 has rotated to a first rotation position and a second rotation position. After steps S511a, S511b and / or S511c are executed, step S512 is executed, and the storage device 15 records the first rotation position and the second rotation position, and simultaneously records the time when the target blade 3 rotates to the first rotation position and the second rotation position. Next, step S513 is executed, and processing unit 14 calculates the rotational speed of target blade 3 based on the distance between the first rotational position and the second rotational position, and based on the time difference between the target blade 3 rotating to the first rotational position and the second rotational position. In one embodiment, at least one of steps S511a, S511b, and S511c may be executed, or at least two of steps S511a, S511b, and S511c may be executed, or all three of steps S511a, S511b, and S511c may be executed.

[0053] Regarding steps S511a and S511b, in one embodiment, a specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) and / or infrared sensor 11, in conjunction with the processing unit 14, can track the dynamics of the target blade 3 by detecting its approximate outline, thereby obtaining the first rotation position and the second rotation position of the target blade 3, and is not limited thereto. Regarding step S511c, in one embodiment, an image sensor 12, in conjunction with the processing unit 14, can track the dynamics of the target blade 3 by using special markings on the target blade 3 or specific parts of the target blade 3 (e.g., the tip of the blade, and is not limited thereto), thereby obtaining the first rotation position and the second rotation position of the target blade 3. In one embodiment, the first rotation position and the second rotation position can be, for example, any two positions of the target blade 3 within a specific sensing range R1, and is not limited thereto.

[0054] Regarding steps S512 and S513, in one embodiment, the rotation speed may be measured in revolution(s) per minute (RPM), but is not limited thereto.

[0055] Furthermore, in one embodiment, the processing unit 14 can calculate the rotational speed of the target blade 3 using the first rotational position and the second rotational position obtained by at least two of the specific sensors 13 (millimeter-wave sensor 131 and / or laser sensor 132), infrared sensor 11, and image sensor 12, and compare the calculated rotational speeds for multiple verifications to improve accuracy, and is not limited thereto. Further, the blade rotational speed calculated in step S513 can also be used to calculate the current wind speed, thereby deriving the power generation efficiency of the wind turbine 2, and is not limited thereto.

[0056] Thus, the process by which the processing unit 14 generates the rotational speed monitoring results of the target blade 3 can be understood.

[0057] The present invention may also have different embodiments. Figure 6 is a system architecture diagram of the monitoring system 1 according to another embodiment of the present invention. The architecture of the monitoring system 1 in Figure 6 is generally similar to that in Figure 1, so the following mainly describes the differences. As shown in Figure 6, the monitoring system 1 may also include a Lidar 16 for measuring the wind field around the wind turbine 2. In addition, the wind turbine 2 may also include a mechanism adjustment unit 27 and a control unit 28, wherein the control unit 28 may be coupled to the processing unit 14 and the mechanism adjustment unit 27 respectively, and the mechanism adjustment unit 27 may be connected to or coupled to the rotating shaft 24.

[0058] In one embodiment, the light source 16 measures the wind field around the wind turbine 2 using the Doppler effect, but is not limited thereto. In one embodiment, the light source 16 can be located at any suitable position on the wind turbine 2, such as the rotating shaft 24, the nacelle 23, or other suitable positions, and is not limited thereto.

[0059] In one embodiment, the mechanism adjustment unit 27 can be connected to the rotating shaft 24 for adjusting the orientation of the rotating shaft 24 in space. In one embodiment, the mechanism adjustment unit 27 can employ various feasible mechanical operating mechanisms, such as gears, shafts, etc., and is not limited thereto. Since the details of how to actually use mechanical devices to adjust the orientation of the rotating shaft 24 are not the focus of this invention and fall within the scope of prior art, this invention will not describe them in detail. In one embodiment, the control unit 28 can be used to control the mechanism adjustment unit 27 to operate in order to adjust the orientation of the rotating shaft 24, and is not limited thereto. In one embodiment, the control unit 28 can be, for example, a microcontroller, but is not limited thereto.

[0060] In some cases, the fan assembly 25 of the wind turbine 2 may shift due to excessive surrounding wind. By using the LiDAR 16 in conjunction with the millimeter-wave sensor 131 and / or the laser sensor 132, the processing unit 14 can determine the wind strength and direction around the wind turbine 2 and correct the shift of the fan assembly 25 accordingly. Figure 7 is a flowchart of the operation of the LiDAR 16 in conjunction with the millimeter-wave sensor 131 and / or the laser sensor 132 according to an embodiment of the present invention. Please also refer to Figures 1 to 6.

[0061] As shown in Figure 7, step S71 is first executed, whereby the LiDAR 16 measures the wind field around the wind turbine 2. Additionally, step S72 is executed, where a specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132) senses the deformation or flexural displacement monitoring results of one or more blades. After steps S71 and S72 are completed, step S73 is executed, whereby the processing unit 14 calculates real-time wind force and direction information based on the wind field information measured by the LiDAR 16 and the deformation or flexural displacement monitoring results of the one or more blades sensed by the specific sensor 13 (millimeter-wave sensor 131 and / or laser sensor 132). Next, step S74 is executed, whereby the processing unit 14 determines a yaw angle of the fan assembly 25 in space based on the real-time wind force and direction information, and generates an adjustment command based on that yaw angle. Next, step S75 is executed, and the mechanism adjustment unit 17 receives the adjustment command and adjusts the orientation of the rotating shaft 24 in space according to the adjustment command. In this way, the orientation of the fan assembly 25 can be corrected.

[0062] In one embodiment, when a suspected infringing product is discovered, the present invention can at least determine whether it falls within the protection scope of the present invention by examining the presence or absence of components, component configuration, mechanism observation and / or operation mode of the suspected infringing product, and is not limited thereto.

[0063] Details or features of the various embodiments of the present invention may be arbitrarily mixed and matched as long as they do not violate the spirit of the invention or conflict with it.

[0064] The system and method of the present invention can solve the problems of the prior art.

[0065] The above embodiments are merely examples for the convenience of illustration. The scope of the rights claimed by this invention should be determined by the claims of the patent application, and not limited to the above embodiments. [Simplified Explanation of the Diagram]

[0066] Figure 1 is a system architecture diagram of a monitoring system for wind turbine blade health diagnosis according to an embodiment of the present invention. Figure 2 is a flowchart of the main operation of the monitoring method performed by the monitoring system according to an embodiment of the present invention. Figure 3A is a flowchart of the detailed steps of the monitoring method according to the first embodiment of the present invention. Figure 3B is a flowchart of the detailed steps of the monitoring method according to the second embodiment of the present invention. Figure 4 is a schematic diagram of the distribution of millimeter-wave signals at different time points according to an embodiment of the present invention. Figure 5 is a flowchart of the detailed steps of the monitoring method according to the third embodiment of the present invention. Figure 6 is a system architecture diagram of a monitoring system according to another embodiment of the present invention. Figure 7 is a flowchart of the operation steps of a light source, a millimeter-wave sensor, and / or a laser sensor according to another embodiment of the present invention.

Claims

1. A monitoring system for health diagnosis of wind turbine blades, used to monitor at least one target blade (3) of a wind turbine (2), comprising: an infrared sensor (11); an image sensor (12); a specific sensor (13); and at least one processing unit (14), coupled to or communicating with the infrared sensor (11), the image sensor (12), and the specific sensor (13); wherein, The at least one processing unit (14) determines whether the at least one target blade (3) has entered a specific sensing range (R1) based on the sensing results of the infrared sensor (11) and the image sensor (12). When the at least one processing unit (14) determines that the at least one target blade (3) has entered the specific sensing range (R1), the specific sensor (13) receives a specific reflection signal reflected from the at least one target blade (3), and the at least one processing unit (14) generates a monitoring result based on the specific reflection signal.

2. The monitoring system as claimed in claim 1, wherein the particular sensor (13) comprises at least one millimeter-wave sensor (131) and / or at least one laser sensor (132), and the particular reflection signal comprises a millimeter-wave reflection signal and / or a laser reflection signal.

3. The monitoring system as claimed in claim 2, wherein the monitoring results include at least one of a deformation or flexural displacement monitoring result of the at least one target blade (3) and a rotational speed monitoring result of the target blade (3).

4. The monitoring system as claimed in claim 3, wherein the at least one processing unit (14) determines whether the at least one target blade (3) has entered a specific sensing range (R1) based on the sensing results provided by the infrared sensor (11) and the image sensor (12) as follows: when the infrared sensor (11) senses that an object has entered an infrared light sensing range (R11), the at least one processing unit (14) activates or wakes up the image sensor (12), and determines whether the object is the at least one target blade (3) based on a visible light image provided by the image sensor (12).

5. The monitoring system as claimed in claim 4, wherein the specific sensor (13) receives the specific reflection signal at different time points, the at least one processing unit (14) tracks at least one specific area on the at least one target blade (3) by means of the specific reflection signal at the different time points, and generates the deformation or deflection monitoring result of the at least one target blade (3) based on at least one of the distance change, speed change and angle change of the at least one specific area on the at least one target blade (3).

6. The monitoring system as claimed in claim 2, wherein the image sensor (12) provides an image signal at different time points, and the at least one processing unit (14) compares the target blade (3) in the image signal at the different time points to determine whether the at least one target blade (3) has deformed or shifted.

7. The monitoring system as claimed in claim 5, wherein the specific sensor (13) receives the specific reflection signal at different time points, the at least one processing unit (14) tracks the position of the at least one target blade (3) by means of the specific reflection signal at the different time points, and generates the rotational speed monitoring result of the at least one target blade (3) based on a time interval between the at least one target blade (3) rotating from a first rotation position (P1) to a second rotation position (P2) and the distance between the first rotation position (P1) and the second rotation position (P2).

8. The monitoring system as claimed in claim 7, wherein the infrared sensor (11) provides an infrared sensing signal at different time points and / or the image sensor (12) provides an image signal at different time points, and the at least one processing unit (14) tracks the position of the at least one target blade (3) based on the infrared sensing signal and / or the image signal at the different time points, thereby determining the rotational speed of the at least one target blade (3).

9. The monitoring system as claimed in claim 2, further comprising a LiDAR (16) for detecting the wind field around the wind turbine (2).

10. The monitoring system as claimed in claim 3, further comprising a light source (16), a computing unit (4) and a mechanism adjustment unit (5), wherein the light source (16) is used to detect the wind field around the wind turbine (2), the computing unit (4) is used to calculate an instantaneous wind force and wind direction information based on the wind field information detected by the light source (16) and the deformation or flexural displacement monitoring result of the at least one target blade (3), and to determine a maximum displacement direction of the at least one target blade (3) based on the instantaneous wind force and wind direction information, and the mechanism adjustment unit (5) adjusts the orientation of a blade shaft of the wind turbine (2) based on the maximum displacement direction of the at least one target blade.

11. A monitoring method for health diagnosis of wind turbine blades, used to monitor at least one target blade (3) of a wind turbine (2), the method being executed through a monitoring system (1) for wind turbine blade health diagnosis, wherein the monitoring system (1) includes an infrared sensor (11), an image sensor (12), a specific sensor (13), and at least one processing unit (14), wherein the method includes the steps of: determining, by the at least one processing unit (14), whether the at least one target blade (3) has entered a specific sensing range (R1) based on the sensing results of the infrared sensor (11) and the image sensor (12); when the at least one processing unit (14) determines that the at least one target blade (3) has entered the specific sensing range (R1), receiving, by the specific sensor (13), a specific reflection signal reflected from the at least one target blade (3); and generating a monitoring result by the at least one processing unit (14) based on the specific reflection signal; wherein, The particular sensor (13) includes at least one millimeter-wave sensor (131) and / or at least one laser sensor (132), and the particular reflection signal includes a millimeter-wave reflection signal and / or a laser reflection signal.

Images